JP7240564B1 - Evaluation device and evaluation system - Google Patents

Abstract

The evaluation device is an evaluation device that evaluates the operation status of a plant including a plurality of facilities, and derives an evaluation value based on a display unit and two reference values, and outputs the two reference values and the evaluation value. and an evaluation unit that displays the relationship between and in a three-dimensional graph on the display unit, and the evaluation unit displays the relationship including information indicating the time-series change in the evaluation value accompanying the time-series change in the two reference values , the two reference values are values related to the operating status of two facilities or two facilities included in one facility, and the three-dimensional graph displays one of the two reference values , the other of the two reference values is the other one of the three axes, and the evaluation value is the remaining one of the three axes.

Description

The present invention relates to an evaluation device and an evaluation system.

2. Description of the Related Art A method is known in which the operating status of a device can be grasped based on the output of a sensor provided in the device (for example, Patent Literature 1).

JP 2019-178625 A

In plants, there is also a demand to grasp the operation status of the plant by installing sensors in the devices of various facilities included in the plant. On the other hand, in a large-scale plant such as a power plant, an integrated operation is established when the operational states of various facilities included in the plant interact with each other. In such plants, if something goes wrong, it tends to be caused by a portion of the entire facility or equipment. Also, an abnormality in one part may affect other parts of the whole. However, with the method of Patent Document 1, it is difficult to grasp the relationship between the state of the part concerned, the state of the facility or equipment as a whole, and the state of other parts of the whole.

The present invention provides an evaluation device and an evaluation system that make it easier to grasp the relationship between the state of a part of a facility or equipment, the state of the entire facility or equipment including that part, and the state of other parts of the whole. for the purpose.

In order to solve the above-described problems and achieve the object, the evaluation apparatus of the present invention is an evaluation apparatus that evaluates the operation status of a plant including a plurality of facilities, and includes a display unit and two reference values. an evaluation unit that derives an evaluation value based on causing the display unit to display the relationship including information indicating the time-series change in the evaluation value accompanying the time-series change in the value, and the two reference values are two of the facilities or one of the facilities. is a value related to the operating status of one facility, and the three-dimensional graph has one of the two reference values as one of the three axes, and the other of the two reference values as the other one of the three axes, Let the evaluation value be the remaining one of the three axes.

As a desirable aspect of the present invention, the information indicating time-series changes in the evaluation value includes information indicating at least two points in time.

As a desirable aspect of the present invention, the information indicating the time-series change in the evaluation value includes information indicating the direction of change in the evaluation value between at least two points in time.

As a desirable aspect of the present invention, the evaluation unit causes the display unit to further display information indicating a correspondence relationship between the evaluation value and the threshold value of the evaluation value.

As a desirable aspect of the present invention, an acquisition unit is provided for acquiring sensing values individually output by a plurality of sensors provided in a plurality of facilities included in each facility, and the evaluation unit includes a plurality of sensing values output by the plurality of sensors. A first process for calculating a first evaluation value indicating the stability of the facility based on the sensing value of the facility is performed for each facility, and the Lp norm of the first evaluation value of the plurality of facilities is used in the facility including the plurality of facilities. and performing a second process of calculating a second evaluation value of the sensing item for each group of the first evaluation values classified by type of sensing value, and calculating the second evaluation value corresponding to the sensing item in one facility. A third process for calculating a third evaluation value indicating the stability of the one facility as the Lp norm is performed for each facility, and the Lp norm of the third evaluation values of a plurality of facilities indicates the stability of the plant. A fourth process for calculating a fourth evaluation value is performed, the first value being the weighting value of the Lp norm in the second process, the second value being the weighting value of the Lp norm in the third process, and the fourth Unlike the third value that is the Lp norm weighting value in the process, the sensing value in the first process functions as the reference value, the first evaluation value in the first process functions as the evaluation value, The first evaluation value in the second process functions as the reference value, the second evaluation value in the second process functions as the evaluation value, and the second evaluation value in the third process functions as the reference value. , the third evaluation value in the third process functions as the evaluation value, the third evaluation value in the fourth process functions as the reference value, and the fourth evaluation value in the fourth process functions as the evaluation value.

As a desirable aspect of the present invention, the first value is greater than the second value and the third value, and the second value is greater than the third value.

As a desirable aspect of the present invention, the evaluation unit performs the second processing for each of the plurality of sensing items for some or all of the plurality of facilities, and performs the second processing for each of the plurality of sensing items. In the third process of , the third evaluation value is calculated as the Lp norm of the second evaluation values of the plurality of sensing items.

As a desirable aspect of the present invention, the evaluation unit performs the second processing for one sensing item on some or all of the plurality of facilities, In the third process, the third evaluation value is calculated as the Lp norm of the second evaluation value of the one sensing item.

As a desirable aspect of the present invention, the sensor detects at least one of temperature, vibration, water pressure and water level of installed equipment.

As a desirable aspect of the present invention, the evaluation device according to any one of claims 1 to 9, and a terminal provided with a notification unit that performs notification related to evaluation of the operation status of the plant.

According to the present invention, it becomes easier to grasp the relationship between the status of a part of a facility or equipment, the overall status of the facility or equipment including that part, and the status of other parts in the whole.

FIG. 1 is a diagram showing a main configuration example of an evaluation system according to an embodiment. FIG. 2 is a block diagram showing a configuration example of the evaluation device. FIG. 3 is a graph showing an example of the relationship between sensing values and evaluation values. FIG. 4 is a graph showing an example of the relationship between time-series changes in sensing values and time-series changes in evaluation values. FIG. 5 is a diagram showing an example of a correspondence relationship between a plurality of facilities included in the plant 2, items sensed in the facilities, facilities to be sensed, and the like. FIG. 6 is a graph schematically showing an example of calculation of a total plant evaluation value based on two evaluation values. FIG. 7 is a graph schematically showing an example in which p=1 in calculating the comprehensive evaluation value r of the plant 2 based on the two evaluation values. FIG. 8 is a graph schematically showing an example of setting p=10 in calculating the comprehensive evaluation value r of the plant 2 based on the two evaluation values. FIG. 9 shows the total evaluation value when p= ₁ , the total evaluation value when p=2, and the total evaluation value when p=10 when the evaluation values x ₁ =10 and x 2 =10. and a comprehensive evaluation value when p=infinity (∞). FIG. 10 is a graph for explaining examples when different comprehensive evaluation values are calculated. FIG. 11 is a diagram showing an example of notification regarding evaluation of the operating status of the plant. FIG. 12 is a graph showing an example of changing the upper limit of variance according to season. FIG. 13 is a graph showing time-series changes in the evaluation value of one piece of equipment. FIG. 14 is a graph showing time series changes in the evaluation value of one piece of equipment different from FIG. FIG. 15 is a graph showing time-series changes in the overall evaluation value of the facility having the two facilities described with reference to FIGS. 13 and 14. In FIG. FIG. 16 is a two-dimensional graph showing the overall evaluation values shown in FIG. 15, with the evaluation values shown in FIG. 13 on the horizontal axis and the evaluation values shown in FIG. 14 on the vertical axis. FIG. 17 is a two-dimensional graph showing a distribution of comprehensive evaluation values r different from FIG. FIG. 18 is a three-dimensional graph showing the distribution of comprehensive evaluation values. FIG. 19 is a three-dimensional graph showing the distribution of comprehensive evaluation values from an angle different from that in FIG. FIG. 20 is a three-dimensional graph showing the distribution of comprehensive evaluation values from an angle different from that in FIG. FIG. 21 is a diagram showing an example of a three-dimensional graph showing a part of the distribution of comprehensive evaluation values on the r12b side where the comprehensive evaluation value r is higher. FIG. 22 is a diagram showing an example of a three-dimensional graph showing a part of the distribution of comprehensive evaluation values on the r12b side where the comprehensive evaluation value r is higher. FIG. 23 is a schematic diagram showing an example of the relationship between data analysis by the k-means method and comprehensive evaluation values.

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a diagram showing a main configuration example of an evaluation system 1 according to an embodiment. The evaluation system 1 is a system that evaluates the operating status of the plant 2 .

As shown in FIG. 1, the evaluation system 1 includes a plurality of sensors 4a, 4b, 4c, 4d, . The plant 2 includes a plurality of facilities 3a, 3b, . The facilities 3a, 3b, . . . function individually, so that the plant 2 functions as a whole. As a specific example, if the plant 2 is a hydroelectric power plant, the facilities 3a, 3b, . The specific form of the facilities 3a, 3b, . . . will correspond to the purpose of the plant 2. When comprehensively describing the facilities 3a, 3b, . The number of facilities 3 may be three or more.

The facility 3a shown in FIG. 1 includes a plurality of facilities 9a, 9b, . Moreover, the facility 3b shown in FIG. 1 includes a plurality of facilities 9c, 9d, . Facilities 9a, 9b, 9c, 9d, . For example, when the facility 3a is a power generator, the specific forms of the facilities 9a, 9b, . When the facilities 9a, 9b, 9c, 9d, . Each facility 3 thus includes a plurality of facilities 9 . A more detailed specific example of the equipment 9 will be described later with reference to FIG.

Also, the facilities 3a, 3b, . . . include a plurality of sensors 4a, 4b, 4c, 4d, . That is, the plurality of sensors 4a, 4b, 4c, 4d, . A plurality of sensors 4a, 4b, 4c, 4d, . The specific aspects of each of the sensors 4a, 4b, 4c, 4d, . When the sensors 4a, 4b, 4c, 4d, .

A detailed example of the sensor 4 will be described later with reference to FIG. In FIG. 1, one sensor 4 is provided for one facility 9, but a plurality of sensors 4 may be provided for one facility 9. A sensor 4 may be provided so as to be detectable by the sensor 4 . Items common to the plurality of facilities 9 include, for example, the weather, temperature, humidity, etc. of the area where the plant 2 is located.

The number of sensors 4a, 4b, 4c, 4d, . However, the plant 2 includes a plurality of facilities 3 each including a plurality of installations 9 and a plurality of sensors 4 .

In the embodiment, as shown in FIG. 1, a facility 3a is provided with a communication unit 5a. A communication unit 5b is provided in the facility 3b. When the communication units provided in each facility 3 such as the communication units 5a and 5b are not distinguished, they may be referred to as the communication unit 5 in some cases.

A communication unit 5 transmits the output of the sensor 4 to the evaluation device 10 . The communication unit 5 according to the embodiment has a circuit or the like for functioning as a NIC (Network interface controller) compatible with a predetermined communication protocol. Examples of the communication protocol include TCP/IP (Transmission Control Protocol/Internet Protocol), but the applicable communication protocol is not limited to this and can be changed as appropriate. The communication unit 5 transmits data generated according to the output of the sensor 4 to the evaluation device 10 via communication with the evaluation device 10 .

Hereinafter, the term "output data" refers to data generated according to the output of the sensor 4. FIG. The output data includes information indicating sensing values detected by the sensor 4 . A sensing value is a value derived according to the state of the equipment 9 provided with the sensor 4 . For example, if the sensor 4 is a temperature sensor, the sensing value indicates temperature (eg temperature in degrees Celsius). Moreover, when the sensor 4 is a vibration sensor, the sensing value indicates the magnitude of vibration. In addition, the specific content of the sensing value will depend on the specific mode of the sensor 4 .

In FIG. 1, one communication unit 5 is shown for one facility 3. However, some or all of the plurality of sensors 4a, 4b, . . . Arrangements may be provided for communicating with. Further, although not shown, when the output of the sensor 4 is analog, an analog/digital conversion circuit for generating output data that can be transmitted via digital communication by the communication unit 5 is provided between the sensor 4 and the communication unit 5 or It is provided in the sensor 4 .

In addition, among the configurations of the plant 2 , the configurations included in the evaluation system 1 according to the embodiment are the sensor 4 and the communication unit 5 . Facility 3 and facility 9 are illustrated to explain the configuration in which sensor 4 is provided in plant 2 .

