JP3693644B2 - Equipment operating state acoustic monitoring method and equipment operating state acoustic monitoring apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a facility operation state acoustic monitoring method and facility operation state acoustic monitoring apparatus that collect and monitor facility operation sounds.
[0002]
[Prior art]
Monitoring of the operating state of facilities such as power plants, chemical plants, equipment, etc. is performed by a supervisor who patrols and listens to the operating sound unique to each facility. That is, acoustic diagnosis using human auditory sense is widely used as a method for confirming the operation state of facilities performed during daily patrols such as patrol and inspection.
[0003]
However, acoustic diagnosis using hearing requires that the loudness and timbre of normal operation generated from the equipment be sufficiently memorized, and it must be a skilled monitor who constantly monitors the equipment. It couldn't be done. In addition, it is difficult to convey the sound diagnosis status, feelings and results judged by a skilled monitor to another monitor, and while it is easy to handle as a diagnostic method for detecting the occurrence of an abnormality, it is difficult to pass on the technology. For this reason, in order to solve the problems of the conventional acoustic diagnosis, an attempt of acoustic diagnosis that can be easily handled by a person other than a skilled engineer has started.
[0004]
As an acoustic diagnosis method that does not depend on humans, it replaces the auditor's auditory judgment, records the specific sounds generated by each equipment, performs frequency analysis, and quantitatively determines the size and tone of normal sounds generated from the equipment. In other words, there is a method for determining whether or not an abnormality has occurred in the equipment by comparing with a signal in a state that seems to be a preset abnormal sound.
[0005]
However, in large-scale plants, inspection inspections of equipment are still regularly carried out by supervisors, and in order to perform acoustic diagnosis using frequency analysis during inspection inspections, a microphone and a small recording device are used. It is necessary to collect the sound of the equipment to be monitored by carrying an acoustic diagnostic device. At that time, since the microphone is recorded at a position with a certain distance to the monitored equipment, the acoustic diagnostic device attenuates the driving sound detected as the distance from the monitored equipment increases. On the other hand, noise such as environmental noise increases, so the S / N ratio deteriorates and the operating sound of the monitored equipment is accurately captured. It was difficult to monitor accurately.
[0006]
As a countermeasure against ambient noise, a monitoring method has been proposed in which the peak frequency of abnormal sound of equipment is used as a monitoring frequency, and the external sound at this monitoring frequency is removed and monitored (see, for example, Patent Document 1). Further, there is described an apparatus for detecting abnormality by monitoring the loudness of information collected by a microphone or the like as the level of a sound signal or the tone color as a frequency component (see, for example, Patent Document 2).
[0007]
[Patent Document 1]
JP-A-9-229762 ([0023] to [0028] FIGS. 4 and 5).
[Patent Document 2]
JP-A-11-118593 ([0004], [0005]).
[0008]
[Problems to be solved by the invention]
However, from the supervisor who manages the facilities, there are many other facilities in the vicinity of the facilities to be monitored in the facilities such as various plants and various equipment. A monitoring technique that can accurately determine whether there is an abnormality in the target facility is desired.
[0009]
The present invention has been made to solve such a conventional problem, and can accurately collect the operation sound from the monitoring target equipment, and can be accurately distinguished and monitored from the noise such as the operation sound from other surrounding equipment. It is an object of the present invention to provide a facility operation state monitoring method and facility operation state sound monitoring device.
[0010]
[Means for Solving the Problems]
In order to achieve the above object, the present invention provides a facility operating state sound monitoring method and a facility operating state sound monitoring device having the following configuration.
[0011]
  The facility operating state sound monitoring method of the present invention includes a step of collecting operation sounds specific to the facility to be monitored by a first sound sensor for a predetermined period and outputting a first operation sound signal; Performing a frequency analysis process on one driving sound signal to output a first acoustic spectrum signal indicating the relationship of the frequency to the volume output value; and a peak frequency at which the volume output value exhibits a peak from the first acoustic spectrum signal. And extracting the first volume output value of the peak frequency as normal operation state monitoring information and storing it in a storage device in advance, and at a monitoring position away from the monitoring target facility from the first sound sensor. Providing a second sound sensor, collecting operation sounds of the monitoring target equipment for the predetermined period, and outputting a second operation sound signal; and frequency of the second operation sound signal. Analyzing and outputting a second acoustic spectrum signal indicating the relationship of the frequency to the volume output value; and the second volume output value of the peak frequency read from the storage device as the second acoustic spectrum. A step of extracting from the signal, and obtaining a correlation between the first sound volume output value and the second sound volume output value read from the storage device, and determining an operation state of the monitoring target equipment by the second sound sensor. Outputting monitoring position information to be monitored, storing the monitoring position information in the storage device, and in monitoring, the position of the monitoring position information of the facility to be monitoredThe secondProviding and monitoring a sound sensor; andSecondAnd comparing the monitoring output from the sound sensor with the monitoring information of the normal operation state stored in the storage device to monitor whether there is an abnormality in the monitoring target equipment.
[0012]
According to this invention, since the peak frequency obtained from the driving sound information from the sound sensor provided near the monitoring target facility is used as the monitoring frequency for monitoring by the second sound sensor output signal, the driving sound from the monitoring target facility is monitored. Can be collected accurately, and can be monitored with distinction from driving sounds from other surrounding equipment. Further, since the monitoring position information is output, when the monitoring person who monitors the facility patrols and monitors, by bringing the second sound sensor, the second position is detected in the monitoring target facility. A sound sensor can be provided, and monitoring can be performed under the same conditions even when the monitor is changed, even when the monitor is changed, even when the monitor is brought.
[0013]
Since the installation position of the first sound sensor is in contact with the container wall surface of the monitoring target facility, the operation sound from the monitoring target facility can be collected accurately, and environmental sounds such as driving sounds from other surrounding facilities can be collected. Can be accurately distinguished and monitored.
[0014]
The monitoring position information is obtained by performing at least one of a deviation process and a coherence analysis process of the first sound volume output value and the second sound volume output value at the peak frequency from the first sound spectrum signal and the second sound spectrum signal. Since it is the obtained information, it is possible to automatically confirm the installation position of the second sound sensor each time monitoring is performed.
[0015]
The monitoring position information is obtained when the information obtained by at least one of the deviation process and the coherence analysis process of the first sound volume output value and the second sound volume output value falls within a predetermined range. Information on the distance from the second sound sensor to the second sound sensor, the monitoring position information can be read from the storage device each time monitoring is performed, and the second sound sensor can be installed at the read monitoring position. Can be monitored under the same conditions.
[0016]
The monitoring of the facility to be monitored by the second sound sensor detects the volume output value of the peak frequency obtained from the monitor output by the second sound sensor and the normal operation state stored in the storage device. Compared with monitoring information, when different signals are output, abnormal information is output, displayed on the display device, and stored in the storage device. Therefore, even in the second sound sensor located away from the monitoring target facility Since the determination is made from the volume output value based on the peak frequency extracted from the first sound sensor closest to the monitored facility, the operation sound of the monitored facility can be stably monitored without being affected by noise.
[0017]
The extraction of the peak frequency is to output a frequency at which the volume output value shows a peak from a frequency curve diagram with respect to the volume output value obtained by performing frequency analysis processing on the first sound sensor output signal. Therefore, the operation sound of the monitored equipment is the peak frequency directly or from a nearby location, the operation sound from the monitored equipment can be collected accurately, and it can be accurately distinguished from the environmental sound such as the operation sound from other surrounding equipment. Can be monitored.
[0018]
The monitoring position information is output from the frequency characteristic curve diagram information for each volume signal generated by performing frequency analysis processing on each of the first and second sound sensor output signals, for each volume signal value at the peak frequency. Since the coherence analysis process outputs the monitoring position information of the second sound sensor, the second sound sensor is always installed at the monitoring position to monitor the operating state of the monitored equipment under the same conditions. can do.
[0019]
Since the monitoring position information is distance information from the monitoring target facility to the second sound sensor, by storing this distance information, monitoring is performed in a state that is always close to the same state even if the monitor changes. The operating state of the target equipment can be monitored.
[0020]
Since the monitoring information of the normal operation state and the monitoring position information of the second sound sensor are displayed on a display device, the monitoring position information is read from the storage device and displayed on the display device. The second sound sensor can be installed at a predetermined position at the position. Therefore, regardless of who the monitoring person is, the second sound sensor can monitor the operation state of the monitoring object facility under the same conditions as when the first sound sensor is brought into contact with the monitoring object facility. it can.
[0021]
The monitoring of the facility to be monitored by the second sound sensor detects the volume output value of the peak frequency obtained from the monitor output by the second sound sensor and the normal operation state stored in the storage device. Compared with monitoring information, when different signals are output, abnormal information is output, displayed on the display device and stored in the storage device, so that the first sound sensor is brought into contact with the monitoring target equipment and Similarly, the occurrence of abnormal operating conditions can be accurately monitored by the second sound sensor.
