CN117611128A - A health management system and method considering single spare parts guarantee - Google Patents

Abstract

The invention provides a health management system and a method taking single spare part guarantee into consideration, belonging to the technical field of reliability engineering, wherein the system comprises a first processing subsystem, a second processing subsystem and a third processing subsystem, wherein the first processing subsystem is used for calculating PDF of the residual life of a part according to monitored single spare part degradation data, and calculating a cumulative distribution function and a reliability function of the residual life of the part; the second processing subsystem is used for determining optimal spare part acquisition and part replacement time through the multi-objective joint decision model based on the cumulative distribution function, the reliability function and the prediction result; and the third processing subsystem is used for judging whether the optimal spare part acquisition time is before the next state monitoring time, if so, installing and maintaining activities according to the predicted optimal spare part acquisition and part replacement time, otherwise, acquiring state monitoring data of the next monitoring time, and returning to the first processing subsystem. The invention solves the problems that the engineering equipment is inevitably subjected to health management and has overlong downtime and the usability of the system cannot be ensured.

Description

A health management system and method considering single spare parts guarantee

Technical field

The invention belongs to the technical field of reliability engineering, and in particular relates to a health management system and method considering single spare parts guarantee.

Background technique

In the era of Industrial Revolution 4.0, some concepts have emerged in parallel with this new revolution, such as Predictive Maintenance (PM), which plays a key role in sustainable manufacturing and production systems by introducing digital models of machine maintenance. . Due to the development of sensing technology, the data extracted from the production process has increased exponentially. PM can minimize machine downtime and associated costs, maximize machine lifecycle, and improve production quality and pace. This approach is often characterized by a very precise workflow, starting with project understanding and data collection and ending with the decision-making phase.

The first two parts focus on the large amount of monitoring data generated during the operation of complex systems. A lot of prediction information can be analyzed through missing data generation and RUL prediction methods. Forecast information includes RUL's probability density function (ProbabilityDensity Function, PDF), cumulative distribution function (Cumulative Distribution Function, CDF), reliability function (Reliability Function, RF), etc. Based on the RUL prediction information, maintenance staff can perform condition-based maintenance (CBM) on complex systems, formulate reasonable health management activities in advance for problems that may arise during operation, and implement preventive and remedial measures in advance.

Existing research mainly focuses on the situation of no-inventory spare parts guarantee, using multiple conditions such as operating cost and availability as constraint targets to carry out joint decision-making on condition-based maintenance and spare parts acquisition. However, in actual situations, there are sudden failures in the running components of complex systems, which increases the risk of downtime and maintenance costs, causing the system to fall out of a healthy state. In order to minimize the replacement time of components, reduce the risk of downtime, and improve its availability, a spare parts inventory system is urgently needed. Operation strategies to extend the operating life of the system.

Contents of the invention

In view of the above-mentioned deficiencies in the prior art, the present invention provides a sensitive text representation method based on contrastive learning, which solves the original semantic distortion caused by existing data enhancement and the inefficiency of contrastive learning training caused by the existing method using a computer vision contrast framework. , which in turn affects the problem of low text representation quality.

In order to achieve the above objects, the technical solutions adopted by the present invention are:

This solution provides a health management system that considers single spare parts guarantee, including:

The first processing subsystem is used to calculate the PDF of the remaining life of the component based on the monitored single spare part degradation data, and calculate the cumulative distribution function and reliability function of the remaining life of the component based on the PDF;

The second processing subsystem is used to determine the optimal spare parts acquisition and component replacement time through a multi-objective joint decision-making model based on the cumulative distribution function, reliability function and prediction results, according to the actual working environment requirements for component availability;

The third processing subsystem is used to determine whether the optimal spare parts acquisition time is before the next status monitoring time based on the determination result. If so, install maintenance activities based on the predicted optimal spare parts acquisition and component replacement time. Otherwise, obtain the next monitoring time. The status monitoring data at all times is returned to the first processing subsystem.

The beneficial effects of the present invention are: based on the predicted remaining life information, the present invention proposes a single inventory spare part single component operation system, taking into account high availability and low cost, a joint decision-making method for spare parts acquisition and spare parts replacement. Taking cost and availability as decision-making objectives at the same time, taking spare parts acquisition and component replacement time as decision variables, and handling the trade-off between cost and availability through the constructed decision boundary, minimizing the cost while meeting availability requirements, and obtaining multi-degraded parts. Optimum spare parts acquisition and component replacement time. Finally, the effectiveness of the proposed method was verified through the aero-engine data set. The obtained decision-making results can help operation and maintenance personnel arrange spare parts and replacement activities scientifically and reasonably, which has certain engineering application value.