The evaluation device 10 evaluates the operation status of the plant 2 . As shown in FIG. 1, the evaluation device 10 includes a communication unit 11, an acquisition unit 12, an evaluation unit 13, and a notification unit . The communication unit 11 has a circuit etc. for functioning as a NIC corresponding to the communication protocol of the communication unit 5 . The communication unit 11 communicates with the communication unit 5 and receives output data. In the embodiment, the communication line between the communication section 5 and the communication section 11 is a wireless communication line using electromagnetic waves such as radio waves. Thus, in the embodiment, the plurality of sensors 4a, 4b, 4c, 4d, . . .

Acquisition unit 12 acquires sensing values individually output by a plurality of 4a, 4b, 4c, 4d, . . . Specifically, the acquisition unit 12 acquires the sensing value by acquiring the output data received by the communication unit 11 . The evaluation unit 13 evaluates the operation status of the plant 2 based on the sensing values acquired by the acquisition unit 12 . The reporting unit 14 reports the evaluation of the operation status of the plant 2 evaluated by the evaluating unit 13 .

The evaluation device 10 according to the embodiment is an information processing device provided to function as a communication unit 11, an acquisition unit 12, an evaluation unit 13, and a notification unit 14. FIG. A configuration example of such an information processing apparatus will be described with reference to FIG.

FIG. 2 is a block diagram showing a configuration example of the evaluation device 10. As shown in FIG. The evaluation device 10 includes a communication section 11 , a storage section 31 , a calculation section 32 , an input section 33 and an output section 34 . The storage unit 31 stores software programs and data read in various processes performed by the evaluation device 10 . Hereinafter, when it is described as a program, etc., it refers to the software program and data. Specifically, the storage unit 31 includes a storage device such as a hard disk drive (HDD), a solid state drive (SSD), or a flash memory, and stores the data in the storage device. do.

Although not shown, the terminal 20 (see FIG. 1) also includes a communication unit, a storage unit, and a storage unit that can function in the same manner as the communication unit 11, the storage unit 31, the calculation unit 32, the input unit 33, and the output unit 34 included in the evaluation device 10. It is an information processing apparatus including a unit, a calculation unit, an input unit, and an output unit. Specifically, the terminal 20 is, for example, a portable terminal such as a smart phone or a tablet, but is not limited to this, and may be a stationary PC (Personal Computer).

In FIG. 2, an evaluation program 31a, sensing data 31b, and evaluation criteria data 31c are illustrated as examples of programs and the like. The evaluation program 31 a is a software program for causing the calculation unit 32 to function as the acquisition unit 12 and the evaluation unit 13 . The sensing data 31b is cumulative data obtained by accumulating output data acquired over time. The evaluation criteria data 31c is data referred to when evaluating based on the sensing data 31b.

The calculation unit 32 includes a calculation device such as a CPU (Central Processing Unit), functions as the acquisition unit 12 and the evaluation unit 13 by calculation processing using the above-described programs, etc., and performs various types of processing. Further, in the embodiment, the calculation unit 32 generates and updates the sensing data 31b in accordance with acquisition of output data. As a result, the calculation unit 32 can acquire a sensing value at an arbitrary timing in the past from the sensing data 31b.

The input unit 33 has a configuration for receiving input operations from the administrator of the evaluation device 10 . Such a configuration includes, for example, a keyboard, a mouse, etc., but is not limited to this, and various input devices can be employed.

The output unit 34 performs various outputs according to the content of processing performed by the calculation unit 32 . The output unit 34 includes, for example, a display output unit 34a, an audio output unit 34b, and the like. The display output unit 34a has one or more display devices such as a liquid crystal display and an organic EL (Electro Luminescence) display, and displays an image according to the processing performed by the calculation unit 32. FIG. The audio output unit 34b has a configuration capable of outputting audio, such as a speaker, and outputs audio according to the processing performed by the calculation unit 32. FIG. The display output unit 34 a and the audio output unit 34 b function as the notification unit 14 .

The processing contents of the evaluation unit 13 will be sequentially described below. First, the evaluation unit 13 performs conversion processing. Conversion processing is processing for converting a sensing value into an evaluation value.

The conversion process will be described below with reference to FIGS. 3 and 4. FIG. FIG. 3 is a graph showing an example of the relationship between sensing values and evaluation values. In FIG. 3, the distribution of the sensing values is shown by a histogram, and the evaluation values corresponding to the sensing values are shown by a line graph.

The sensing values exemplified in FIG. 3 are normal sensing values employed as the basis for deriving the evaluation values. The evaluation value is determined based on the distribution of normal sensing values indicated by the histogram. A normal value range for the sensing value can be derived by kernel density estimation based on the distribution.

Specifically, most of the sensing values illustrated in FIG. 3 are distributed within the range of 25-50. Here, let tau (τ) be the median value of the sensing values indicated by the dashed-dotted line C1, and sigma (σ) be the variance of the sensing values. In this case, the lower limit E1 of the normal range of sensing values can be expressed as τ-σ. Also, in this case, the upper limit E2 of the normal value range of the sensing value can be expressed as τ+σ. When a sensing value outside the normal value range of the sensing value from the lower limit E1 to the upper limit E2 is detected, it is suggested that the facility 9 provided with the sensor 4 that outputs the sensing value is likely to have some kind of abnormality. be.

The evaluation value of the embodiment is expressed as a percentage, and the higher the value, the greater the possibility that some abnormality has occurred in the equipment 9 . For example, when the sensing value falls below the lower limit E1, the evaluation value is derived as a value exceeding zero. FIG. 3 illustrates a setting in which the evaluation value gradually increases as the sensing value decreases from τ-σ to τ-3σ. Further, FIG. 3 shows τ−3σ as the lower evaluation limit E3. Also, when the sensing value exceeds the upper limit E2, the evaluation value is derived as a value exceeding zero. FIG. 3 illustrates a setting in which the evaluation value gradually increases as the sensing value increases from τ+σ to τ+3σ. In addition, FIG. 3 shows τ+3σ as the evaluation upper limit E4. Thus, in the embodiment, within the range from the evaluation lower limit E3 to the lower limit E1 and within the range from the upper limit E2 to the evaluation upper limit E4, the evaluation value increases according to the degree of deviation of the sensing value from the normal value range. is set to That is, the larger the evaluation value, the more the sensing value deviates from the normal value range, and the higher the possibility that some abnormality has occurred in the facility 9 provided with the sensor 4 that outputs the sensing value. It is suggested. When a sensing value equal to or lower than the evaluation lower limit E3 or a sensing value equal to or higher than the evaluation upper limit E4 is obtained, the evaluation value saturates at the maximum (100).

Thus, the evaluation value of the embodiment is a value that indicates the operating status of the equipment 9 . The evaluation value of the embodiment is derived as a percentage from the most stable state (0) including the normal operation of the equipment 9 to the most alarming state (100) including the abnormal operation of the equipment 9.

The evaluation unit 13 converts the sensing values into evaluation values based on the median value (τ) and variance (σ). The process of converting the sensing value into the evaluation value in this way functions as "the first process of calculating the first evaluation value indicating the stability of the equipment 9 based on the plurality of sensing values output by the plurality of sensors 4". . Data indicating the median (τ) and the variance (σ) are individually provided for each of the plurality of sensors 4 . Such data is included in the evaluation criteria data 31c. Although the sensing values shown in FIG. 3 are examples of the amount of vibration of the bearing provided in the generator, the sensing values are not limited to this example. For example, when the sensor 4 is a temperature sensor, the temperature of the sensing target is obtained as the sensing value. Moreover, when the sensor 4 is a pressure sensor, a value indicating the detected pressure is obtained as the sensing value. A sensing value corresponding to a specific correspondence of the sensor 4 is obtained without being limited to these. The median value (τ) of the sensing values is a value corresponding to the combination of the detection item of each sensor 4 and the detection target. The variance (σ) is a value corresponding to the combination of the detection item of each sensor 4 and the detection target. The evaluation unit 13 individually performs a process of converting sensing values into evaluation values for each of the plurality of sensors 4 .

The setting of the lower limit E1 and upper limit E2 of the normal value range and the evaluation lower limit E3 and evaluation upper limit E4 based on the median value (τ) and variance (σ) of the sensing values is merely an example and is limited to this. However, coefficients and specific formulas can be changed as appropriate.

FIG. 4 is a graph showing an example of the relationship between time-series changes in sensing values and time-series changes in evaluation values. Here, it is assumed that the sensing values shown in FIG. 4 and the sensing values shown in FIG. 3 are sensed by the same equipment 9 .

As shown in FIG. 4, the sensing value in the time period before time period T1 is within the normal value range of 25 to 50, and is generally stable at a value significantly below 50. The evaluation value for this time period is a relatively low value even if it is 0 or a value exceeding 0. On the other hand, in the time period T1, the sensing value significantly exceeds 50, then falls below 25, remains below 25 for a while, and then rises to suddenly exceed 50. changing erratically. The evaluation value in such time period T1 significantly exceeds 0, suggesting the possibility that some abnormality has occurred in the facility 9 provided with the sensor 4 that outputs the sensing value.

In most of the blank time period T2 that occurs after the time period T1, the sensing value is 0, but it was determined that there was some abnormality in the equipment 9 based on the evaluation value in the time period T1. Due to this, the facility 9 or the plant 2 including the facility 9 was stopped. During shutdown of the plant 2, no evaluation value is derived either. The output of sensing values and the derivation of evaluation values are restarted by restarting the plant 2 including the equipment 9 after the blank time zone T2. In the example shown in FIG. 4, the sensing value and the evaluation value temporarily fluctuate immediately after restarting operation, but are stable thereafter.

Output of sensing values and derivation of evaluation values are performed for each sensing value of the plurality of sensors 4a, 4b, 4c, 4d, . Here, in the embodiment, derivation processing for deriving the evaluation value for each sensing value as the first evaluation value is defined as the first processing. That is, the evaluation unit 13 performs the first process for each piece of equipment 9 based on multiple sensing values output by the multiple sensors 4 . The first evaluation value is an evaluation value indicating the stability of the equipment 9 .

In addition, the evaluation unit 13 of the embodiment periodically performs the first processing according to a predetermined processing cycle. The time length of the processing cycle is arbitrary, but it is desirable that it be a short period (for example, less than 1 second) so that when an abnormality occurs in the equipment 9, the evaluation value immediately indicates the abnormality.

(Evaluation example 1)
Hereinafter, an example of a series of processes related to evaluation of the plant 2 in the embodiment will be described as evaluation example 1 with reference to FIGS. 5 to 9. FIG. FIG. 5 is a diagram showing an example of a correspondence relationship between a plurality of facilities 3 included in the plant 2, items sensed by the facilities 3, facilities 9 to be sensed, and the like. Although FIG. 5 exemplifies the correspondence when the plant 2 is a hydroelectric power plant, it is not limited to this. Each item shown in FIG. 5 corresponds to the plant 2 .

The sensing value of the processing part P01 in FIG. 5 becomes the evaluation value of the processing part P03 by the processing of the processing part P02, which is the first processing.

As shown in FIG. 5, the plant 2 (see FIG. 1) includes generator, water turbine and waterway installations 3 (FIG. 1). The generator is connected to the water wheel and generates power according to the rotation of the water wheel. A water wheel rotates according to the flow of water supplied using a water channel. Conduit refers to the supply and discharge of the water concerned.

The generator is connected to the water wheel and generates power according to the rotation of the water wheel. A water wheel rotates according to the flow of water supplied using a water channel. Conduit refers to the supply and discharge of the water concerned. The generator senses one or more sensing items, such as temperature and vibration. In water turbines, sensing is performed for items such as water pressure and vibration. In a waterway, sensing is performed for the water level item. In this way, a sensing item is a group of sensing values, and is classified according to the type of sensing value.