[0022]
The equipment operation state acoustic monitoring apparatus of the present invention collects operation sounds specific to the object to be monitored whose operation state is monitored from operation sounds using a first sound sensor, and the monitoring object is obtained from the first sound sensor. The first acoustic spectrum showing the relationship between the frequency and the volume output value by performing frequency analysis processing on the first volume signal and the second volume signal collected by the second sound sensor provided at the monitoring position away from the facility And the second acoustic spectrum information, the peak frequency information in the first acoustic spectrum, and the first sound volume output value at the peak frequency in the first acoustic spectrum are extracted and information indicating the normal operation state is stored. A second volume output value of the peak frequency is extracted from the storage device and the second acoustic spectrum information, and the first volume output value and the second volume output value are compared. If there is a correlation, means for storing the position of the second sound sensor as the monitoring position of the monitoring target equipment in the storage device, and reading the monitoring position information stored in the storage device, The operation sound of the monitoring target facility is monitored by the second sound sensor provided, and the monitoring volume signal monitored by the second sound sensor and the data indicating the normal operation state stored in the storage device And means for determining the presence / absence of an abnormality compared to the above.
[0023]
According to the present invention, it is possible to accurately collect the operation sound from the monitoring target facility even if there are other facilities in operation in the vicinity, and to accurately monitor the operation sound from the other surrounding facilities.
[0024]
Since the peak frequency in the first volume signal is a frequency at which the volume output value shows a peak in a curve diagram showing the relationship of the frequency to the volume output value, the signal has a larger S / N ratio than ambient noise. Thus, the installation of the second sound sensor at the monitoring position and the monitoring of the monitoring target equipment can be performed accurately.
[0025]
DETAILED DESCRIPTION OF THE INVENTION
Next, an embodiment of the facility operating state monitoring method according to the present invention will be specifically described with reference to FIGS. First, in order to understand the whole operation state monitoring method of equipment, it demonstrates with reference to the flowchart of FIG.
[0026]
The feature of the operation state monitoring method of this equipment is that a first sound sensor is provided in contact with or close to the equipment to be monitored in order to detect the operation sound unique to the equipment to be monitored without being relatively influenced by ambient noise. To obtain second monitoring data for the sound sensor. Further, the feature of the method for monitoring the operating state of the facility is that the second sensor is located at a monitoring position farther from the facility to be monitored than the first sound sensor under substantially the same conditions as the state monitored by the first sound sensor. The sound sensor 4 is provided so that it can be monitored. Furthermore, the feature of the operation state monitoring method of the facility is that the peak frequency is obtained from the operation sound signal of the first sound sensor output in order to output substantially the same condition as the state monitored by the first sound sensor, This peak frequency is defined as a driving sound signal, and a monitoring position of the second sound sensor used for monitoring and a determination signal for monitoring the driving sound of the monitoring target equipment are output.
[0027]
The facility operating state sound monitoring method 1 as shown in FIG. 1 is the one shown in FIG. 1, in which the first sound sensor 2 is installed directly or close to the monitoring target facility 3 and the operation sound specific to the monitoring target facility 3 is collected. Monitoring data is collected so that monitoring can be performed by the second sound sensor 4 provided at a monitoring position away from the monitoring target facility 3 from the first sound sensor under substantially the same conditions as the state. The monitoring data collecting step 5 and the monitoring step 6 for monitoring the operation state of the monitoring target equipment 3 based on the collected monitoring data.
[0028]
The monitoring data collection process 5 collects the operation sound from the monitoring target facility 3 by the first sound sensor 2 provided relatively near the monitoring target facility 3 and the second sound sensor 4 provided far away. A driving sound collecting step 11 for outputting a driving sound signal, a peak frequency extracting step 12 for extracting a peak frequency indicating the peak of the first volume output value from the output signal of the first sound sensor 2, and a second The second volume output value of the peak frequency is extracted from the output signal of the sound sensor 4, and the operation state of the monitoring target equipment 3 by the second sound sensor 4 is monitored from the first and second volume output values. The monitoring position information output step 13 outputs the monitoring position information as monitoring data.
[0029]
The monitoring step 6 sets the second sound sensor 4 to the position of the monitoring position information by a method of monitoring the operation state of the monitoring target equipment 3 only by the second sound sensor 4 based on the monitoring data. The monitoring start step 15 and the monitoring step 16 for monitoring the operation state of the monitoring target equipment 3 by monitoring the volume output value of the peak frequency from the output signal of the second sound sensor 4.
[0030]
The monitoring target facility 3 is a facility such as a power plant, a chemical plant, or a device, for example, and is an operating device whose operating state is monitored. The first sound sensor 2 is an element that is provided directly or in the vicinity of a monitoring target part of the monitoring target facility 3 and converts driving sound into an electrical signal, and is, for example, an acceleration sensor, a piezoelectric sensor, a speaker, or the like. The vicinity of the monitoring target facility 3 is a position that is in direct contact with or directly away from a portion of the monitoring target facility 3 where a failure is particularly expected, and noise such as operation noise from other surrounding facilities. The operation sound of the monitoring target equipment 3 can be detected with a larger S / N ratio.
[0031]
The second sound sensor 4 is a sensor that is used for daily monitoring, and is an element that converts the operation sound of the monitoring target equipment 3 into an electrical signal, for example, a microphone. It is provided at a position close to the monitoring target facility 3 in a safety passage or a safety zone. The second sound sensor 4 is provided at a position farther from the monitoring target facility 3 than the first sound sensor 2, and the operation sound from the surrounding facilities and the work sound of the construction are noise. On the contrary, the detected sound volume of the driving sound from the monitoring target equipment 3 is output as a sound volume attenuated by a distant distance. Although the second sound sensor 4 in the monitoring data collection process 5 and the second sound sensor 4 in the monitoring process 6 are described with the same reference numerals, they may be the same sound sensor or other sound sensor. Alternatively, the second sound sensor 4 in the monitoring data collection step 5 is for outputting the monitoring position condition of the second sound sensor 4 in the monitoring step 6.
[0032]
The operation sound collection step 11 outputs the operation sound of the monitoring target equipment 3 as a signal larger than the environmental sound such as operation sound and work sound from other surrounding equipment. Therefore, the first sound sensor 2 and the second sound sensor 4 have, for example, a hood for providing directivity as means for outputting the operation sound of the monitoring target facility 3 with a good S / N ratio. It is desirable to be provided.
[0033]
The peak frequency extraction step 12 converts the output signal of the first sound sensor 2 into a digital signal, and then performs a frequency analysis process to show the first acoustic spectrum of the characteristic curve diagram showing the relationship between the volume output value and the frequency. And a peak frequency at which the first sound volume output value peaks is extracted from the first acoustic spectrum. The peak frequency of the first acoustic spectrum is a frequency obtained from the driving sound signal collected from the position closest to the monitoring target facility 3, and is a characteristic sound of electrical information that represents the characteristics of the driving sound of the monitoring target facility 3. is there.
[0034]
The monitoring position information output step 13 outputs the monitoring position information of the second sound sensor 4 so as to be substantially monitored by the first sound sensor 2 and stores it in the storage device. This is a process that enables monitoring of the operation sound of the monitoring target equipment 3 by the sound sensor 4. In the monitoring start step 15, only the second sound sensor 4 is installed at the monitoring position from the monitoring position information stored in the storage device or the monitoring position obtained based on the monitoring position information output step 13. The monitoring step 16 extracts the volume output value of the peak frequency from the operation sound signal detected by the second sound sensor 4, and monitors the operation state of the monitoring target equipment 3 by monitoring the extracted volume output value. To do.
[0035]
That is, the facility operating state sound monitoring method 1 is monitored from the first sound sensor 2 that is in direct contact with the monitored facility 3 as basic data in order to prevent erroneous monitoring due to noise such as environmental sounds from the surroundings. The characteristic curve diagram showing the relationship between the frequency for the volume output value detected by the second sound sensor 4 that outputs the characteristic of the operation sound of the target equipment 3 at the peak frequency and is provided relatively far away is the S / N ratio. Although worse, by monitoring the volume output value corresponding to the peak frequency from this characteristic curve diagram, it is possible to monitor the operating state of the monitored facility 3 that is resistant to noise. The monitoring of the operation state of the monitoring target equipment 3 which is resistant to noise means that it can be detected without overlooking the abnormal operation of the monitoring target equipment 3.