Further, the expression of the PDF of the remaining life of the component is as follows:

in, PDF indicating the remaining life of the component, /> Indicates the total number of samples,/> Indicates the number of samples,/> Represents the probability density function PDF of a sampling result based on the optimal variational distribution,/> Indicates the predicted component RUL result,/> Represents the input single spare part degradation monitoring data,/> Represents the training set input data,/> Indicates the corresponding RUL label of the training set,/> Represents a sampling result based on the optimal variational distribution.

The beneficial effect of the above further solution is that obtaining the PDF of the remaining life is a prerequisite for health management and maintenance activities. Through this expression, the results of the remaining life predicted by the deep learning method can be unified with the statistical method.

Furthermore, the expression of the cumulative distribution function is as follows:

The expression of the reliability function is as follows:

in, Represents the cumulative distribution function,/> Represents the reliability function,/> Express/> RUL results predicted at time,/> Represents the input of new single spare part degradation monitoring data,/> Represents the cumulative distribution density function,/> Indicates time,/> represents the differential symbol.

The beneficial effect of the above further solution is that information such as the cumulative distribution function and reliability function based on the PDF of the remaining life can better serve the subsequent multi-objective joint decision-making model.

Furthermore, the second processing subsystem includes:

The first calculation module is used to construct a long-term average cost model and a long-term average availability model based on the cumulative distribution function and the reliability function respectively;

The second calculation module is used to build a multi-objective joint decision-making model based on the long-term average cost model and the long-term average availability model;

The third calculation module is used to obtain the prediction results of the remaining service life of a single spare part;

The fourth calculation module is used to determine the optimal spare parts acquisition and component replacement time through a multi-objective joint decision-making model based on the prediction results of the remaining service life of a single spare part and based on the requirements for component availability in the actual working environment.

The beneficial effect of the above further solution is that the present invention effectively determines the optimal spare parts acquisition and component replacement time by constructing a multi-objective joint decision-making model.

Furthermore, the expression of the long-term average cost model is as follows:

in, Represents the long-run average cost model,/> and/> Respectively represent the spare parts acquisition and component replacement time that need to be optimized,/> Represents the expected cycle cost,/> Indicates the expected cycle length,/> Represents the total cost of obtaining spare parts,/> Indicates the first/> time,/> Indicates the length of time between spare parts acquisition and spare part arrival, /> Represents the inventory maintenance cost per unit time, Represents the reliability function,/> Express/> RUL results predicted at time,/> Represents the time to complete the ineffective replacement activity,/> Indicates the time to complete preventive replacement activities,/> represents the cumulative distribution function of component replacement time, Represents the cumulative distribution function,/> Represents preventive replacement costs,/> Represents the cumulative distribution density function,/> Represents the replacement cost of failure,/> Indicates the cost of single monitoring,/> Indicates the cost of obtaining spare parts,/> Represents the total sustaining cost of inventory parts,/> Indicates inventory cost,/> Indicates replacement cost,/> Indicates the monitoring time point,/> Represents preventive replacement costs.

Furthermore, the expression of the long-term average availability model is as follows:

in, Represents the long-term average availability model,/> Indicates the expected running time of the component,/> Represents the distribution function of acquisition time.

Furthermore, the expression of the multi-objective joint decision-making model is as follows:

in, Represents the availability threshold.

Furthermore, the expression of the optimal spare parts acquisition and component replacement time is as follows:

in, Represent the optimal spare parts acquisition and component replacement time respectively,/> represent the lowest and highest points of acquisition time and replacement time, respectively.

The present invention provides a health management method considering single spare parts guarantee, which includes the following steps:

S1. Based on the monitored single spare part degradation data, calculate the PDF of the remaining life of the component, and calculate the cumulative distribution function and reliability function of the remaining life of the component;

S2. Based on the cumulative distribution function, reliability function and prediction results, and according to the actual working environment requirements for component availability, determine the optimal spare parts acquisition and component replacement time through a multi-objective joint decision-making model;

S3. Based on the determination result, determine whether the optimal spare parts acquisition time is before the next status monitoring time. If so, install maintenance activities based on the predicted optimal spare parts acquisition and component replacement time. Otherwise, obtain the status monitoring data at the next monitoring time. And return to step S1.

The beneficial effects of the present invention are: the present invention provides a joint decision-making research method for spare parts acquisition and replacement under the guarantee of a single spare part. Health management of engineering equipment will inevitably face the actual situation of excessive downtime and inability to guarantee system availability. Based on the predicted remaining life distribution of aero-engines, a multi-objective joint decision-making model that considers both cost and availability is established to maximize component availability while minimizing maintenance costs, thereby determining the optimal spare parts acquisition and component replacement time. In order to achieve better health management results, health management of engineering equipment will inevitably face the problems of long downtime and inability to guarantee system availability.