As mentioned above, each facility 3 includes multiple facilities 9 . In the embodiment, as shown in FIG. 5, a plurality of facilities 9 are subject to sensing for each sensing item. The equipment 9 in FIG. 5 is an example and is not limited to this. It suffices if there is at least one facility 9 to be sensed for each sensing item.

Specifically, FIG. 5 shows, as equipment 9 for temperature sensing in the generator, a stator coil, cooling water for water-cooling the generator, and cooling air for air-cooling the generator. and are exemplified. Further, FIG. 5 illustrates two main bearings (upper bearing and lower bearing) among a plurality of bearings that rotatably support the turbine shaft of the generator as equipment 9 for sensing vibrations in the generator. ing. Moreover, FIG. 5 exemplifies an iron pipe that guides water supplied to the water turbine and a casing that houses the water turbine as equipment 9 for sensing water pressure in the water turbine. Sensing the water pressure refers to sensing the water pressure inside these facilities 9 . Moreover, FIG. 5 exemplifies one main bearing among the bearings that rotatably supports the water turbine and an upper cover as equipment 9 for sensing vibrations in the water turbine. The upper cover is a structure provided to cover the casing of the water turbine from above. In addition, FIG. 5 illustrates, as equipment 9 for sensing the water level in the waterway, a dam that stores and discharges water to be supplied to the water turbine, and a water outlet into which the water discharged from the water turbine flows. Sensing the water level refers to sensing the water level of the water that flows into these facilities 9 and is stored.

In this way, a sensing value is obtained from the plant 2 for each item sensed by the sensor 4 of each facility 9 . When the sensing item is temperature, the sensor 4 that outputs the sensing value is a temperature sensor. When the sensing item is vibration, the sensor 4 that outputs the sensing value is a vibration sensor. Sensors 4 corresponding to other sensing values are similarly provided for other sensing values.

Sensing values se01 to se03 are sensing values of the temperature item. The sensing value se01 changes according to the temperature of the generator stator. The sensing value se02 changes according to the temperature of the cooling water of the generator. The sensing value se03 changes according to the temperature of the cooling air for the generator.

A sensing value se04 and a sensing value se05 are sensing values of the vibration item. The sensing value se04 changes according to the vibration of the upper bearing of the generator. The sensing value se05 changes according to the vibration of the lower bearing of the generator.

A sensing value se06 and a sensing value se07 are sensing values of the water pressure item. The sensing value se06 changes according to the water pressure of the iron pipes of the water turbine. The sensing value se07 changes according to the water pressure in the casing of the water turbine.

A sensing value se08 and a sensing value se09 are sensing values of the vibration item. The sensing value se08 changes according to the vibration of the bearings of the water turbine. The sensing value se09 changes according to the vibration of the upper cover of the water turbine.

A sensing value se10 and a sensing value se11 are sensing values of the water level item. The sensing value se10 changes according to the water level of the dam of the channel. The sensing value se11 changes according to the water level of the outlet of the channel.

For example, in FIG. 5, the evaluation value _x1 is derived from the sensing value se01, the evaluation value _x2 is derived from the sensing value se02, and the evaluation value _x3 is derived from the sensing value se03. These evaluation values x ₁ , x ₂ and x ₃ are respectively derived by the first processing performed by the evaluation unit 13 .

Thus, the evaluation values x ₁ , x ₂ , . . . , x _n are, for example, the sensing values Se01, . , is derived from the sensing value sen. That is, n corresponds to, for example, the number of facilities 9 that are targets of one sensing item in one facility 3 . To give a specific example, n=3 (evaluation values x ₁ , x ₂ , x ₃ ) from the viewpoint of sensing the temperature item in the generator shown in FIG.

Next, the evaluation unit 13 calculates a comprehensive evaluation value r based on the multiple evaluation values x ₁ , x ₂ , . . . , _xn . A method of calculating the comprehensive evaluation value r will be described below with reference to FIGS. 5 and 6. FIG.

FIG. 6 is a graph schematically showing a calculation example of the overall evaluation value r based on the two evaluation values _x1 and _x2 . In the description with reference to FIG. 6, the evaluation of the plant 2 is represented by the comprehensive evaluation value r.

The evaluation unit 13 calculates a comprehensive evaluation value r based on the following formula (1). Here, x can be expressed as in Equation (2). _{x 1 ,} x _{2 , .} Also, p in Equation (1) is a weighted value of the Lp norm, and is set to a real number of 1 or more.

The comprehensive evaluation value r calculated based on the formulas (1) and (2) is greater than or equal to the maximum value of the multiple evaluation values referred to in the calculation. In the following description of the comprehensive evaluation value r, when the term “maximum evaluation value” is used, each of the plurality of facilities 3 referred to in calculating the comprehensive evaluation value r based on the formulas (1) and (2) of the evaluation values (x ₁ , x ₂ , . . . , x _n ). For example, when only one evaluation value is greater than 0 and all the other evaluation values are 0, the overall evaluation value r is equal to the one evaluation value according to Equation (1). On the other hand, when multiple evaluation values are greater than 0, according to Equation (1), the total evaluation value r is a value exceeding each of the multiple evaluation values greater than 0.

n in equations (1) and (2) corresponds to the number of evaluation values. As will be described later, p in Expression (1) is appropriately set to a value (for example, a first value u, a second value v, and a third value w, which will be described later) according to the evaluation stage. In order to facilitate understanding, the case where n=2 and p=2 will be described below with reference to FIG. 6 as an example.

As shown in FIG. 6, when n=2 and p=2, the evaluation value _x1 is 30, and the evaluation value _x2 is 40, the evaluation unit 13 calculates the overall evaluation value r =50. In the calculation example shown in FIG. 6, the “maximum evaluation value” of the two evaluation values x ₁ and x ₂ is the evaluation value x ₂ , and the difference between the overall evaluation value r and the evaluation value x ₂ is 10. . Thus, when the two evaluation values x ₁ and x ₂ are both greater than 0, the comprehensive evaluation value r is a value exceeding the two evaluation values x ₁ and x ₂ . Here, in the embodiment, the larger the evaluation value, the greater the possibility that some abnormality has occurred in the object from which the evaluation value was derived. Therefore, when there are multiple pieces of equipment 9 with large evaluation values, the evaluation unit 13 lowers the evaluation of the operation status of the plant 2 below the evaluation indicated by the evaluation value of a single piece of equipment 9 . In the embodiment, when the "evaluation value" is large, the "evaluation" decreases, and when the "evaluation value" is small, the "evaluation" increases. There is a correlation of

As the value of p approaches 1, the overall evaluation value r tends to increase relative to the “maximum evaluation value”. More specifically, when the value of p is 1 or a value close to 1 (for example, less than 2), the greater the number of facilities 3 from which non-zero evaluation values are derived, the more the “maximum evaluation value” Comprehensive evaluation value r is relatively large. An example of such a case will be described with reference to FIG.

FIG. 7 is a graph schematically showing an example in which p=1 in calculating the overall evaluation value r of the plant 2 based on the two evaluation values x ₁ and x ₂ . When n=2 and p=1, the evaluation value _x1 is 30, and the evaluation value _x2 is 40, the evaluation unit 13 calculates the overall evaluation value r=70 by Equation (1). In the calculation example shown in FIG. 7, the “maximum evaluation value” of the two evaluation values x ₁ and x ₂ is the evaluation value x ₂ , and the difference between the overall evaluation value r and the evaluation value x ₂ is 30. ing.

FIG. 8 is a graph schematically showing an example of setting p=10 in calculating the overall evaluation value r of the plant 2 based on the two evaluation values x ₁ and x ₂ . As the value of p increases, there is a tendency that the overall evaluation value r and the "maximum evaluation value" are closer to each other. Description will be made below with reference to FIG.

When n = 2 and p = 10, and one of the two evaluation values x ₁ and x ₂ has an evaluation value x ₁ of 30 and the other evaluation value x ₂ of 40, the evaluation unit 13 uses the formula Calculate r=40.2197415 by (1). In the calculation example shown in FIG. 8, the “maximum evaluation value” of the two evaluation values x ₁ and x ₂ is the evaluation value x ₂ , and the difference between the overall evaluation value r and the evaluation value x ₂ is 0.2197415. It has become.

7, 8, and FIG. 10, which will be described later, are diagrams corresponding to the range of the limited area L where the evaluation values _x1 and _x2 are 0 or more among the areas shown in FIG. In the embodiment, the evaluation values x ₁ , x _{2 ,} .

In FIG. 9, n=2, the evaluation value _x1 is 10, _x2 is 10, and the total evaluation value r for each different p value is calculated based on the above formula (1), described. When p=1, the comprehensive evaluation value r is 20.0. When p=2, the comprehensive evaluation value r is 14.1. When p=10, the comprehensive evaluation value r is 10.7. When p=infinity (∞), the comprehensive evaluation value r is 10.0. For the purpose of simplifying the comparison of the comprehensive evaluation values r due to different values of p, the values of the comprehensive evaluation values r are rounded off to the second decimal place in FIG.

As described with reference to FIGS. 6 to 9, the relationship between the evaluation value and the comprehensive evaluation value r changes according to the value of the weighting value (p). That is, the larger the weighting value (p) exceeds 1, the smaller the difference between the "maximum evaluation value" and the comprehensive evaluation value r. Also, the closer the weighting value (p) is to 1, the greater the difference between the "maximum evaluation value" and the comprehensive evaluation value r.

For example, assuming that each of a plurality of evaluation values is not 0, there is a case where it is desired that the synergistic effect of these evaluation values appears in the total evaluation value r. In this case, by setting the weighting value (p) to 1 or a value closer to 1, the total evaluation value r can be calculated as a significantly larger value than each of the plurality of evaluation values. Such a way of thinking can be applied, for example, to the calculation of the comprehensive evaluation value r of the entire plant 2 based on the evaluation values of each of the plurality of facilities 3 . In other words, when an abnormality occurs in a plurality of facilities 3 included in the plant 2, compared with the case where an abnormality occurs in one facility 3 among the plurality of facilities 3, the total evaluation value indicating the abnormality of the entire plant 2 It becomes easier to calculate r as a larger value. Therefore, a situation in which one facility 3 out of the plurality of facilities 3 has an abnormality and a situation in which each of the plurality of facilities 3 has an abnormality can be distinguished from each other by the magnitude of the comprehensive evaluation value r. As a result, it is possible to grasp the situation that an abnormality has occurred in each of the plurality of facilities 3 based on the magnitude of the comprehensive evaluation value r.

On the other hand, if multiple factors are not considered, the weighting value (p) is set to a larger value. For example, in the comprehensive evaluation value r of the temperature of the facility 9 sensed by the generator, it is sufficient to know whether there is a facility 9 out of the normal temperature range and the degree of deviation from the normal temperature range. In this case, since the evaluation value is 0 if the temperature is within the normal temperature range, it is possible to determine whether or not there is equipment 9 outside the normal temperature range depending on whether the overall evaluation value r is 0 or not. Also, the "degree of deviation from the normal temperature range" can be grasped from the magnitude of the comprehensive evaluation value r. Such a concept can be applied to the calculation of the comprehensive evaluation value r of the sensing item.

Based on this way of thinking, the second process, the third process, and the fourth process, which will be described later, use the same equations (1) and (2), but are set with different weighting values (p). ing.

The evaluation unit 13 performs the process of calculating the comprehensive evaluation value r multiple times based on the above formulas (1) and (2). Specifically, the evaluation unit 13 performs a first process, a second process, a third process, and a fourth process. As described above, the first process is the process of converting the sensing value into the evaluation value by the conversion process described above. In the second process, the overall evaluation value r calculated in the first process is selected for each sensing item, and the selected overall evaluation value r calculated in the first process is used as the evaluation value x ₁ , x ₂ , . . . , _xn .