[0036]
As the means for collecting the operation sound unique to the monitoring target facility, when the first sound sensor is brought into contact with the monitoring target facility and the housing of the monitoring target facility, the driving sound is detected by the second sound sensor at the monitoring position. If the peak frequency during normal operation is different when the sound is collected, it is preferable that the vicinity of the monitoring target sound generator of the monitoring target facility is not touched.
[0037]
Next, the specific embodiment of FIG. 1 will be described with reference to FIGS. The same parts as those in FIG. 1 will be described using the same reference numerals. FIG. 2 is a circuit configuration diagram for explaining the monitoring data collection device 20 for performing the monitoring data collection step 5. In this embodiment, the monitoring data collection device 20 for outputting the monitoring position information of the second sound sensor 4 used for daily monitoring for monitoring the operation state of the monitoring target equipment 3, and the monitoring data It comprises an equipment operation state acoustic monitoring device 21 for monitoring the operation state of the monitoring target equipment 3 based on the monitoring data obtained from the data collection device 20. An embodiment of the monitoring data collection device 20 is shown in FIG. 2, and an embodiment of the facility operating state acoustic monitoring device 21 is shown in FIG.
[0038]
First, an embodiment of the monitoring data collection device 20 will be described. The operation state of the monitoring target equipment 3 is monitored by operation sound, and this operation sound is detected by the first sound sensor 2 and the second sound sensor 4. The first sound sensor 2 is set in the vicinity of the monitoring target part in order to collect the operation sound unique to the monitoring target facility 3. For example, the acceleration sensor as the contact-type acoustic input device 25 has a wall surface in the vicinity of the monitoring target part. It is provided in contact with. The second sound sensor 4 is provided in a monitoring position, for example, a safety zone, away from the first sound sensor 2 from the monitoring target portion of the monitoring target facility 3, and a microphone is used as the non-contact type sound input device 26. Provided. The operation sound collection step 11 is performed by the first sound sensor 2 and the second sound sensor 4. The collection period of the operation sound of the monitoring target facility 3 by the first sound sensor 2 and the second sound sensor 4 is preferably a predetermined period.
[0039]
An A / D conversion processing unit 34 for converting an analog signal into a digital signal is connected to the output terminal of the contact-type acoustic input device 25, and the output terminal of the A / D conversion processing unit 34 is converted into a digital signal. An acoustic spectrum signal indicating the relationship of the frequency to the sound volume output value from the driving sound signal is connected to a frequency analysis processing unit 35 for creating a frequency characteristic curve diagram for the sound pressure level as shown in FIG. The output terminal of the frequency analysis processing unit 35 has a positive peak frequency extraction unit 36 for extracting the peak frequency of the waveform indicating the peak characteristic representing the characteristics of the operation sound of the monitored facility 3 in the frequency characteristic curve diagram. It is connected. A coherence analysis processing unit 37 for outputting the monitoring position of the non-contact acoustic input device 26 is connected to the output terminal of the positive peak frequency extraction unit 36.
[0040]
On the other hand, an A / D conversion processing unit 41 for converting an analog signal into a digital signal is connected to an output terminal of the non-contact acoustic input device 26, and a digital signal is connected to an output terminal of the A / D conversion processing unit 41. Connected to a frequency analysis processing unit 42 for creating a frequency characteristic curve diagram with respect to a sound pressure level as shown in FIG. Yes. The sound pressure level value in the frequency characteristic curve diagram of FIG. 4 is attenuated by a distance corresponding to a distance far from the monitored facility 3 compared to the peak value of the sound pressure level in the frequency characteristic curve diagram of FIG. The level is detected.
[0041]
The output terminal of the frequency analysis processing unit 42 is connected to a pseudo peak frequency extraction unit 43 for outputting the sound pressure level corresponding to the peak frequency as the sound pressure level of the pseudo peak frequency in the frequency characteristic curve diagram. . A peak frequency deviation processing unit 44 for outputting a difference between the sound pressure level values of the peak frequency and the pseudo peak frequency is connected to an output terminal of the pseudo peak frequency extracting unit 43. A measurement position pass / fail judgment processing unit 45 for outputting measurement position information of the non-contact acoustic input device 26 from outputs of the coherence analysis processing unit 37 and the peak frequency deviation processing unit 44 is connected.
[0042]
The equipment operating state acoustic monitoring device 21 is provided with a distance measuring device 47 such as a laser distance measuring device in order to measure the distance between the monitoring target equipment 3 and the non-contact type acoustic input device 26. The measured distance information indicates a position for installing the non-contact acoustic input device 26 when monitoring the monitoring target equipment 3. An output terminal of the distance measuring device 47 is connected to an A / D conversion processing unit 48 for converting the measured distance information into a digital signal. The distance information output from the A / D conversion processing unit 48 indicates whether the measurement position is good or bad. It is connected so as to be output to the determination processing unit 45. The output of each processing unit and extraction unit is connected to a storage device for storing, for example, a data storage unit 49 and a display device 50.
[0043]
Furthermore, on the equipment operating state acoustic monitoring device 21, information stored in the data storage unit 49, such as the name of the monitoring target equipment 3, the installation location (address), the monitoring date / time, the name of the supervisor, and the like are displayed on the display device 50. An input device 51 is provided for inputting content selection information and the like.
[0044]
Block diagrams other than the monitoring target facility 3, the contact-type acoustic input device 25, the non-contact-type acoustic input device 26, the distance measuring device 47, the input device 51, and the display device 50 are surrounded by a dotted line, and the portion surrounded by the dotted line is A computer such as a personal computer 52 can be used. When configured with the personal computer 52, each operation procedure is controlled by a CPU (control device) 54 that operates according to a program stored in advance in the memory 53. In this way, the monitoring data collection device 20 is configured.
[0045]
The contact-type sound input device 25 is a first sound sensor 2 that converts an acoustic signal of a driving sound emitted from the monitoring target facility 3 into an electric signal as a driving sound signal, for example, an acceleration sensor. As a means for detachably attaching to the cover, there is, for example, a sensor integrally provided with a magnet, a rod-shaped jig or the like. The non-contact type sound input device 26 is carried by the second sound sensor 4 that detects the sound signal of the driving sound emitted from the monitoring target equipment 3 as a driving sound signal at a relatively distant position, for example, a patrol position of the monitoring person. For example, a microphone is convenient and easy to operate.
[0046]
In the distance measuring device 47, when the measurement position pass / fail judgment processing unit 45 outputs the monitoring position of the non-contact type acoustic input device 26, the CPU 54 controls the distance measuring device 47 to control the contact type acoustic input device 25 or the monitoring target equipment 3. The distance to the non-contact type sound input device 26 is measured, and the current distance information converted into a digital signal is transmitted to the measurement position pass / fail judgment processing unit 45 that outputs the monitoring position of the non-contact type sound input device 26. The distance measuring device 47 is not limited to the laser distance measuring device, and may be another distance meter or a tape measure. In this case, the measured distance is input from the input device 51 and registered in the data storage unit 49 as distance information. The distance information is information indicating the monitoring position of the non-contact type sound input device 26 and may be any information as long as the non-contact type sound input device 26 can be installed in the next monitoring, for example, the reference position of the monitoring target equipment 3 Information indicating the direction from the distance, distance, and the like.
[0047]
The CPU 54 stores the digitized driving sound signal, the distance information, and the like in the data storage unit 49 in association with the monitoring target facility 3 and controls them to be displayed on the display device 50 when read out when requested. The input device 51 is a terminal through which the monitor inputs input information such as the name of the monitored facility 3, the position (address) where the monitored facility is installed, the date of measurement, the name of the monitor, etc. The CPU 54 stores the input information input from the keyboard in the data storage unit 49 in association with the operation sound information of the monitoring target equipment 3 and simultaneously displays it on the display device 50.
[0048]
The A / D (analog / digital) conversion processing units 34 and 41 digitally convert analog sound signals, which are operation sound information of the monitoring target equipment 3, detected by the contact-type sound input device 25 and the non-contact-type sound input device 26. It is converted into acoustic information in the form.
[0049]
The frequency analysis processing units 35 and 42 perform frequency analysis on the digitized driving sound signal by fast Fourier transform and output an acoustic spectrum signal indicating the relationship of the frequency to the sound volume output value. For example, FIG. 3 and FIG. A characteristic curve diagram showing the relationship between the sound pressure level and the frequency as shown in FIG. The relationship of the frequency with respect to the volume output value is not limited to the graphs as shown in FIGS. 3 and 4, but may be a table.
[0050]
The positive peak frequency extraction unit 36 extracts a peak frequency at which the sound pressure level has a peak, for example, a plurality of peak frequencies, based on the acoustic spectrum of the contact acoustic input device calculated by the frequency analysis processing unit 35. In the frequency characteristic curve diagram with respect to the sound pressure level as shown in FIGS. 3 and 4, the CPU 54 has a waveform (Pmax₁, Pmax₂, Pmax₃... Pmax_n) Is extracted and output to the coherence analysis processing unit 37, the pseudo peak frequency extraction unit 43, and the like, the data is stored in the data storage unit 49 and displayed on the display device 50 at the same time.
[0051]
The positive peak frequency extraction unit 36 can also be configured as shown in FIG. In FIG. 3, the peak frequency extraction means 36a showing the peak of the sound pressure level, in FIG. 6, the frequency extraction means 36b showing the peak of the sound pressure level higher than the average value of the sound pressure level in the entire frequency band, and the specific frequency in FIG. Any one of the frequency extraction means 36c showing a peak of the sound pressure level higher than the average value of the sound pressure levels for each band can be selected according to the environment of the monitoring target equipment 3. For example, the positive peak frequency extraction unit 36 selects the peak frequency extraction unit 36a when the distance between the adjacent monitoring target equipments 3 is long and the non-contact acoustic input device 26 can perform monitoring with a good S / N ratio. To do. The positive peak frequency extraction unit 36 has a sound pressure level higher than the sound pressure level average value for each specific frequency band when the distance between adjacent monitoring target equipments 3 is close and the S / N ratio is poor by the non-contact acoustic input device 26. It is desirable to select the frequency extraction means 36c. In the case of an intermediate S / N ratio, it is desirable that the positive peak frequency extraction unit 36 selects the frequency extraction means 36b that shows a peak of a sound pressure level higher than the sound pressure level average value in the entire frequency band. The positive peak frequency extraction unit 36 may be configured by any one of the frequency extraction units 36a, 36b, and 36c.
[0052]
As a means for extracting the peak frequency based on the acoustic spectrum collected and processed by the contact-type acoustic input device 25, for example, as shown in FIG. 3, a method of extracting an arbitrary number of peak frequencies from the entire analysis frequency band in descending order of frequency level. And calculating a mean value of all analysis frequency bands as shown in FIG. 6 and extracting a frequency having a frequency level larger than the average value as a peak frequency, and for each analysis frequency band as shown in FIG. The peak frequency is extracted by a method of calculating an average value and extracting a frequency having a frequency level higher than the average value as a peak frequency.
[0053]
As a result, the peak frequency is extracted from the acoustic spectrum collected and processed by the contact-type acoustic input device 25. As shown in FIG. 3, an arbitrary number of frequencies showing peaks having a large sound pressure level are extracted from the entire analysis frequency band. By doing so, the characteristic frequency component of the normal operation sound of the monitoring object equipment 3 can be extracted. In addition, the peak frequency can be extracted by calculating an average value of sound pressure levels in all analysis frequency bands as shown in FIG. 6 and extracting a frequency having a sound pressure level larger than the average value as a peak frequency. Furthermore, when calculating the average value of the heights of the respective frequencies in the entire analysis frequency band and setting the frequency higher than the calculated average value as the peak frequency, the range that can be detected by the non-contact acoustic input device 26 is derived. Data as an index can be extracted.
[0054]
The peak frequency extraction means 36a is configured so that the CPU 54 determines the peak frequency Pmax in the descending order of frequency from the entire analysis frequency band in the frequency characteristic curve diagram with respect to the sound pressure level shown in FIG.₁, Pmax₂, Pmax₃... Pmax_nIs output to the coherence analysis processing unit 37 and the pseudo peak frequency extraction unit 43.
[0055]
The frequency extraction means 36b showing the peak of the sound pressure level higher than the sound pressure level average value in the entire frequency band is calculated by the CPU 54 in the frequency characteristic curve diagram for the sound pressure level shown in FIG. The peak frequency Pmax calculated by 1 and higher than the straight line S indicating the calculated average value₁, Pmax₂, Pmax₃... Pmax_nIs output to the coherence analysis processing unit 37 and the pseudo peak frequency extraction unit 43. The sound pressure level frequency extracting means is executed for each specific frequency band.
[0056]
The frequency extraction means 36c showing the peak of the sound pressure level higher than the average value of the sound pressure levels obtained for each specific frequency band is selected by the CPU 54 for each analysis frequency band in the frequency characteristic curve diagram for the sound pressure level shown in FIG. The sound pressure level Pmaxa indicating a peak value higher than the average value S calculated for each analysis frequency band by calculating the average value indicated by the straight line S for each analysis frequency band according to Formula 2.₁, Pmaxb₁, Pmaxc₁... Pmaxn₁Is extracted as a peak frequency and output to the coherence analysis processing unit 37 and the pseudo peak frequency extraction unit 43.
[0057]
[Formula 1]
Formula for calculating the average value of all analysis frequency bands
Analysis frequency band f₁~ F_n
Number of frequency bands ... N
Average value .... fave
Sound pressure level at peak frequency ... pmax₁~ Pmax_n
f₁+ F₂+ ... f_n/ N = fave
fave <pmax₁~ Pmax_n
[0058]
[Formula 2]
Average value calculation formula between arbitrary frequency bands fa
Arbitrary analysis frequency band ... fa₁~ Fa_n
Number of frequency bands ... Na
Average value .... fave
Sound pressure level at peak frequency ... pmaxa₁~ Pmaxa_n
fa₁+ Fa₁+ ... fa_n/ Na = fave₁
fave₁<Pmaxa₁~ Pmaxa_n
Thereafter, the calculation formula between the frequency bands fa to fn is the same as the above formula.
[0059]
The pseudo peak frequency extraction unit 43 calculates the sound pressure level of the same peak frequency from the acoustic spectrum of the non-contact type acoustic input device 26 based on each peak frequency input by the CPU 54 from the positive peak frequency extraction unit 36. Extracted and output to the peak frequency deviation processing unit 44, the coherence analysis processing unit 37, the data storage unit 49, the display device 50, and the like. The acoustic spectrum of the non-contact acoustic input device 26 output is a frequency characteristic curve diagram with respect to the sound pressure level as shown in FIG. 4, for example. The acoustic spectrum of FIG. 4 has a larger sound pressure level at the separation distance from the monitoring target equipment 3 than the acoustic spectrum of FIG. 3, and a large S / N ratio is output.
[0060]
The coherence analysis processing unit 37 obtains coherence for the sound pressure level of each peak frequency extracted by the CPU 54 using the positive peak frequency extraction unit 36 and the pseudo peak frequency extraction unit 43. The coherence analysis unit strengthens the phase correlation. For all frequency bands. The coherence analysis processing unit 37 outputs “1” if there is a correlation and “0” if there is no correlation. Correlated “1” indicates that the sound pressure level value of the peak frequency extracted from the acoustic spectrum of the output of the contact acoustic input device 25 and the sound pressure level value of the peak frequency extracted from the acoustic spectrum of the output of the non-contact acoustic input device 26. When there is a correlation, the measurement position pass / fail determination processing unit 45 determines the current position of the non-contact acoustic input device 26 as the monitoring position. The information on the monitoring position is distance information from the monitoring target facility 3, and is stored in the data storage unit 49 in association with the monitoring target facility name.
[0061]
The peak frequency deviation processing unit 44 obtains a deviation of each sound pressure level in the entire frequency band at the peak frequency extracted by the CPU 54 using the positive peak frequency extracting unit 36 and the pseudo peak frequency extracting unit 43. The peak frequency deviation processing unit 44 outputs “0” if there is no deviation in each sound pressure level, and the measurement position is the same as the operation sound of the monitoring target equipment 3 detected by the contact acoustic input device 25. The determination processing unit 45 determines. Further, if there is a deviation in each sound pressure level, “1” is output, and the measurement position pass / fail determination processing unit 45 determines that the operation sound of the monitored facility 3 detected by the contact-type acoustic input device 25 is not the same.
[0062]
That is, if the coherence analysis processing unit 37 outputs that there is a correlation, and the peak frequency deviation processing unit 44 matches with no deviation in the sound pressure level, the measurement position pass / fail judgment processing unit 45, and the CPU 54 detects the contact-type acoustic input device 25. Is determined to be the same as the driving sound of the monitoring target equipment 3 detected, and is determined as the monitoring position of the driving sound. As shown in FIG. 8, the relationship between these is shown in FIG. 8A where the sound pressure level value of the extracted peak frequency in the frequency characteristic curve diagram with respect to the sound pressure level of the output of the coherence analysis processing unit 37 is shown. The waveform indicates the characteristic sound during normal operation of the monitoring target equipment 3. The waveform shown in FIG. 8 can be output by, for example, processing with a clamp circuit having an average value as a clamp level in the waveform shown in FIG.