In other words, in the second process, the first evaluation values are classified for each type of sensing value in the facility 3 including the plurality of facilities 9 and treated as a group of the first evaluation values. In FIG. 5, temperature and vibration are sensing items in a generator, for example. Therefore, the generator temperature evaluation values x ₁ , x ₂ and x ₃ and the generator vibration evaluation values x ₁ and x ₂ are classified and treated as different groups. In addition, water pressure and vibration are sensing items for water turbines. Therefore, the evaluation values x ₁ and x ₂ of the water pressure of the water turbine and the evaluation values x ₁ and x ₂ of the vibration of the water turbine are classified and treated as different groups. The sensing item for waterways is the water level, and there are no other sensing items. Therefore, the channel water level evaluation values x ₁ and x ₂ are grouped.

In the third process, the comprehensive evaluation value r calculated in the second process is selected for each facility 3, and the selected comprehensive evaluation value r calculated in the second process is used as the evaluation value x ₁ , x ₂ , . . . , _xn . In the fourth process, all the comprehensive _evaluation values r calculated in the third process are used as the evaluation values x ₁ , x ₂ , .

First, an overview of the second process will be described with reference to FIG. The value before the second processing is represented by processing part P03 in FIG. The second process is represented by process part P04 in FIG. The value after the second processing is represented by processing part P05 in FIG. In the example shown in FIG. 5, the evaluation unit 13 evaluates the evaluation value of the processing part P03 of FIG. is subjected to the second processing. The weighting value (p) in the second process is the first value u. This is indicated by "Lp norm (p=u)" of processing part P04 in FIG.

First, the second processing for the temperature of the generator will be described. The evaluation unit 13 processes the evaluation values x ₁ , x ₂ and x ₃ derived from the temperature sensing values se01, se02 and se03 of the generator shown in FIG. Therefore, n in this case is 3. The evaluation unit 13 applies the evaluation values x ₁ , x ₂ , x ₃ , n=3, and p=u to the above equations (1) and (2) to calculate the overall evaluation value r. .

Next, the second processing for vibration of the generator will be described. The evaluation unit 13 processes the evaluation values x ₁ and x ₂ derived from the sensed vibration values se04 and se05 of the generator shown in FIG. Therefore, n in this case is two. The evaluation unit 13 applies the evaluation values x ₁ and x ₂ , n=2, and p=u to the above equations (1) and (2) to calculate the overall evaluation value r.

Next, the second process for the hydraulic pressure of the water turbine will be described. The evaluation unit 13 processes the evaluation values x ₁ and x ₂ derived from the sensing values se06 and se07 of the water pressure in the water turbine shown in FIG. Therefore, n in this case is two. The evaluation unit 13 applies the evaluation values x ₁ and x ₂ , n=2, and p=u to the above equations (1) and (2) to calculate the overall evaluation value r.

Next, the second processing for vibration of the water turbine will be described. The evaluation unit 13 processes the evaluation values x ₁ and x ₂ derived from the sensed vibration values se08 and se09 of the water turbine shown in FIG. Therefore, n in this case is two. The evaluation unit 13 applies the evaluation values x ₁ and x ₂ , n=2, and p=u to the above equations (1) and (2) to calculate the overall evaluation value r.

Next, the second processing for the water level of the channel will be described. The evaluation unit 13 processes the evaluation values x ₁ and x ₂ derived from the sensing values se10 and se11 of the water level in the channel shown in FIG. Therefore, n in this case is two. The evaluation unit 13 applies the evaluation values x ₁ and x ₂ , n=2, and p=u to the above equations (1) and (2) to calculate the overall evaluation value r.

Next, an overview of the third process will be described. The relationship between the third processing and the values before and after the third processing is represented, for example, within the range from processing part P07 to processing part P09 in FIG. As described above, the third process is a process of calculating the comprehensive evaluation value r indicating the stability of each facility 3 . In the case of the example shown in FIG. 5, the evaluation unit 13 individually performs the third process for the generator, water turbine, and waterway. The weighting value (p) in the third process is the second value v. This is indicated by "Lp norm (p=v)" of processing part P08 in FIG.

Next, the third processing for each facility 3 will be described with reference to FIG. First, the 3rd process about a generator is demonstrated. The evaluation unit 13 treats the overall evaluation value r of the generator temperature calculated in the second process as the evaluation value _x1 in the third process. Also, the evaluation unit 13 treats the overall evaluation value r of the vibration of the generator as the evaluation value _x2 in the third process. Therefore, n in this case is two. In _this way, the total evaluation value r calculated in the second process is treated as the evaluation values x ₁ , x ₂ , . showing. The evaluation unit 13 applies the evaluation values x ₁ and x ₂ , n=2, and p=v to the above equations (1) and (2) to calculate the comprehensive evaluation value r.

Next, the 3rd process about a water turbine is demonstrated. The evaluation unit 13 treats the comprehensive evaluation value r of the hydraulic pressure of the water turbine calculated in the second process as the evaluation value _x1 in the third process. In addition, the evaluation unit 13 treats the comprehensive evaluation value r of vibration of the water turbine as the evaluation value _x2 in the third process. Therefore, n in this case is two. The evaluation unit 13 applies the evaluation values x ₁ and x ₂ , n=2, and p=v to the above equations (1) and (2) to calculate the comprehensive evaluation value r.

Next, the 3rd process about a waterway is demonstrated. The evaluation unit 13 treats the comprehensive evaluation value r of the water level of the channel calculated in the second process as the evaluation value _x1 in the third process. Therefore, n in this case is 1. The evaluation unit 13 applies the evaluation value _x1 , n=1, and p=v to the above equations (1) and (2) to calculate the comprehensive evaluation value r. Note that when n=1, the comprehensive evaluation value r is the same as the evaluation value _x1 .

Next, an overview of the fourth process will be described. The value before the fourth processing is represented by processing part P11 in FIG. The fourth process is represented by process part P12 in FIG. The value after the fourth processing is represented by processing part P13. The weighting value (p) in the fourth process is the third value w. This is indicated by "Lp norm (p=w)" of processing part P12 in FIG.

In the case of the example shown in FIG. 5, the plant 2 includes generators, water turbines, and water channels as a plurality of facilities 3 . The evaluation unit 13 treats the overall evaluation value r of the generator calculated in the third process as the evaluation value _x1 . The evaluation unit 13 also treats the overall evaluation value r of the water turbine calculated in the third process as the evaluation value _x2 . Also, the evaluation unit 13 treats the comprehensive evaluation value r of the waterway calculated in the third process as the evaluation value _x3 . Therefore, n in this case is 3. The evaluation unit 13 applies the evaluation values x ₁ , x ₂ , x ₃ , n=3, and p=w to the above equations (1) and (2) to calculate the comprehensive evaluation value r. . A comprehensive evaluation value r shown in the processing part P13 of FIG. 5 indicates the stability of the plant 2 .

A first value u that is the weighting value (p) in the second process, a second value v that is the weighting value (p) in the third process, and a third value w that is the weighting value (p) in the fourth process , are predetermined respectively. Data indicating the first value u, the second value v, and the third value w are included in the evaluation criteria data 31c. Here, the first value u that is the weighting value (p) in the second process, the second value v that is the weighting value (p) in the third process, and the third value v that is the weighting value (p) in the fourth process The value w is different. As a result, the weighting of the overall evaluation value r for each sensing item in the stability evaluation of the plant 2 and the weighting of the overall evaluation value r for each facility 3 are made different from the weighting of the overall evaluation value r of the plant 2. be able to.

The relationship between the first value u, the second value v, and the third value w in the embodiment is u>v>w. When the value of p is 1 or a value close to 1 (for example, less than 2), the more evaluation values that are not 0, the larger the comprehensive evaluation value r is relative to the “maximum evaluation value”. Here, the case where a plurality of non-zero evaluation values x ₁ , x _{2 ,} . This is the case when it has occurred.

First, an event example in which the comprehensive evaluation value r of the plant 2 is calculated by the fourth process will be given. In such an event example, the evaluation values x ₁ , x _{2 ,} . Here, it _is assumed that some kind of abnormality occurs in a plurality of facilities 3, and a plurality of evaluation values x ₁ , x ₂ , . In this case, there is a high possibility that the plant 2 is in a more serious situation than in other cases where one facility 3 out of the plurality of facilities 3 is malfunctioning. In this case, it is desirable to represent the seriousness of the situation of the plant 2 by increasing the comprehensive evaluation value r more than in other cases. Therefore, in the embodiment, the weighting value in calculating the comprehensive evaluation value r of the plant 2 is set to a smaller weighting value (for example, p=w). As a result, a larger comprehensive evaluation value r is calculated in this case. As a result, when some kind of abnormality occurs in a plurality of facilities 3, the serious situation of the plant 2 can be represented appropriately.

Next, an event example in which the comprehensive evaluation value r of one facility 3 is calculated by the third process will be given. In this event example, the evaluation values x ₁ , x ₂ , _. reflected. Here, it _is assumed that a situation that can be regarded as abnormal from the viewpoint of a plurality of sensing items occurs, and a plurality of evaluation values x ₁ , x ₂ , . In this case, attention should be paid more attention than in other cases where an abnormality is occurring only in terms of one sensing item, but it is strictly related to one facility 3 . Here, the evaluation value of each sensing item at the stage of calculating the comprehensive evaluation value r of one facility 3 is higher than the evaluation value of each facility 3 at the stage of calculating the comprehensive evaluation value r of the plant 2. "Strength of relevance to overall evaluation" is considered to be weak. Therefore, in the embodiment, the weighting value (for example, p=v) in calculating the comprehensive evaluation value r of each facility 3 is set larger than the weighting value (for example, p=w) in calculating the comprehensive evaluation value r of the plant 2. are doing.

Next, an event example in which a comprehensive evaluation value r of one sensing item in a certain facility 3 is calculated by the second process will be given. In this event example, the evaluation values x ₁ , x _{2 ,} . 9 situation is reflected. Here, it is assumed that an abnormality occurs in a plurality of facilities 9 and a plurality of non-zero evaluation values x ₁ , x ₂ , . . . , x _n are included. In this case, more attention should be paid than in other cases in which an abnormality occurs only from the viewpoint of one sensing item. Here, the "evaluation value obtained by converting the sensing value" referred to at the stage of calculating the comprehensive evaluation value r for each sensing item is the sensing item at the stage of calculating the comprehensive evaluation value r of each facility 3. It is considered that the "strength of relevance to the evaluation of the entire plant 2" is weaker than the evaluation value for each plant. Therefore, in the embodiment, the weighting value (for example, p=u) in calculating the comprehensive evaluation value r of the sensing item is set larger than the weighting value (for example, p=v) in calculating the comprehensive evaluation value r for each facility 3. are doing.

In this way, by setting the weighting value (p), the overall evaluation value r that more appropriately considers the evaluation value referred to in each calculation process and the "strength of relationship with the evaluation of the entire plant 2" can be calculated in each of the second, third and fourth processes. In the embodiment, for example u=100, v=50, and w=25, but it is not limited to this. For example, w may be a value closer to 1 (eg, 5) or a value greater than 25. Other values may be smaller or larger. Also, the relationship between the first value u, the second value v, and the third value w does not necessarily have a relationship of u>v>w. When the relationship with the evaluation of the entire plant 2 is stronger, etc., u, v , w may be set.

A flow until the comprehensive evaluation value r of the plant 2 is calculated from the sensing values will be described with reference to FIG. First, as shown in processing part P01, sensors 4 individually perform sensing in each facility 9, and output sensing values se01 to se11. The acquisition unit 12 acquires the sensing values se01 to se11.

As indicated by the processing part P02, the evaluation unit 13 performs the conversion processing described above as the first processing. As a result, as shown in processing part P03, evaluation values corresponding to each of the sensing values se01 to se11 are derived.