[0063]
The sound pressure level value of the peak frequency extracted in the frequency characteristic curve diagram with respect to the sound pressure level of the output of the peak frequency deviation processing unit 44 has a waveform shown in FIG. A waveform in which the sound pressure level value is compensated by adding an attenuation amount (dotted line value) corresponding to a distance difference between the non-contact acoustic input device 26 and the monitoring target equipment 3 to the waveform shown in FIG. FIG. 8C shows the waveform, and the CPU 54 determines that the waveform in FIG. 8A matches the waveform in FIG. 8C. That is, the coincidence of the waveform in FIG. 8A and the waveform in FIG. 8C indicates that the non-contact acoustic input device 26 has been able to detect a characteristic sound during operation of the monitoring target equipment 3. The position of the non-contact type sound input device 26 that can detect the characteristic sound of the monitoring target facility 3 is the monitoring position of the non-contact type sound input device 26 that can measure the operation sound of the monitoring target facility 3 by the non-contact type sound input device 26. It is. When the measurement position pass / fail determination processing unit 45 determines that the waveforms in FIG. 8A and FIG. 8C match, the CPU 54 informs the distance measurement device 47 of the current position of the non-contact acoustic input device 26. And a control command for measuring the distance between the monitoring target equipment 3 and the distance measuring device 47 outputs distance information. This distance information is monitoring position information at which the non-contact acoustic input device 26 can monitor the operation sound of the monitoring target equipment 3. That is, when the coherence analysis processing unit 37 outputs that there is a correlation, and the peak frequency deviation processing unit 44 outputs the coincidence of the sound pressure levels of the waveforms in FIG. 8A and FIG. The unit 45 outputs the monitoring position information of the non-contact acoustic input device 26.
[0064]
The coherence analysis processing unit 37 outputs no correlation, the peak frequency deviation processing unit 44 outputs a sound pressure level deviation, and if they do not match, the measurement position pass / fail judgment processing unit 45 controls the contact type sound input under the control of the CPU 54. It is determined that the sound is different from the driving sound of the monitoring target facility 3 detected by the device 25, and is determined as being unsuitable as the monitoring position of the driving sound and not the monitoring position of the driving sound. These processing results are simultaneously displayed on the display device 50 when the CPU 54 stores them in the data storage unit 49.
[0065]
As an example of obtaining monitoring position information, the coherence analysis processing unit 37 has been described based on the presence / absence of correlation, and the peak frequency deviation processing unit 44 is based on the presence / absence of sound pressure level deviation. It is desirable to obtain a permissible range in advance and determine the monitoring position when entering this range. That is, the monitoring position information is obtained from the monitoring target facility when the information obtained by the deviation or coherence analysis processing of the first sound volume output value and the second sound volume output value falls within a predetermined range. It is desirable to use distance information to the sound sensor.
[0066]
In the above embodiment, the monitoring position information is obtained from the output values of the coherence analysis processing unit 37 and the peak frequency deviation processing unit 44. However, at least one of the monitoring position information can be obtained.
[0067]
The measurement position pass / fail determination processing unit 45 executes coherence determination means 45a and peak frequency deviation determination means 45b as shown in FIG. In the former coherence determination means 45a, the CPU 54 determines the sound pressure level at the peak frequency of the contact acoustic input device 25 from the coherence analysis processing unit 37 and the sound pressure level of the output of the non-contact acoustic input device 26 at this peak frequency. When there is a correlation, “1” is output, and when there is no correlation, a result of “0” is output. Based on this, when the correlation is “1”, the position of the non-contact acoustic input device 26 is “monitored”. When the correlation is “0”, the position of the non-contact acoustic input device 26 is determined as the “monitoring inappropriate position”.
[0068]
In the latter peak frequency deviation determining means 45b, the CPU 54 sends the sound pressure level at the peak frequency of the contact type acoustic input device 25 from the peak frequency deviation processing unit 44 and the sound output from the non-contact type acoustic input device 26 at this peak frequency. The difference between the sound pressure level compensated for the distance from the pressure level is taken, and when there is no difference in the sound pressure level, “0” is output, and the sound pressure level of the contact-type sound input device 25 output and the non-contact-type sound input device 26 output The sound pressure level is the same sound as the operation sound of the monitored equipment and is output. Regarding this determination result, the CPU 54 evaluates that the current position of the non-contact type sound input device 26 is appropriate as a monitoring position, stores it in the data storage unit 49 and simultaneously displays it on the display device 50.
[0069]
Further, when there is a difference in sound pressure level, the CPU 54 outputs “1”, and the sound pressure level output from the contact acoustic input device 25 and the sound pressure level output from the non-contact acoustic input device 26 are abnormal and are monitored. For example, it is determined that the output is not an operation sound of the facility 3 and is regarded as an environmental sound. Regarding this determination result, the CPU 54 evaluates that the current position of the non-contact acoustic input device 26 is inappropriate as a monitoring position, and stores this evaluation in the data storage unit 49 and simultaneously displays it on the display device 50. In this case, the monitor can move the position of the non-contact type sound input device 26, and can select a position at which an evaluation that the monitoring position is appropriate is displayed on the display surface of the display device 50. In this specification, the operation of moving the position of the input device 26 and selecting the monitoring position is defined as tuning.
[0070]
In the distance compensation, the contact-type sound input device 25 and the non-contact-type sound input device 26 have a separation distance difference with respect to the monitored facility 3 that is a sound source, and a sound pressure level difference at each peak frequency is generated accordingly. Therefore, the non-contact type sound input device 26 output waveform is compensated accordingly. This compensation amount is indicated by a dotted line in FIG.
[0071]
In this way, in the measurement position pass / fail determination processing section 45, the CPU 54 outputs “1” with the coherence determination means 45a to determine “monitoring position”, and the peak frequency deviation determination means 45b determines the sound pressure level difference. If the sound pressure level of the contact-type sound input device 25 and the sound pressure level of the non-contact-type sound input device 26 are evaluated to be the same sound and the operation sound of the monitoring target equipment, the monitoring position information is output. Is stored in the data storage unit 49 and displayed on the display device 50 at the same time. When the CPU 54 makes other evaluations, the measurement position pass / fail judgment processing unit 45 outputs an inappropriate monitoring and stores it in the data storage unit 49 and simultaneously displays it on the display device 50.
[0072]
In other evaluations, when the CPU 54 outputs a correlation of “0” by the coherence determination unit 45a and the sound pressure level difference of “1” by the peak frequency deviation determination unit 45b, the correlation similarly shows “0”. For example, when the sound pressure level difference is “0”, the correlation is “1”, and the sound pressure level difference is “1”.
[0073]
In other words, the peak frequency deviation determination means 45b for determining based on the result of the peak frequency deviation processing unit 44 makes the determination according to Equation 4. The coherence determination unit 45a determines the measurement propriety by determining the correlation between the peak frequencies using Equation 3 based on the coherence result calculated by the coherence analysis processing unit 37.
In the peak frequency deviation determination means 45b, the deviation relationship between the peak frequencies is determined by Equation 4 based on the deviation of the peak frequency calculated by the peak frequency deviation processing unit 44, and the monitoring position of the non-contact type acoustic input device 26 is determined. Make a decision. The measurement position pass / fail judgment processing unit 45 outputs the monitoring position of the non-contact acoustic input device 26.
[0074]
[Formula 3]
Coherence function of peak frequency ... cpmax₁~ Cpmax_n
0 <cpmax₁~ Cpmax_n≦ 1
[0075]
[Formula 4]
Sound pressure level deviation at peak frequency deviation between contact type and non-contact type ... Δpmax₁~ Δpmax_n
Sound pressure level at contact peak frequency ... pmax₁~ Pmax_n
0 ≦ Δpmax₁~ Δpmax_n<Pmax₁~ Pmax_n
[0076]
The measurement position pass / fail determination processing unit 45 determines the pass / fail of the measurement position by setting a threshold value for the coherence analysis result by the coherence analysis processing unit 37 or the peak frequency deviation result by the peak frequency deviation. The measurement position pass / fail determination processing unit 45 has been described with respect to an example of AND operation of the coherence determination unit 45a and the peak frequency deviation determination unit 45b, but the S / N ratio is determined by the contact-type acoustic input device 25 and the non-contact-type acoustic input device 26. If good detection is possible, either one of the coherence determination unit 45a or the peak frequency deviation determination unit 45b may be used.
[0077]
The display device 50 includes a digitized acoustic signal from the A / D conversion processing units 34 and 41, an acoustic spectrum output from the frequency analysis processing units 35 and 42, a peak frequency extracted by the positive peak frequency extracting unit 36, and a pseudo signal. The sound pressure level of the peak frequency of the peak frequency extraction unit 43, the coherence result of the coherence analysis processing unit 37, the calculation result of the peak frequency deviation processing unit 44, the measurement pass / fail result of the measurement position pass / fail determination processing unit 45, and the result The distance measured by the distance measuring device 47, the name of the monitoring target facility input from the input device 51, the location of the monitoring target facility, the date of measurement, and the like are displayed.