The evaluation values of the processing part P03 are treated as evaluation values x ₁ , x ₂ , . . . , x _n of the second processing. As shown in processing part P04, the evaluation unit 13 performs a second process with p=u. As a result, as shown in processing part P05, a comprehensive evaluation value r for each sensing item of facility 3 is calculated. In FIG. 5, in the calculation of the total evaluation value r for each sensing item performed in the second process, the range related to the calculation process of each total evaluation value r is indicated by a thick rectangle within the range from the processing part P03 to the processing part P05. ing. The thick-line rectangles are individually indicated for each sensing item of the temperature of the generator, the vibration of the generator, the water pressure of the water turbine, the vibration of the water turbine, and the water level of the water channel.

As shown by the relationship between the processing part P05 and the processing part P07 sandwiching the processing part P06, the comprehensive evaluation value r calculated in the second processing is the evaluation values x ₁ , x ₂ , . , _xn . As indicated by processing part P08, the evaluation unit 13 performs a third processing with p=v. As a result, a comprehensive evaluation value r for each facility 3 is calculated as shown in processing part P09. In FIG. 5, in the calculation of the comprehensive evaluation value r for each facility 3 performed in the third process, the range related to the calculation process of each comprehensive evaluation value r is indicated by a thick rectangle within the range from the processing part P07 to the processing part P09. ing. The heavy-line rectangles are individually shown for each facility 3 of the generator, the water turbine, and the waterway.

As shown by the relationship between the processing part P09 and the processing part P11 sandwiching the processing part P10, the comprehensive evaluation value r calculated in the third processing is the evaluation values x ₁ , x ₂ , . , _xn . As shown in processing part P12, the evaluation unit 13 performs a fourth processing with p=w. As a result, the comprehensive evaluation value r of the plant 2 is calculated as shown in processing part P13. In FIG. 5, the range related to the calculation of the comprehensive evaluation value r of the plant 2 performed in the fourth process is indicated by a thick rectangle within the range from the process part P11 to the process part P13.

The processing related to the calculation of the comprehensive evaluation value r of the plant 2 is not limited to obtaining the comprehensive evaluation value r as a comprehensive index of the stability of the plant 2. It can also be used as a process for specifying specific contents. Hereinafter, evaluation example 2 different from evaluation example 1 described with reference to FIGS. 5 to 9 will be described with reference to FIG.

(Evaluation example 2)
FIG. 10 is a graph for explaining an example in which comprehensive evaluation values r1, r2, r3, and r4, which are different comprehensive evaluation values r, are calculated. In the description with reference to FIG. 10, evaluation values x ₁ and x ₂ are evaluation values of two facilities 3 .

The comprehensive evaluation value r1 of the plant 2 is the comprehensive evaluation value r when x ₁ =10 and x ₂ =10, and is approximately 11.5. The comprehensive evaluation value r2 of the plant 2 is the comprehensive evaluation value r when x ₁ =90 and x ₂ =10, which is approximately 90.0 (90.00030843). The comprehensive evaluation value r3 of the plant 2 is the comprehensive evaluation value r when x ₁ =20 and x ₂ =90, which is approximately 90.0 (90.0097525). The comprehensive evaluation value r4 of the plant 2 is the comprehensive evaluation value r when _x1 =90 and _x2 =90, which is approximately 103.4 (103.3828519). Thus, the comprehensive evaluation value r1, the comprehensive evaluation value r2, the comprehensive evaluation value r3, and the comprehensive evaluation value r4 are calculated from different evaluation values _x1 and _x2 , respectively. That is, when the comprehensive evaluation value r1 is calculated, when the comprehensive evaluation value r2 is calculated, when the comprehensive evaluation value r3 is calculated, and when the comprehensive evaluation value r4 is calculated, the two The situation of each facility 3 is different. Note that the description with reference to FIG. 10 assumes that p=5, but this is just an example and the value of p is not limited to this.

For example, the evaluation value _x1 is the evaluation value corresponding to the overall evaluation value r of the generator calculated in the third process, and the evaluation value _x2 corresponds to the overall evaluation value r of the water turbine calculated in the third process. Assume that it is an evaluation value. In this case, when the comprehensive evaluation value r4 is calculated in the fourth process, there is a high possibility that the plant 2 has a complex problem such as some kind of abnormality occurring in both the generator and the water turbine.

As in the above example, the evaluation value _x1 is the evaluation value corresponding to the overall evaluation value r of the generator calculated in the third process, and the evaluation value _x2 is the overall evaluation of the turbine calculated in the third process. Assume that it is an evaluation value corresponding to the value r. Here, when the comprehensive evaluation value r2 is calculated, there is a high possibility that some abnormality has occurred in the generator. Further, when the comprehensive evaluation value r3 is calculated, there is a high possibility that some abnormality has occurred in the water turbine. As described above, the facility 3 in which the abnormality actually occurs differs between when the comprehensive evaluation value r2 is calculated and when the comprehensive evaluation value r3 is calculated. On the other hand, from the viewpoint of the degree of stability of the plant 2 suggested by the magnitude of the comprehensive evaluation value r, both the comprehensive evaluation value r2 and the comprehensive evaluation value r3 are approximately 90.0, and there is almost no difference. This is because when viewed from the viewpoint of "the level at which the abnormality of one facility 3 should be dealt with", the time when the comprehensive evaluation value r2 is calculated is equivalent to the time when the comprehensive evaluation value r3 is calculated. .

In other words, when looking only at the magnitudes of the comprehensive evaluation value r2 and the comprehensive evaluation value r3 of “approximately 90.0”, the calculation of the comprehensive evaluation value r2 and the calculation of the comprehensive evaluation value r3 can be distinguished. not. Actually, when the comprehensive evaluation value r2 is calculated, it is considered that some measures should be taken for the generator. Moreover, when the comprehensive evaluation value r3 is calculated, it is considered that some measures should be taken for the water turbine. Therefore, in Evaluation Example 2, considering the case where the objects to be actually dealt with are different even if the values of the comprehensive evaluation value r are equal in terms of magnitude, what was the determinant factor of the value of the comprehensive evaluation value r? It is desirable to provide supplementary information on whether A specific example of such information provision will be described later with reference to FIG. 11 .

Note that the overall evaluation value r4 described above is approximately 103.4, which is greater than the overall evaluation value r2 and the overall evaluation value r3 (about 90.0). This is because when the comprehensive evaluation value r4 is calculated, it is "a level at which it is necessary to deal with anomalies in a plurality of facilities 3", so the comprehensive evaluation is at a "level at which it is sufficient to deal with anomalies in one facility 3". It is calculated as a value larger than the value r2 and the comprehensive evaluation value r3. The difference between the comprehensive evaluation value r4, the comprehensive evaluation value r2, and the comprehensive evaluation value r3 is similar to the concept of the comprehensive evaluation value r of the plant 2 in Evaluation Example 1 described above.

Note that when the overall evaluation value r1 is calculated based on the evaluation values x ₁ and x ₂ similar to the above example, when the overall evaluation value r2, the overall evaluation value r3, and the overall evaluation value r4 are calculated, In comparison, it can be said that there is no major problem with the stability of Plant 2.

As described above, the example of the value of n in the description with reference to FIG. 10 is n=2, but it is not limited to this. In fact, the evaluation unit 13 evaluates the value _of n, that is, the evaluation values x ₁ , x ₂ , . An evaluation value r can be calculated.

In addition, the estimation of the cause of the abnormality is not limited to three units of facilities, and can be performed in smaller units. For example, by setting the evaluation values x ₁ , x _{2 ,} . It can be estimated whether there is a problem related to the event. To give a more specific example, the evaluation value _x1 is the evaluation value corresponding to the overall evaluation value r of the temperature of the generator calculated in the second process, and the evaluation value _x2 is the evaluation value calculated in the second process. Assume that it is an evaluation value corresponding to the overall evaluation value r of the vibration of the generator. In this case, when the comprehensive evaluation value r2 is calculated, there is a high possibility that some abnormality has occurred in the generator that generates heat (for example, bearings, etc.) and that heat is being generated. Also, when the comprehensive evaluation value r3 is calculated, there is some abnormality in the structure that causes vibration in the generator (for example, a rotating mechanism such as a turbine shaft of the generator or a water wheel that is another facility 3 connected to the turbine). is likely to occur. In a similar way of thinking, by setting the evaluation values x ₁ , x _{2 ,} . can be estimated.

The specific configurations and numbers of the facilities 3, sensing items, and equipment 9 of the plant 2 illustrated in the descriptions with reference to FIGS. 5 to 10 are merely examples for explanation. The specific configuration of the plant 2 is not limited to this example, and can be made according to the actual state of the plant 2 .

In addition, the comprehensive evaluation value r may be a value exceeding the upper limit (100) of the percentage. This makes it easier to grasp the degree of abnormality when the plant 2 is abnormal.

Next, notification related to evaluation of the operating status of the plant 2 will be described. FIG. 11 is a diagram showing an example of notification relating to evaluation of the operating status of the plant 2. In FIG. The display output contents by the display output unit 34a include a first display area D1 and a second display area D2.

Sensing values indicated by the outputs of the plurality of sensors 4a, 4b, 4c, 4d, . . . are displayed in the first display area D1. In the first display area D1 shown in FIG. 11, the sensing value indicated by the output from each of the 18 sensors 4 is displayed in analog meter form, but this is an example of display form and is limited to this. not. The number and type of sensing values displayed in the first display area D1, the display format in the first display area D1 and the second display area D2, the layout, and other specific aspects are arbitrary. In addition, when the number of sensors 4 that cannot be displayed collectively in the first display area D1 is provided in the plant 2, a scroll bar or the like is additionally provided to enable selection of the sensing value to be displayed in the first display area D1. .

The second display area D2 includes a plant stability display portion M1 and a trend graph display portion M2. In the plant stability display section M1, a display corresponding to the comprehensive evaluation value r indicating the stability of the operating state of the plant 2 is performed. Specifically, for example, the latest value of the comprehensive evaluation value r indicating the soundness of the plant 2 is reflected in the plant stability display section M1. That is, the total evaluation value r calculated as the fourth evaluation value is displayed on the plant stability display section M1. As exemplified in FIG. 11, the plant stability display section M1 is displayed to be larger and more conspicuous than each sensing value included in the first display area D1.

In FIG. 11, the evaluation value is displayed in the form of an analog meter called "stability meter" in the plant stability display section M1, but this is an example of a form of display and is not limited to this. can be changed as appropriate. In the plant stability display section M1 shown in FIG. 11, the "warning level" is indicated when the comprehensive evaluation value r reaches 100[%]. The "alarm level" indicates whether an alarm is issued by the audio output unit 34b or the like to indicate that some kind of abnormality has occurred that threatens the stable operation of the plant 2, or the level of abnormality is such that an alarm is issued even if the alarm is not issued. It shows what happened. The analog meter of the plant stability display unit M1 exemplifies 120 [%] as a value exceeding 100 [%], but the exceeding value is not essential, and the value exceeding 100 [%] The value may be less than 120[%] or may be more than 120[%]. Conversely, it paradoxically indicates that the plant 2 is in an ideal stable state when the total stability is less than 100[%] and the fluctuation is small.

In addition, in FIG. 11, the character string "real time" is attached in the plant stability display portion M1. Such a character string is such that the overall evaluation value r displayed by the plant stability display unit M1 is for a short period of time (for example, (less than 1 second). The character strings "alarm level" and "real time" are not essential, and the specific contents of the character strings can be changed as appropriate.

The display of the plant stability display section M1 makes it easier for a person who confirms the display contents (manager, worker, etc. in the plant 2) to grasp the operation status of the plant 2. FIG. In addition, an additional display indicating whether the evaluation value displayed on the plant stability display section M1 is based on the evaluation value of a single facility 3 or the evaluation value of a plurality of facilities 3 is further provided by the plant It may be performed in the stability display section M1 or in the vicinity of the plant stability display section M1. This makes it easier to intuitively grasp the facility 3 that should be dealt with when the plant 2 has an abnormality.