[0078]
In other words, the display device 50 displays an input information display for displaying information input at the time of monitoring, a threshold setting display, a display of conditions input for retrieving data from the data storage unit 49, and a search result. Data display, measurement result display of the measurement position pass / fail judgment processing unit 45, display of the judgment result of the judgment processing unit 62, and display of the contents of the database stored in the data storage unit 49. The threshold value setting display indicates the items for setting the threshold value in the coherence analysis result or peak frequency deviation result used in the measurement position pass / fail judgment processing unit 45 and the peak frequency comparison result used in normal sound (or abnormal sound) correctness determination. An item for inputting a threshold value is displayed, and an item for inputting a threshold value of a coherence analysis result or a peak frequency comparison result is displayed. The data search condition display displays items for inputting the monitoring target facility name and the monitoring date / time, and the data search result display displays the monitoring target facility name, the monitoring date / time, and the measurable distance.
[0079]
Measurement result display includes time-series display that displays the digital signal converted by the A / D conversion processing unit in time series, time-series progress display that displays a plurality of measurement results superimposed, and processing by the frequency analysis processing unit Frequency analysis display that displays the measured results, frequency analysis progress display that displays multiple measurement results superimposed, and peak frequency display that displays the results processed by the positive peak frequency extraction processing unit and the pseudo peak frequency extraction processing unit And a coherence analysis display for displaying a result processed by the coherence analysis processing unit. The determination result display is a measurement availability display that displays the result of the distance determination processing unit, the normal sound availability display that displays the result of the normal sound correctness determination processing unit, and the database display is the data stored in the data storage unit on the display device. Display.
[0080]
The data storage unit 49 measures the distance from the monitoring target facility 3 to the contact-type sound input device 25 and the non-contact-type sound input device 26, and the monitor inputs the measurement distance from the input device 51. Distance information, monitored facility name and measurement date and time input by the input device 51, acoustic signals digitized by the A / D conversion processing units 34 and 41, and acoustic spectra of the frequency analysis processing units 35 and 42 , The peak frequency of the positive peak frequency extraction unit 36, the peak frequency of the pseudo peak frequency extraction unit 43, the calculation result of the peak frequency deviation processing unit 44, the measurement availability result of the measurement position pass / fail judgment processing unit 45, and the like Store.
[0081]
In this way, monitoring of the output of the measurement position pass / fail determination processing unit 45 can determine the monitoring position at which the non-contact acoustic input device 26 monitors the characteristic sound during operation of the monitoring target equipment 3. The distance between the monitoring target facility 3 and the noncontact acoustic input device 26 that can measure the characteristic sound of the monitoring target facility 3 with the noncontact acoustic input device 26 indicates the monitoring position of the noncontact acoustic input device 26. .
[0082]
Next, an embodiment of the facility operating state acoustic monitoring device 21 for diagnosing the presence or absence of abnormality of the monitored facility 3 by monitoring the operation sound of the monitored facility 3 with the non-contact acoustic input device 26 is illustrated. 10 and FIG. 11 will be described. The same parts as those in FIG. 1 to FIG. 9 are described with the same reference numerals, and detailed description thereof will be omitted because it is duplicated. In this embodiment, the operation sound of the monitoring target equipment 3 is monitored with the same performance as that monitored by the contact acoustic input device 25 only by the non-contact acoustic input device 26. In this example, the non-contact acoustic input device 26 is installed at the output monitoring position to monitor the operation sound of the monitoring target equipment 3. Equivalent to monitoring with the contact-type sound input device 25 is that the S / N ratio of the monitoring signal by the non-contact-type sound input device 26 is deteriorated, but stable monitoring can be performed without being affected by noise. . Stable monitoring is achieved by extracting the peak frequency from the signal monitored by the contact-type sound input device 25 and using the sound pressure level of the monitor signal by the non-contact-type sound input device 26 as a monitor signal based on this peak frequency. Is done.
[0083]
The operating state acoustic monitoring device 21 of such an equipment includes a non-contact type acoustic input device 26, a distance measuring device 47, an input device 51, a personal computer 52, and a display device 50. The personal computer 52 includes, for example, an A / D conversion processing unit 34, a frequency analysis processing unit 35, a pseudo peak frequency extraction unit 43, a peak frequency correction processing unit 61, a determination processing unit 62, a data storage unit 49, and an A / D conversion processing unit 48. , A distance comparison processing unit 65, a search processing unit 66, a memory 53, a CPU 54, and the like.
[0084]
The non-contact type sound input device 26 detects an operation sound generated when the monitoring target facility 3 is operated and outputs an electric signal, and is a microphone, for example. The installation position of the microphone is contactless while monitoring the display position of the display device 50 by the monitoring position output from the monitoring data collection device 20 and stored in the data storage unit 49 or the operation state acoustic monitoring device 21 of the equipment. This is a position where the installation position of the acoustic input device 26 is moved and tuned to the monitoring position. That is, the installation of the non-contact acoustic input device 26 at the monitoring position read from the data storage unit 49 is performed by measuring the distance from the monitoring target facility 3 by using the distance measuring device 47 based on the monitoring position read from the data storage unit 49. Can be found and installed at the monitoring location. As a result, the plant supervisor patrols a large number of equipments sequentially while carrying the equipment operating state acoustic monitoring device 21, reads out the corresponding monitoring positions at the respective monitoring target equipment 3 positions from the data storage unit 49, and performs this monitoring. The non-contact type sound input device 26 is installed at the position, and the operation state of the monitoring target equipment 3 can be monitored from the operation sound.
[0085]
The A / D conversion processing unit 34 converts an analog acoustic signal obtained by detecting the operation sound of the monitoring target facility 3 by the non-contact acoustic input device 26 into a digital format and outputs the digital signal to the frequency analysis processing unit 35. The frequency analysis processing unit 35 analyzes the frequency of the acoustic signal digitized by the A / D conversion processing unit 34 by fast Fourier transform and outputs the acoustic spectrum to the pseudo peak frequency extraction unit 43.
[0086]
The pseudo peak frequency extraction unit 43 reads the peak frequency data of the monitoring target equipment 3 from the data storage unit 49, and the data storage unit stores the sound pressure level of the read peak frequency in the output waveform of the non-contact acoustic input device 26. 49, and when it is displayed on the display device 50, it is simultaneously output to the peak frequency correction processing unit 61. The output waveform of the pseudo peak frequency extraction unit 43 is a waveform as shown in FIG. The peak frequency correction processing unit 61 is shown in FIG. 8C in which a dotted line corresponding to the distance between the monitoring target facility 3 and the non-contact acoustic input device 26 is added to the waveform as shown in FIG. FIG. 8A shows the output value of the positive peak frequency extraction unit 36 of the output system of the contact-type acoustic input device 25 of the monitoring target equipment 3 stored in the data storage unit 49. The deviation from the sound pressure level value obtained from the waveform as shown is taken. If there is no deviation as a result of the deviations of FIGS. 8A and 8C, the peak frequency correction processing unit 61 outputs a waveform as shown in FIG. If there is a deviation as a result of the deviation, the peak frequency correction processing unit 61 moves the position of the non-contact acoustic input device 26 so that the deviation is eliminated, and adjusts the position to obtain a monitoring position.
[0087]
The determination processing unit 62 includes the waveform as shown in FIG. 8C from the peak frequency correction processing unit 61 and the waveform of the monitoring target facility 3 stored in the data storage unit 49 when it is normal (or abnormal). If they match, it is confirmed that they are in a normal operating state (if they match, an abnormal state is output). In this way, the determination processing unit 62 determines the strength of the correlation between the sound pressure level at the peak frequency detected by the non-contact acoustic input device 26 and the sound pressure level at the peak frequency read from the data storage unit 49. judge. That is, the determination processing unit 62 determines whether it is in a normal operation state or an abnormal operation state by determining whether there is a difference from the normal sound (or abnormal sound).
[0088]
The determination processing unit 62 correlates the sound pressure level of the peak frequency corrected for the separation distance by the peak frequency correction processing unit 61 and the sound pressure level of the peak frequency of the monitoring target equipment 3 read from the data storage unit 49. It is possible to determine whether the normal sound (or abnormal sound) is correct by setting a threshold value for the peak frequency deviation result due to the time deviation. The threshold value has a function of setting the threshold value on the display surface of the display device 50.
[0089]
The distance measuring device 47 is, for example, a laser distance measuring device, which is attached to the non-contact type acoustic input device 26, measures the distance from the monitoring target part of the monitoring target facility 3 to the non-contact type acoustic input device 26, Output distance information. When the distance information is converted into a digital signal by the A / D conversion processing unit 48 and input to the distance comparison processing unit 65, the distance information is simultaneously input and stored in the data storage unit 49.
[0090]
The search processing unit 66 searches and reads out the monitoring position data of the monitoring target equipment 3 from the data storage unit 49 and outputs it to the distance comparison processing unit 65. The distance comparison processing unit 65 compares the distance information measured by the distance measuring device 47 with the monitoring position data read from the data storage unit 49, and recognizes that the current position of the non-contact acoustic input device 26 is the monitoring position when they match. The monitoring position is output to the determination processing unit 62. The determination processing unit 62 displays on the display surface of the display device 50 that the current position of the non-contact acoustic input device 26 is the monitoring position. The input device 51 is, for example, a keyboard, and inputs a monitoring target equipment name and measurement date and time, and inputs and stores them in the data storage unit 49.
[0091]
The search processing unit 66 can measure the distance (monitoring position data) of the same monitoring target facility 3 from the past data stored in the data storage unit 49 in association with the monitoring target facility name input from the input device 51. The positive peak frequency processing result and the peak frequency deviation processing result of the detection output waveform by the contact-type acoustic input device 25 are searched. The past data includes various data output by the monitoring data collection device 20 and data output by the facility operating state sound monitoring device 21.
[0092]
The distance comparison processing unit 65 inputs the monitoring target facility name in the search processing unit 66 to search and extract the monitoring target facility name related data from the data storage unit 49, the monitoring target facility 3 and the current monitoring target facility 3. The distance between the non-contact acoustic input devices 26 is measured by the distance measuring device 47, and a distance deviation (Formula 5) is obtained based on the input distance data. The monitoring position of the non-contact type sound input device 26 is to adjust the position by moving the position of the non-contact type sound input device 26 so that the deviation value becomes “0”. The determination when the deviation value is “0” is performed by the determination processing unit 62 and displayed on the display device 50. The monitor can install the non-contact type sound input device 26 at the optimum position while looking at the display screen of the display device 50.
[0093]
[Formula 5]
Current position data of the non-contact acoustic input device 26 collected by the distance measuring device 47... K₁
Measurement position of the non-contact acoustic input device 26 read from the data storage unit 49... K
Distance deviation .... DELTA.K
KK₁= ΔK
[0094]
The measurement position of the non-contact acoustic input device 26 is moved so that ΔK becomes “0”. As a result, when ΔK becomes “0”, the non-contact type sound input device 26 is in substantially the same monitoring state as the monitoring by the contact type sound input device 25. In this manner, the positioning of the non-contact acoustic input device 26 at the monitoring position is completed.
[0095]
When the CPU 54 outputs a determination result that the installation position of the non-contact acoustic input device 26 is a monitoring position, the determination processing unit 62 uses the acoustic spectrum calculated by the frequency analysis processing unit 35 to generate data. Search is performed by inputting the monitoring target facility name from the storage unit 49 from the input device 51, and the same peak frequency as the extracted positive peak frequency processing result is extracted. The pseudo peak frequency extraction unit 43 is a sound pressure level value pmax corresponding to the peak frequency extracted from the acoustic spectrum output from the frequency analysis processing unit 35.₁~ Pmax_nIs output to the peak frequency correction processing unit 61.
[0096]
The peak frequency correction processing unit 61 calculates the sound pressure level value pmax corresponding to the extracted peak frequency.₁~ Pmax_nIn addition, the attenuation of the sound pressure level corresponding to the distance between the non-contact type sound input device 26 and the contact type sound input device 25 is added to substantially adjust the monitoring sensitivity by the contact type sound input device 25.
[0097]
That is, the peak frequency correction processing unit 61 adds the peak frequency deviation processing result searched and extracted from the data storage unit 49 by the search processing unit 66 to the extracted peak frequency (Formula 6), and the pseudo contact peak frequency The sound pressure level is obtained.
[0098]
[Formula 6]
Peak frequency deviation processing result retrieved from the data storage unit 49 by the search processing unit 66... Δpmax₁~ Δpmax_n
Sound pressure level at the peak frequency detected by the non-contact acoustic input device 26 this time ... dpmax₁~ Dpmax_n
Pseudo contact peak frequency with added deviation: apmax₁~ Apmax_n
If
dpmax₁+ Δpmax₁... dpmax_n+ Δpmax_n
= Apmax₁~ Apmax_n
[0099]
As shown in FIG. 11, the determination processing unit 62 is based on the distance determination means 62 a that determines whether or not the distance measured as the monitoring position based on the result of the distance comparison processing unit 65 and the output of the peak frequency correction processing unit 61. Next, the presence or absence of abnormal noise in the monitoring target equipment 3 is determined by the equipment operating state diagnosis means 62b.
[0100]
The distance determination unit 62 a determines a measurable distance by Equation 7 based on the result of the distance comparison processing unit 65.
[Formula 7]
Distance deviation ・ ・ ・ ΔK
ΔK = 0
[0101]
The facility operation state diagnosis means 62b is configured to detect the sound pressure level at the pseudo contact type peak frequency calculated by the peak frequency correction processing unit 61 by the CPU 54 and the sound of the positive peak frequency extracted from the data storage unit 49 by the search processing unit 66. Compared with the pressure level, the difference in the sound pressure level at the peak frequency is determined, and the presence or absence of abnormal sound is determined. If the CPU 54 has a difference in sound pressure level, the determination processing unit 62 outputs that there is a possibility of abnormal operation, and if there is no difference in sound pressure level, the normal operation is in progress. When it is stored in the data storage unit 49 in association with the date and time, it is displayed on the display device 50 at the same time.
[0102]
The display device 50 includes an acoustic signal digitally processed by the A / D conversion processing unit 34, an acoustic spectrum calculated by the frequency analysis processing unit 35, a peak frequency of the pseudo peak frequency extraction unit 43, and a peak frequency correction process. The peak frequency of the unit 61, the monitoring position information of the past non-contact acoustic input device 26 extracted from the data storage unit 49, the positive peak frequency processing result and the peak frequency deviation processing result, and the monitoring target facility input by the input device 51 The name, the measurement date and time, the monitoring position availability result of the distance determination means 65a, the result of the equipment operation state diagnosis means 62b, and the like are displayed.
[0103]
The data storage unit 49 includes an acoustic signal digitally processed by the A / D conversion processing unit 34, an acoustic spectrum calculated by the frequency analysis processing unit 35, a peak frequency of the pseudo peak frequency extraction unit 43, and a distance measuring device. The monitoring position information input by 47, the monitoring equipment name and measurement date input by the input device 51, and the result of the peak frequency correction processing unit 61 are stored.
[0104]
Next, an operation state monitoring method for the monitoring target equipment 3 by the equipment operation state acoustic monitoring device 21 will be described with reference to the flowchart of FIG. The same parts as those in FIGS. 1 to 11 are denoted by the same reference numerals, and detailed description thereof is omitted. The monitor circulates, for example, the power plant in which the monitoring target facilities 3 are arranged, carrying the facility operating state acoustic monitoring device 21 shown in FIG. 10 (F-1). At this time, the supervisor sets the operation state acoustic monitoring device 21 of the equipment in the positioning mode for installing the non-contact acoustic input device 26 at the monitoring position (F-2). When reaching a predetermined monitoring position, the monitor installs the non-contact acoustic input device 26 at the expected monitoring position of the monitoring target equipment 3 and adjusts the position (F-3). When the monitor can move to a predetermined monitoring position while looking at the display screen of the display device 50, it is displayed on the display screen as the optimum monitoring position (F-4).
[0105]
Next, the supervisor switches the operation state acoustic monitoring device 21 of the facility to the operation state monitoring mode of the monitoring target facility 3 (F-5). The operation sound (F-6) of the monitoring target equipment 3 detected by the non-contact type sound input device 26 is converted into a digital signal by the A / D conversion processing unit 34, and this digital signal is sent to the frequency analysis processing unit 35. Then, the acoustic spectrum of the frequency characteristic curve diagram with respect to the sound pressure level is input to the pseudo peak frequency extraction unit 43 (F-7).
[0106]
The pseudo peak frequency extraction unit 43 reads the positive peak frequency information extracted from the acoustic spectrum of the contact acoustic input device 25 output from the data storage unit 49, and outputs the frequency analysis processing unit 35 output corresponding to the positive peak frequency. The sound pressure level of the acoustic spectrum is extracted and output to the peak frequency correction processing unit 61 as the sound pressure level information of the pseudo peak frequency (F-8).