The trend graph display section M2 shows time-series changes within a certain period of time in the plant stability display section M1. In the embodiment, the evaluation unit 13 causes the storage unit 31 to store at least data indicating evaluation values derived within the certain period of time, and refers to the data when displaying the trend graph display unit M2. In FIG. 11, the trend graph display section M2 shows a line graph of changes in the values of the plant stability display section M1 during the period from the time point when the display is performed to 48 hours before. The specific display mode of M2 can be changed as appropriate.

In the embodiment, the calculation unit 32 functions as the evaluation unit 13 and performs processing related to determination of display content (display data generation processing), but is not limited to this. The evaluation unit 13 may derive the evaluation value, and a dedicated configuration independent of the evaluation unit 13 may perform processing related to determination of notification content based on the evaluation value.

As shown in FIG. 11, the display output content by the display output unit 34a may further include content that is not included in the first display area D1 and the second display area D2. The display contents shown in FIG. 11 further include a real-time evaluation value display portion A1, an object name display portion A2, and a trend display portion A3 in addition to the first display area D1 and the second display area D2. In the real-time evaluation value display area A1, a list of some or all of the evaluation values of each sensing item of the equipment 9 exceeding 0 is displayed. In the real-time evaluation value display section A1, the evaluation value of the sensing item from which the higher evaluation value was derived may be preferentially displayed, or the higher comprehensive evaluation value r (third evaluation value) is derived. The evaluation value of each other sensing item provided in the facility 3 that has been selected may be preferentially displayed. The object name display area A2 displays information (such as a character string) indicating which equipment 9 and which item the sensing evaluation values listed in the real-time evaluation value display area A1 are obtained from. In the trend display area A3, a display is performed showing the chronological change in the evaluation value selected and displayed as the real-time evaluation value display area A1 within a certain period. In the embodiment, the evaluation unit 13 causes the storage unit 31 to store, as the sensing data 31b, data indicating the sensing evaluation value derived at least within the certain period of time, and refers to the sensing data 31b when displaying the trend display unit A3.

The display content of the first display area D1 may be associated with the estimation of the cause of the abnormality described with reference to FIG. 10 . For example, when there is a high possibility that an abnormality has occurred in a specific facility 3 included in the plant 2, the sensing value of the sensor 4 provided in the facility 9 included in the specific facility 3 has priority over the first display area D1. may be displayed as To give a specific example, when the overall evaluation value r2 is calculated in Evaluation Example 2 described above, the sensing value of the sensor 4 provided in the generator is preferentially displayed in the first display area D1. Further, when the comprehensive evaluation value r3 is calculated in the evaluation example 2 described above, the sensing value of the sensor 4 provided in the water turbine is preferentially displayed in the first display area D1.

Further, the comprehensive evaluation value r that can be reflected in the display contents of the plant stability display section M1 is not limited to the comprehensive evaluation value r as the fourth evaluation value. For example, the display mode of the comprehensive evaluation value r of the entire plant 2, the display mode of the comprehensive evaluation value r for each facility 3, and the display mode of the comprehensive evaluation value r for each sensing item of the facility 3 can be switched. may be provided. A setting input section for such switching may be further provided in the display output contents of the display output section 34a, or a real-time evaluation value displayed in the real-time evaluation value display section A1 or the first display area D1 may be clicked. By operating the display, the display may be switched to display in units of facilities 3 including facilities 9 provided with sensors 4 that output sensing values corresponding to the real-time values, or in units of sensing items of facilities 3 .

In the above, the description has been made with the notification by the output unit 34 included in the evaluation device 10 as an example, but the notification may be performed by the terminal 20 (see FIG. 1). As shown in FIG. 1 , the terminal 20 has a communication section 21 and a notification section 22 . The communication unit 21 has the same configuration as the communication unit 11, and evaluates the data corresponding to the display content of the display output unit 34a and the data corresponding to the audio output content of the audio output unit 34b by communicating with the communication unit 11. Acquired from the device 10 . The notification unit 22 has the same configuration as the output unit 34, and performs display output similar to that of the display output unit 34a described above and audio output similar to that of the audio output unit 34b. As described above, the terminal 20 includes the notification unit 22 that performs notification regarding the evaluation of the operation status of the plant 2 . Moreover, the notification unit 22 includes the same configuration as the display output unit 34a as a configuration for displaying information related to evaluation of the operating status of the plant 2 .

In FIG. 1, the communication unit 11 communicates with the communication unit 5 and the communication unit 21, but the configuration for communicating with the communication unit 5 and the configuration for communicating with the communication unit 21 may be separate configurations. .

In addition, there is air temperature as one of the standards of temperature. Therefore, the evaluation value may be 0 when the temperature is the same as the air temperature (for example, 20 [°C]), but the temperature that is the same as the air temperature is affected by the surrounding temperature. , especially depending on the season. Therefore, it may be conditioned in advance so that the upper limit or lower limit of the variance is changed depending on the season. In this case, data corresponding to the conditioning is included in advance in the evaluation criteria data 31c. In addition, the evaluation device 10 is preliminarily provided with a software program or circuit (such as a timer circuit) that functions as a calendar capable of specifying the season.

FIG. 12 is a graph showing an example of changing the upper limit of variance according to season. FIG. 12 assumes that the temperature is sensed by the sensor 4, which is a temperature sensor. In FIG. 12, a first upper limit E21 that is farther from the evaluation upper limit E4 and a second upper limit E22 that is closer to the evaluation upper limit E4 are individually set. Both the first upper limit E21 and the second upper limit E22 function as the upper limit E2 described above. That is, the variation amount SC between the first upper limit E21 and the second upper limit E22 is the seasonal variation amount of the upper limit E2. Although FIG. 12 shows only the first upper limit E21 and the second upper limit E22 simply, evaluation can be performed according to a software program or circuit (such as a timer circuit) that functions as a calendar that can specify the season. You may make it the part 13 change the upper limit E2 applied in multiple stages or steplessly between the 1st upper limit E21 and the 2nd upper limit E22.

(Evaluation example 3)
Although FIG. 10 describes the comprehensive evaluation value r at a certain point in time, the evaluation unit 13 causes the notification unit 14 to display information based on the chronological change in the comprehensive evaluation value r that changes with the passage of time. You may make it perform to the part 34a. The display output will be described below with reference to FIGS. 13 to 22. FIG.

FIG. 13 is a graph showing time series changes in the evaluation value of one piece of equipment 9 . FIG. 14 is a graph showing time series changes in the evaluation value of one piece of equipment 9 different from FIG. FIG. 15 is a graph showing time series changes in the comprehensive evaluation value r of the facility 3 having the two facilities 9 described with reference to FIGS. 16 is a two-dimensional graph showing the overall evaluation value r shown in FIG. 15, with the evaluation value shown in FIG. 13 as the horizontal axis and the evaluation value shown in FIG. 14 as the vertical axis.

Assume that the evaluation value represented by the graph shown in FIG. 13 is either the evaluation value _x1 or the evaluation value _x2 described with reference to FIG. Also, the evaluation value represented by the graph shown in FIG. 14 is the other of the evaluation value _x1 and the evaluation value _x2 described with reference to FIG. In this case, the comprehensive evaluation value r derived based on the evaluation value _x1 and the evaluation value _x2 is represented by the graph shown in FIG. 15, for example. In the graphs shown in FIGS. 13 to 15, the passage of time indicated by the symbols of times Ta, Tb, Tc, Td, and Te is shown by the flow from one end side to the other end side in the horizontal axis direction, and the evaluation value or The vertical axis indicates the comprehensive evaluation value r.

As shown in FIG. 15, the comprehensive evaluation value r is not constant and changes with the passage of time. When such a comprehensive evaluation value r is expressed on a two-dimensional graph with the evaluation values shown in FIG. 13 as the horizontal axis and the evaluation values shown in FIG. 14 as the vertical axis, the distribution r11 shown in FIG. 16 is obtained. A distribution r11 is a set of comprehensive evaluation values r at multiple time points. The distribution r11 has a substantially stable value within a range close to 0 in the vertical axis direction, but shows that the values vary from less than 20 to more than 80 in the horizontal axis direction.

As shown with reference to FIGS. 13 to 16, when the evaluation value _x1 and the evaluation value _x2 are two reference values, the overall evaluation value r can be expressed as a distribution (for example, distribution r11) on a two-dimensional graph with one of the evaluation values _x1 and _x2 as the vertical axis and the other as the horizontal axis, as shown in FIG. 13 to 16, the vertical and horizontal axes indicate the evaluation value of the facility 9, and the overall evaluation value r is that of the facility 3 having the facility 9, but the present invention is not limited to this. The vertical axis and the horizontal axis may indicate the evaluation value of the facility 3, and the overall evaluation value r may be that of the plant 2 having the facility 3 in question. In this case, the two facilities 3 have a rating value _x1 and a rating value _x2 , which act as two reference values. The comprehensive evaluation value r is for the plant 2 having the two facilities 3 .

FIG. 17 is a two-dimensional graph showing a distribution r12 of comprehensive evaluation values r different from FIG. The distribution r12 shown in FIG. 17 is derived in the same manner as the distribution r11 in that it is the distribution of the comprehensive evaluation value r in a certain time period. The distribution r12 is different from the distribution r11 in that the comprehensive evaluation value r, which had taken a value near the value r12a at the start of the certain time period, changes to the value r12b as time elapses within the time period. It shows that it rose to the side.

The two-dimensional graphs shown in FIGS. 16 and 17 can be expressed as three-dimensional graphs with the comprehensive evaluation value r as the third axis (height axis). A three-dimensional graph corresponding to the two-dimensional graph shown in FIG. 17 will be described below with reference to FIGS. 18 to 22. FIG.

FIG. 18 is a three-dimensional graph showing a distribution r12 of comprehensive evaluation values r. As shown in FIG. 17, the comprehensive evaluation value r at each point in the distribution r12 tends to increase from the value r12a side to the value r12b side. Therefore, when the distribution r12 is represented by a three-dimensional graph, as shown in FIG. 18, the height relative to the origin (0) in the height axis direction is higher on the value r12b side than on the value r12a side. ing. In this way, in the three-dimensional graph, the increase in the comprehensive evaluation value r can be represented in the height axis direction.

A three-dimensional graph can be checked not only from one viewpoint but also from a plurality of viewpoints. An example in which the distribution r12 shown in FIGS. 17 and 18 is viewed from an angle different from that in FIG. 18 will be described below with reference to FIGS. 19 and 20. FIG.

19 and 20 are three-dimensional graphs showing the distribution r12 of the comprehensive evaluation value r from an angle different from that of FIG. The graph shown in FIG. 19 shows the distribution r12 viewed from the origin side of the vertical and horizontal axes in the three-dimensional graph. The graph shown in FIG. 20 shows the distribution r12 viewed from the opposite side of the origin of the vertical and horizontal axes in the three-dimensional graph.

A three-dimensional graph can also show a part of a set of comprehensive evaluation values r such as the distribution r12. Therefore, the three-dimensional graph can be used to pick out areas that show more important information about the condition of the plant 2 . A three-dimensional graph showing a portion of the distribution r12 on the r12b side where the comprehensive evaluation value r is higher will be described below with reference to FIGS. 21 and 22. FIG.