[0107]
Since the non-contact type acoustic input device 26 is located farther from the monitoring target facility 3 that is a sound source than the contact type acoustic input device 25, the peak frequency correction processing unit 61 has a sound pressure level at a pseudo peak frequency. The attenuation due to the separation distance is added, corrected, and output to the determination processing unit 62 (F-9). The determination processing unit 62 reads the sound pressure level value with respect to the peak frequency during normal operation of the monitored facility 3 stored in the data storage unit 49 (F-10), and the sound pressure level and the pseudo peak frequency A sound volume level obtained by adding attenuation to the sound pressure level is compared (F-11). If they match, it is determined that the vehicle is operating normally and displayed on the display device 50 (F-12). When information with the possibility of abnormal operation is stored in the data storage unit 49, it is simultaneously displayed on the display device 50 (F-13). It is desirable to display red when there is an abnormality. In this way, the operating state of the monitoring target equipment can be monitored (F-14).
[0108]
As described above, according to the above embodiment, the monitoring target equipment is monitored based on the peak frequency obtained from the output of the sound sensor provided directly or close to the monitoring target equipment. The monitoring position of the second sound sensor and the monitoring target equipment can be monitored with a signal having a large S / N ratio. As a result, the presence / absence of normal sound and abnormal sound in the monitoring target facility can be monitored with high accuracy and automatically determined.
[0109]
Furthermore, the method of distinguishing between the operation sound of the monitored equipment and the noise from the surroundings, which was a problem in the collection of acoustic data by the second sound sensor such as a microphone, the data collected in the contact state and the non-contact state Since the relationship with the data collected in is characterized, it can be measured at a monitoring position where noise such as environmental sound is not received as much as possible. Furthermore, by storing and registering the data collected at that time in the data storage unit and reading it from the data storage unit, the operating state of the monitored equipment is always maintained at the set monitoring position even if the monitor changes. Can be monitored.
[0110]
【The invention's effect】
As described above, according to the present invention, it is possible to accurately collect the operation sound from the monitoring target facility, and to monitor the operation sound accurately from the noise such as the operation sound from other surrounding facilities.
[Brief description of the drawings]
FIG. 1 is a flowchart for explaining an embodiment of an operation state sound monitoring method for equipment according to the present invention.
FIG. 2 is a circuit configuration diagram for specifically explaining a monitoring data collection process of FIG. 1;
FIG. 3 is a frequency characteristic curve diagram with respect to sound pressure level for specifically explaining the positive peak frequency extracting means of FIG. 2;
4 is a frequency characteristic curve diagram with respect to an output signal sound pressure level obtained by converting the output of the non-contact acoustic input device of FIG. 2 into a digital signal and performing frequency analysis.
5 is a circuit configuration diagram for specifically explaining a positive peak frequency extraction processing unit of FIG. 2; FIG.
6 is a characteristic curve diagram for explaining a peak value of a sound pressure level that is higher than an average value of sound pressure levels in the entire frequency band of FIG.
7 is a characteristic curve diagram for explaining a peak value of a sound pressure level that is higher than an average value of sound pressure levels for each specific frequency band in FIG. 3. FIG.
8 is a peak waveform diagram for explaining the relationship between frequency and sound pressure level for explaining a method for obtaining monitoring position information of the non-contact acoustic input device of FIG. 2;
FIG. 9 is a configuration diagram for explaining a determination method of a measurement position pass / fail determination processing unit in FIG. 2;
10 is a configuration diagram of an acoustic monitoring device for explaining the acoustic monitoring step of FIG. 1;
FIG. 11 is a configuration diagram for explaining a determination processing unit in FIG. 10;
12 is a flowchart for explaining a method of acoustically monitoring a facility to be monitored by the acoustic monitoring apparatus of FIG.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Equipment operating state monitoring method, 2 ... 1st sound sensor, 3 ... Monitoring object equipment, 4 ... 2nd sound sensor, 5 ... Monitoring data collection process, 6 ... Monitoring process, 11 ... Driving sound Collection step, 12 ... Peak frequency extraction step, 13 ... Monitoring position information output step, 15 ... Monitoring start step, 16 ... Monitoring step, 20 ... Monitoring data collection device, 21 ... Equipment operating state acoustic monitoring device, 25 ... Contact Type acoustic input device, 26 ... non-contact type acoustic input device, 34, 41, 48 ... A / D conversion processing unit, 35 ... frequency analysis processing unit, 36 ... positive peak frequency extraction unit, 37 ... coherence analysis processing unit, 42 ... frequency analysis processing unit, 43 ... pseudo peak frequency extraction unit, 44 ... peak frequency deviation processing unit, 45 ... measurement position pass / fail judgment processing unit, 47 ... distance measurement device, 49 ... data storage unit, 50 ... display device 51 ... input device, 52 ... PC, 61 ... peak frequency correction unit, 62 ... determining unit, 65 ... distance comparison processing unit, 66 ... retrieval processing unit.

Claims (8)

Collecting the operation sound unique to the monitoring target facility by the first sound sensor for a predetermined period and outputting the first operation sound signal;
Performing a frequency analysis process on the first driving sound signal to output a first acoustic spectrum signal indicating a relationship of a frequency to a volume output value;
Extracting a peak frequency at which the volume output value shows a peak from the first acoustic spectrum signal and a first volume output value of the peak frequency as monitoring information of a normal operation state, and storing the information in a storage device in advance;
A second sound sensor is provided at a monitoring position farther from the monitoring target facility than the first sound sensor, and the operation sound of the monitoring target facility is collected for the predetermined period to perform the second operation. Outputting a sound signal;
Performing a frequency analysis process on the second driving sound signal to output a second acoustic spectrum signal indicating the relationship of the frequency to the volume output value;
Extracting a second volume output value of the peak frequency read from the storage device from the second acoustic spectrum signal;
The monitoring position information for monitoring the operating state of the monitoring target equipment by the second sound sensor is obtained by obtaining the correlation between the first volume output value and the second volume output value read from the storage device. Storing the monitoring position information in the storage device;
In monitoring, the step of providing and monitoring the second sound sensor at the position of the monitoring position information of the monitoring target equipment;
Monitoring the presence or absence of abnormality of the monitoring target equipment by comparing the monitoring output by the second sound sensor and the monitoring information of the normal operation state stored in the storage device. The operation state acoustic monitoring method of the characteristic equipment.   2. The operation state acoustic monitoring method for equipment according to claim 1, wherein the installation position of the first sound sensor is provided in contact with a container wall surface of the equipment to be monitored.   The monitoring position information is obtained by performing at least one of a deviation process and a coherence analysis process of the first sound volume output value and the second sound volume output value at the peak frequency from the first sound spectrum signal and the second sound spectrum signal. The operation state sound monitoring method for equipment according to claim 1 or 2, wherein the information is obtained information.   The monitoring position information is obtained when the information obtained by at least one of the deviation processing and the coherence analysis processing of the first sound volume output value and the second sound volume output value falls within a predetermined range. The facility operating state sound monitoring method according to any one of claims 1 to 3, wherein the distance information is a distance information from the first sound sensor to the second sound sensor.   5. The facility operating state sound monitoring method according to claim 1, wherein the monitoring position information is distance information from a monitoring target facility to the second sound sensor.   The operation state of the facility according to any one of claims 1 to 4, wherein the monitoring information of the normal operation state and the monitoring position information of the second sound sensor are displayed on a display device. Acoustic monitoring method.   The monitoring of the monitoring target facility by the second sound sensor is performed by the volume output value of the peak frequency obtained from the monitoring output by the second sound sensor and the normal operation state stored in the storage device. The equipment according to any one of claims 1 to 5, wherein the information is compared with monitoring information, and abnormal information is output when a different signal is output, displayed on the display device and stored in the storage device. Driving state acoustic monitoring method. The operation sound unique to the monitoring target facility whose operating state is monitored from the driving sound is collected by the first sound sensor, and the second sound sensor is provided at a monitoring position farther from the monitoring target facility than the first sound sensor. The first sound spectrum and second sound spectrum information indicating the relationship of the frequency to the sound volume output value by subjecting the first sound volume signal and the second sound volume signal collected by the sound sensor to frequency analysis processing, the first sound spectrum information A storage device in which peak frequency information in an acoustic spectrum and a first volume output value at the peak frequency in the first acoustic spectrum are extracted and information indicating a normal operation state is stored;
If the second sound volume output value of the peak frequency is extracted from the second acoustic spectrum information, and the first sound volume output value and the second sound volume output value are compared, and there is a correlation, the second sound sensor. Means for storing in the storage device as the monitoring position of the monitoring target equipment,
The monitoring position information stored in the storage device is read, the operation sound of the monitoring target facility is monitored by the second sound sensor provided at the monitoring position, and the second sound sensor monitors A facility operating state sound monitoring apparatus comprising: a monitoring volume signal and means for determining the presence or absence of an abnormality by comparing with data indicating the normal operation state stored in the storage device.

Images