21 and 22 are diagrams showing an example of a three-dimensional graph showing a part of the distribution r12 of the comprehensive evaluation value r on the value r12b side where the value of the comprehensive evaluation value r is higher. FIGS. 21 and 22 show an excerpt from the distribution r12 in the three-dimensional graph described with reference to FIGS. there is

As shown in FIGS. 21 and 22, in a three-dimensional graph, the relationship between a set of comprehensive evaluation values r such as the distribution r12 and the threshold value of the comprehensive evaluation value r is expressed as a positional relationship between objects in a three-dimensional space. can. 21 and 22 illustrate threshold values Th1, Th2, Th3, Th4, Th5, Th6, and Th7. The comprehensive evaluation value r indicated by the threshold Th1 is 60. That is, if the height in the height axis direction is higher than the threshold Th1, it indicates that the comprehensive evaluation value r exceeds 60. The comprehensive evaluation value r indicated by the threshold Th2 is 70. The comprehensive evaluation value r indicated by the threshold Th3 is 80. The comprehensive evaluation value r indicated by the threshold Th4 is 90. The comprehensive evaluation value r indicated by the threshold Th5 is 100. The comprehensive evaluation value r indicated by the threshold Th6 is 110.

In FIGS. 21 and 22, each of the thresholds Th1, Th2, Th3, Th4, Th5, Th6, and Th7 is drawn as an arc belt-shaped object having width in the radial direction. The comprehensive evaluation value r indicated by the threshold Th7 is 120. The specific values of the thresholds Th1, Th2, Th3, Th4, Th5, Th6, and Th7 as well as the shape, size, and position of the object are merely examples and are not limited to these, and can be changed as appropriate.

Furthermore, on the three-dimensional graph, for some or all of the thresholds Th1, Th2, Th3, Th4, Th5, Th6, and Th7, the total evaluation values at multiple points of time included in the set of total evaluation values r such as the distribution r12 Objects showing the overall evaluation value r at the time when r becomes equal to or exceeds the threshold for the first time can be specially drawn. In addition, when there are a plurality of objects having the comprehensive evaluation value r that are specially drawn on the three-dimensional graph, it is possible to further draw the chronological order of the plurality of objects.

In FIG. 22, an object Ob1 represents an object indicating the first comprehensive evaluation value r at a time equal to or greater than the threshold Th3 among the comprehensive evaluation values r at a plurality of time points included in the distribution r12. Also, an object indicating the first comprehensive evaluation value r at a time equal to or greater than the threshold Th4 among the comprehensive evaluation values r at a plurality of time points included in the distribution r12 is shown as an object Ob2. Also, an object indicating the first comprehensive evaluation value r at a time equal to or greater than the threshold Th5 among the comprehensive evaluation values r at a plurality of time points included in the distribution r12 is shown as an object Ob3. The temporal anteroposterior relationship of the objects Ob1, Ob2, and Ob3 is indicated by an arrow-shaped object Ar. Object Ar indicates that the tip side of the arrow is later in time. That is, FIG. 12 shows that the comprehensive evaluation value r has changed from the object Ob1 side to the object Ob3 side via the object Ob2.

The evaluation unit 13 causes the display output unit 34a of the notification unit 14 to display the two-dimensional graph, the three-dimensional graph, and various objects described with reference to FIGS. 16 to 22 . Further, the notification unit 22 can also perform display output similar to that of the display output unit 34a.

In the embodiment, it is arbitrary as to what operation algorithm is used to realize the various display forms described with reference to FIGS. 16 to 22, but an example is shown below. For example, when the comprehensive evaluation value r exceeds a certain threshold (for example, one of thresholds Th3, Th4, and Th5), the evaluation unit 13 causes the notification unit 14 to make a special notification. The special notification may be, for example, a voice notification, a drawing such as blinking of part or all of the screen described with reference to FIG. 11, or both. After such a special notification is made, when an operation (for example, a click operation, etc.) is performed by the confirmer on the plant stability display unit M1 or the trend graph display unit M2, the evaluation unit 13 The two-dimensional graph described with reference to FIG. 17 or the three-dimensional graph described with reference to FIG. 18 is displayed on the display output unit 34a. In the display by the display output unit 34a, a set (for example, distribution r12) of comprehensive evaluation values r exceeding the certain threshold is displayed. In addition, when the two-dimensional graph described with reference to FIG. 17 is displayed by the display output unit 34a, when an operation (for example, a click operation, etc.) is performed by the confirmer on the two-dimensional graph Alternatively, the evaluation unit 13 may perform a multi-stage operation such that the three-dimensional graph described with reference to FIG. 18 is displayed on the display output unit 34a.

In the state where the three-dimensional graph described with reference to FIG. As shown, there are objects (for example, objects Ob1, Ob2, and Ob3) that indicate the comprehensive evaluation value r at the first point in time when it exceeds the threshold or exceeds the threshold, and a drawing that indicates the temporal context of these objects (for example, , object Ar) is displayed on the display output unit 34a.

The relationship between the operation by the confirmation person and the response of the evaluation unit 13 to the operation illustrated above is merely an example, and is not limited to this, and can be changed as appropriate. For example, when the plant stability display section M1 or the trend graph display section M2 is operated, the display output section 34a may perform drawing as described with reference to FIG. Further, in the state where the three-dimensional graph described with reference to FIG. 18 is displayed, the three-dimensional graph can be displayed at another angle as described with reference to FIGS. It may be provided so that it can be switched to drawing when viewed from above. Similarly, when the three-dimensional graph described with reference to FIG. 22 is displayed, the three-dimensional graph can be viewed from another angle as described with reference to FIG. It may be provided so that it can be switched to drawing in the case of Further, in the embodiment, objects (for example, objects Ob1, Ob2, and Ob3) indicating the comprehensive evaluation value r at the first point in time when the threshold is exceeded or the threshold is exceeded, and a drawing indicating the temporal context of these objects Although (for example, object Ar) is limited to FIG. 22, similar rendering may be performed in each of the two-dimensional graph and three-dimensional graph described with reference to FIGS. 17 to 21. FIG.

Also, the estimation of the cause of the abnormality described with reference to FIG. 10 may be further combined with data analysis such as clustering. An example of data analysis will be described below with reference to FIG.

FIG. 23 is a schematic diagram showing an example of the relationship between the data analysis by the k-means method and the comprehensive evaluation value r. In the description with reference to FIG. 23, the evaluation value _x1 on the horizontal axis is the vibration evaluation value of the generator, and the evaluation value _x2 on the vertical axis is the temperature evaluation value of the generator. , but not limited to this.

The k-means method is a method of non-hierarchical clustering, in which a predetermined number of clusters are assigned to an existing data group, an optimization problem is solved by computation, and the existing data group is divided into the predetermined number of partial data. This is a method of classifying cases into (case classes). Since the algorithm for the optimization problem is the same as a known one, a detailed explanation is omitted.

FIG. 23 exemplifies case class Ga, case class Gb, and case class Gc as classification results. The case class Ga includes comprehensive evaluation values ra, rb, rc, rd, and re. The case class Gb includes comprehensive evaluation values rf, rg, and rh. The case class Gc includes comprehensive evaluation values ri, rj, rk, and rl. The comprehensive evaluation values ra, rb, rc, rd, re, rf, rg, rh, ri, rj, rk, and rl are calculated based on the evaluation values x ₁ and x ₂ when an abnormality occurred in the plant 2 in the past. is the overall evaluation value r. That is, the existing data group in this case is the comprehensive evaluation value r (for example, the fourth evaluation value) calculated when an abnormality occurred in the plant 2 in the past.

A case class Ga is a case class when a failure occurs in the bearing of the generator. In case class Ga, the center of the cluster is set so as to include comprehensive evaluation value r calculated when abnormality occurs in the temperature of the generator indicated by evaluation value _x2 among the existing data groups. A case class Gb is a case class when a failure occurs in the water turbine. In the example class Gb, the center of the cluster is set so as to include the total evaluation value r calculated when abnormality occurs in the vibration of the generator indicated by the evaluation value _x1 among the existing data groups. A case class Gc is a case class in which both the bearing and the water turbine have failed. In the case class Gc, the center of the cluster is set so as to include the comprehensive evaluation value r calculated when both the vibration and the temperature of the generator are abnormal among the existing data groups.

In the k-means method, when a predetermined number of clusters are allocated for classification and the initial positions of the clusters are set at random, the classification tends to be highly dependent on the random value. For this reason, the initial position of the cluster should be manually set so that the correlation between the specific abnormality that actually occurred in the plant 2 and the comprehensive evaluation value r when the abnormality occurred is stronger. , more accurate classification can be performed. Similarly, by setting the number of clusters (predetermined number) to correspond to the types of failures to be classified, more accurate classification can be performed.

By combining such data analysis with the multi-step calculation of the overall evaluation value r of the plant 2 described above, the accuracy of estimating the specific content of the abnormality in the event that some abnormality occurs in the plant 2 can be further improved. . For example, if the latest comprehensive evaluation value r is included in one of the case classes Ga, Gb, and Gc, it is more strongly suggested that an abnormality corresponding to that case class has occurred.

Further, as shown in FIG. 23, even if the latest comprehensive evaluation value r is not included in any of the case classes Ga, Gb, and Gc, the latest comprehensive evaluation value r and the case classes Ga, Gb, and Gc Based on the relationship with each of , it is possible to derive the probability of each case of estimation of what kind of abnormality has occurred. For example, FIG. 23 shows boundary lines La, Lb, and Lf that classify existing data groups in a larger framework than case classes Ga, Gb, and Gc. A boundary line La is a boundary line between a comprehensive evaluation value r indicating a tendency of failure in the bearings of the generator and a comprehensive evaluation value r indicating a tendency of failure in the water turbine. A boundary line Lb is a boundary line between a comprehensive evaluation value r indicating a tendency of failure in the bearing of the generator and a comprehensive evaluation value r indicating a tendency of failure in the bearing and the water turbine. A boundary line Lf is a boundary line between a comprehensive evaluation value r indicating a tendency of failure of the water turbine and a comprehensive evaluation value r indicating a tendency of failure of the bearing of the generator and the water turbine. Since the comprehensive evaluation value r shown in FIG. 23 is included in the range surrounded by the boundary lines La and Lb, the possibility that the bearing of the generator has failed is more likely than other failures. It is suggested relatively strongly.

The boundary lines La, Lb, and Lf may be derived from the positional relationship between the centers of the case classes Ga, Gb, and Gc, or by other methods. Other methods may be non-hierarchical clustering other than the k-means method, hierarchical clustering such as Ward's method, or other clustering methods. Also, the derivation of the case classes Ga, Gb, and Gc is not limited to the k-means method, and other methods may be used.

Furthermore, a so-called machine-learning algorithm may be incorporated into the data analysis for performing classification again by adding the comprehensive evaluation value r, which is newly calculated according to the passage of time, to the data analysis target.

When data analysis as described with reference to FIG. 23 is applied to the embodiment, the storage unit 31 further stores software programs and data for such data analysis. The computing unit 32 reads and executes the software program and data, and further analyzes the data.

As described above, according to the embodiment, the evaluation device 10 is an evaluation device that evaluates the operation status of the plant 2 including the plurality of facilities 3 . The evaluation device 10 has a display unit ( _display output unit 34a, notification unit ₂₂ ) and an evaluation value (eg, a comprehensive evaluation value r) and causes the display unit to display the relationship between the two reference values and the evaluation value in a three-dimensional graph. The evaluation unit 13 causes the display unit to display a relationship (for example, distributions r11 and r12) including information indicating time-series changes in the evaluation values associated with time-series changes in the two reference values. The two reference values are values related to the operating status of two facilities 3 or two facilities 9 included in one facility 3 . As a result, a value indicating the state of a part of the facility 3 or the device 4 (e.g., evaluation value x ₁ ), a value indicating the state of the entire facility 3 or the device 4 including that portion (e.g., comprehensive evaluation value r), and The relationship between the state of other parts of the whole (for example, the evaluation value x ₂ ) and the relationship can be grasped on the three-dimensional graph. Therefore, it becomes easier to grasp the relationship between the state of a part of the facility 3 or the equipment 4, the state of the entire facility 3 or the equipment 4 including the part, and the state of other parts of the whole.

Also, the information indicating the chronological change in the evaluation value (for example, comprehensive evaluation value r) includes information indicating at least two points in time (for example, objects Ob1, Ob2, etc.). This makes it easier to grasp the relationship between the state of a part of the facility 3 or the equipment 4 at multiple points in time, the state of the entire facility 3 or the equipment 4 including the part, and the state of other parts of the whole.

The information indicating the chronological change in the evaluation value (for example, the comprehensive evaluation value r) includes information indicating the direction of change in the evaluation value between at least two points in time (for example, the object Ar). This makes it easier to grasp the temporal context of the state of the facility 3 or the equipment 4 at multiple points in time. Therefore, it becomes easier to grasp changes in the state of the facility 3 or the equipment 4 over time.

In addition, the evaluation unit 13 further provides information indicating the correspondence relationship between the evaluation value (for example, comprehensive evaluation value r) and the threshold value of the evaluation value (for example, threshold values Th1, Th2, Th3, Th4, Th5, Th6, and Th7). It is displayed on the display unit (display output unit 34a, notification unit 22). This makes it easier to manage the state of the facility 3 or the equipment 4 with the threshold as a reference or guideline.

The evaluation device 10 also includes an acquisition unit 12 that acquires sensing values individually output by a plurality of sensors 4 provided in a plurality of facilities 9 included in each facility 3 . The evaluation unit 13 performs a first process, a second process, a third process, and a fourth process. As a result, the person who confirms the overall evaluation value r such as the fourth evaluation value can grasp the overall stability of the plant 2 more than in the case of simply acquiring the sensing values and notifying them in an enumerated manner. easier. Also, a first value u that is the weighting value (p) in the second process, a second value v that is the weighting value (p) in the third process, and a third value that is the weighting value (p) in the fourth process w and are different. As a result, the weighting of the overall evaluation value r for each sensing item in the stability evaluation of the plant 2 and the weighting of the overall evaluation value r for each facility 3 are made different from the weighting of the overall evaluation value r of the plant 2. be able to. For this reason, for those that directly indicate the evaluation of the entire plant 2, such as the fourth evaluation value, for example, by setting the weighting value to 1 or a value closer to 1, the synergistic effect of the evaluation values referred to in the calculation appears in the comprehensive evaluation value r. Therefore, in the evaluation of the entire plant 2 using the fourth evaluation value, the presence or absence of multiple factors due to the conditions of the plurality of facilities 3 can be expressed. On the other hand, for values such as the second evaluation value, for which there is less need to consider multiple factors from the viewpoint of the strength of the relationship with the evaluation of the entire plant 2, the weighting value is set to a larger value, for example. As a result, the "presence or absence of abnormality" and the "degree of abnormality" indicated by the evaluation values referred to in the calculation can be represented by the comprehensive evaluation value r. As for the third evaluation value, by setting an intermediate weighting value between the weighting value in the second process and the weighting value in the fourth process, the relationship with the evaluation of the entire plant 2 is stronger than the second evaluation value. , the third evaluation value can be calculated as a value having a weaker relationship with the evaluation of the entire plant 2 than the fourth evaluation value. In this way, in deriving the index for grasping the overall stability of the plant 2, weighting according to the strength of the relationship with the evaluation of the plant 2 as a whole can be arbitrarily performed. Therefore, it is possible to further assist the confirmation person in grasping the status of the plant 2 .

Also, the first value u is greater than the second value v and the third value w. Also, the second value v is greater than the third value w. As a result, the fourth evaluation value that directly indicates the evaluation of the entire plant 2 can be calculated as a value in which the multiple factors of the evaluation value are more likely to appear than the second evaluation value and the third evaluation value. Moreover, the 3rd evaluation value stronger than the 2nd evaluation value with the evaluation of the whole plant 2 can be calculated as a value in which the compound factor of an evaluation value appears more easily than a 2nd evaluation value.

In addition, the evaluation unit 13 performs the second processing for each of the plurality of sensing items for some or all of the plurality of facilities 9 (for example, generators and water turbines), and the second processing is performed for each of the plurality of sensing items. In the third process of facility 3, a third evaluation value is calculated as the Lp norm of the second evaluation values of the plurality of sensing items. In addition, the evaluation unit 13 performs the second processing for one sensing item on some or all of the plurality of facilities 3 (for example, waterways), and performs the second processing on the third processing of the facility 3 for which the second processing is performed for one sensing item. In the process, the third evaluation value is calculated as the Lp norm of the second evaluation value of the one sensing item. As a result, the processing corresponding to the number of facilities 9 included in the facility 3 can be performed more accurately.

Further, as shown in the example of FIG. 5, the sensor 4 detects the temperature, humidity, pressure, vibration, water level, capacity, flow rate, operating speed, and opening/closing state of the opening/closing part (for example, the inlet valve) of the installed equipment 9. detect at least one of them. This makes it possible to use more diverse information as information relating to the operating status of the plant 2 .

Also, some or all of the plurality of sensors 4 and the acquisition unit 12 are connected via wireless communication. This makes it easier to relax the installation conditions of the evaluation device 10 with respect to the plant 2 . In addition, it becomes easier to relax physical constraints for providing the sensor 4 . That is, since the sensor 4 and the evaluation device 10 do not necessarily have to be directly connected, the degree of freedom regarding the installation of the sensor 4 and the evaluation device 10 is further improved.

In addition, the evaluation value is derived as a percentage from the most stable state including the normal operation of the facility 9 to the state including the abnormal operation of the facility 3, such as the above-described alarm level warning situation. It is a value that indicates the operating status of the facility 3 that has been set. As a result, the operational status of the plant 2 can be grasped with more intuitive numerical information represented within the range from 0 to 100. It should be noted that the "stable state" does not only refer to an ideal operating state without any problems, but as long as the state is maintained, the equipment 9 and the facility 3 including the equipment 9 can continue to operate without problems. Refers to general condition. Hereinafter, when the equipment 9 or the like is described, the equipment 9 and the facility 3 including the equipment 9 are referred to. In addition, the "most alarming state" is set as the management level of the plant 2. For example, an alarm indicating the need for immediate action to eliminate the abnormality, such as an alarm level, is notified by voice or the like. This corresponds to the operation status of the equipment 9, etc., but is not limited to this. The voice notification of the warning is not essential, and it is natural that the operation status of the equipment 9, etc., which is considered to be the "most alarming state", is changed based on a temporary or newly occurring accident or the like.

In addition, the evaluation device 10 of the embodiment includes a notification unit 14 that performs notification related to evaluation of the operation status of the plant 2 . As a result, even a person who does not have the terminal 20 can get an opportunity to recognize the contents of the notification.

The notification unit 14 also includes a display output unit 34a that displays information related to evaluation of the operating status of the plant 2 . As a result, a manager, an operator, or the like of the plant 2 can easily check the information related to the evaluation of the operation status of the plant 2 at any timing by visually recognizing the display output unit 34a.

It should be noted that the above embodiment is merely an example, and can be modified as appropriate without departing from the technical features of the present invention. For example, the evaluation device 10 is not limited to an information processing device, and may be a dedicated device provided to perform at least the functions of the acquisition unit 12 and the evaluation unit 13 among the functions described above.

The path for transmitting the output of the sensor 4 to the evaluation device 10 is not limited to wireless communication between the communication units 5 and 11 . A part or all of the communication line between the communication section 5 and the communication section 11 may include a wired communication circuit, or may include a public communication line network such as the Internet. Also, part or all of the route may be a dedicated direct connection bus.

In addition, in FIG. 3, the order of increase/decrease of the evaluation value is the first order, but the order of increase/decrease of the evaluation value is arbitrary. Also, the relationship between the level of the evaluation value and the level of the evaluation of the evaluation target (the plant 2, the facility 3, and the sensing items of the facility 3) may be reversed from the above description.

Further, what the sensor 4 detects is not limited to the temperature, vibration, water pressure, and water level of the equipment 9 . For example, the sensor 4 may detect at least one of the humidity of the facility 9, the pressure (not limited to water pressure), the capacity, the flow rate, the operating speed of the movable structure, and the opening/closing status of the opening/closing portion. However, other sensing items may also be detected.

1 evaluation system 2 plant 3 facility 4 sensor 9 facility 10 evaluation device 5, 11, 21 communication unit 12 acquisition unit 13 evaluation unit 14, 22 notification unit 20 terminal r comprehensive evaluation values Th1, Th2, Th3, Th4, Th5, Th6, Th7 threshold x ₁ , x ₂ evaluation value

Claims (7)

An evaluation device for evaluating the operational status of a plant including a plurality of facilities,
a display unit;
an acquisition unit that acquires sensing values individually output by a plurality of sensors provided in a plurality of facilities included in each facility;
an evaluation unit that derives an evaluation value based on two reference values and displays the relationship between the two reference values and the evaluation value in a three-dimensional graph on the display unit;
The evaluation unit causes the display unit to display the relationship including information indicating a time-series change in the evaluation value accompanying a time-series change in the two reference values,
The two reference values are values relating to the operating status of the two facilities or two facilities included in one facility,
The three-dimensional graph is
one of the two reference values is one of the three axes;
the other of the two reference values being the other one of the three axes;
The evaluation value is the remaining one of the three axes,
The information indicating the time-series change in the evaluation value includes information indicating the direction of change in the evaluation value between at least two points in time,
The evaluation unit
performing for each facility a first process of calculating a first evaluation value indicating the stability of the facility based on the plurality of sensing values output by the plurality of sensors;
As the Lp norm of the first evaluation values of the plurality of facilities, calculate the second evaluation value of the sensing item for each group of the first evaluation values classified by type of sensing value in the facility including the plurality of facilities. perform the second processing to
performing a third process for each facility to calculate a third evaluation value indicating the stability of the facility as the Lp norm of the second evaluation value corresponding to the sensing item in the facility;
performing a fourth process of calculating a fourth evaluation value indicating the stability of the plant as the Lp norm of the third evaluation value of a plurality of facilities;
A first value that is the weighting value of the Lp norm in the second process, a second value that is the weighting value of the Lp norm in the third process, and a third value that is the weighting value of the Lp norm in the fourth process Unlike
The sensing value in the first process functions as the reference value, the first evaluation value in the first process functions as the evaluation value,
The first evaluation value in the second process functions as the reference value, the second evaluation value in the second process functions as the evaluation value,
The second evaluation value in the third process functions as the reference value, the third evaluation value in the third process functions as the evaluation value,
The third evaluation value in the fourth process functions as the reference value, and the fourth evaluation value in the fourth process functions as the evaluation value,
Evaluation device. 2. The evaluation device according to claim 1, wherein the evaluation unit causes the display unit to further display information indicating a correspondence relationship between the evaluation value and the threshold value of the evaluation value. the first value is greater than the second value and the third value;
The evaluation device according to claim 1 or 2 , wherein the second value is larger than the third value. The evaluation unit
performing the second processing for each of a plurality of sensing items for some or all of the plurality of facilities;
4. The third evaluation value is calculated as the Lp norm of the second evaluation value of the plurality of sensing items in the third processing of the facility where the second processing is performed for each of the plurality of sensing items. The evaluation device according to any one of 1 . The evaluation unit
performing the second processing on one sensing item for some or all of the plurality of facilities;
4. The third evaluation value is calculated as the Lp norm of the second evaluation value of the one sensing item in the third processing of the facility where the second processing has been performed for one sensing item. The evaluation device according to any one of claims 1 to 3 . The evaluation device according to any one of claims 1 to 5, wherein the sensor detects at least one of temperature, vibration, water pressure and water level of installed equipment. An evaluation system comprising: the evaluation device according to any one of claims 1 to 6 ; and a terminal provided with a notification unit that performs notification relating to evaluation of the operation status of the plant.

Images