Description of drawings

Figure 1 is a schematic diagram of the system structure of the present invention.

Figure 2 is a schematic diagram of the single spare parts guarantee system in this embodiment.

Figure 3 is a schematic diagram of two situations that may occur during the entire life cycle in this embodiment.

Figure 4 is a PDF schematic diagram of the remaining life of different monitoring cycles in this embodiment.

Figure 5 is a schematic diagram of constraints in this embodiment.

Figure 6 is a schematic diagram of single cost decision-making in this embodiment.

Figure 7 is a schematic diagram of the joint decision-making results in this embodiment.

Figure 8 is a schematic flow chart of the method of the present invention.

Detailed ways

The specific embodiments of the present invention are described below to facilitate those skilled in the art to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the technical field, as long as various changes These changes are obvious within the spirit and scope of the invention as defined and determined by the appended claims, and all inventions and creations utilizing the concept of the invention are protected.

Before describing the present invention, the following parameters are explained:

PDF: probability density function;

RUL: remaining useful life.

Example 1

As shown in Figure 1, the present invention provides a health management system that considers single spare parts guarantee, including:

The first processing subsystem is used to calculate the PDF of the remaining life of the component based on the monitored single spare part degradation data, and calculate the cumulative distribution function and reliability function of the remaining life of the component based on the PDF;

The second processing subsystem is used to determine the optimal spare parts acquisition and component replacement time through a multi-objective joint decision-making model based on the cumulative distribution function, reliability function and prediction results, according to the actual working environment requirements for component availability;

The second processing subsystem includes:

The first calculation module is used to construct a long-term average cost model and a long-term average availability model based on the cumulative distribution function and the reliability function respectively;

The second calculation module is used to build a multi-objective joint decision-making model based on the long-term average cost model and the long-term average availability model;

The third calculation module is used to obtain the prediction results of the remaining service life of a single spare part;

The fourth calculation module is used to determine the optimal spare parts acquisition and component replacement time through a multi-objective joint decision-making model based on the prediction results of the remaining service life of a single spare part and based on the requirements for component availability in the actual working environment;

The third processing subsystem is used to determine whether the optimal spare parts acquisition time is before the next status monitoring time based on the determination result. If so, install maintenance activities based on the predicted optimal spare parts acquisition and component replacement time. Otherwise, obtain the next monitoring time. The status monitoring data at all times is returned to the first processing subsystem.

In this embodiment, based on the RUL prediction model, the present invention considers a single inventory spare part single component operation system model. In order to ensure the high availability operation of the entire system, the present invention studies and considers the spare parts of the single spare parts guarantee system based on the RUL remaining service life prediction information. Get-and-replace joint decision-making methods. We respectively introduce the single spare parts guarantee system of single inventory spare parts and single component operation system, and various situations of joint decision-making on replacement and spare parts acquisition according to the situation.

In this embodiment, the single spare parts guarantee system, that is, the single inventory spare parts single component operating system, is aimed at a single large, expensive, replaceable service component with a single inventory capability. In order to reduce system downtime and improve the operating efficiency of the entire system, its spare parts inventory The amount should be no less than one. As shown in Figure 2, in the actual operation cycle, two parts of the same model are first purchased at the same time, one of which is used as a running part and the other as a stock part. The stock parts are stored in the warehouse. After the original service parts suddenly fail, It can be replaced immediately to restore normal operation of the system. A single component is in service and in a running state, while a single component is in inventory redundancy and is in a storage state. Then consider obtaining spare parts for inventory components in advance, that is, after the original component fails and is replaced with one of the inventory spare parts, the inventory can still have a redundant spare part to prevent the new component that has just been replaced from suddenly failing, resulting in a vacant spare parts inventory. In particular, during the operating life cycle of a component, taking into account the cost of acquiring spare parts, the smallest single spare parts inventory system only needs to consider acquiring one component in the later period, except for acquiring two components at the same time for the first time. For the convenience of distinction, the operation of acquiring two parts for the first time is called original parts acquisition, and the operation of one part acquired later is called spare parts acquisition.

In this embodiment, under the single spare parts guarantee system, situation-based replacement is mainly divided into failure replacement and preventive replacement, as follows:

Invalid replacement: if in Momentary single spare part degradation monitoring data/> Exceeding the preset failure threshold/> , it is considered that the component fails, that is, at/> Invalid replacement is performed at all times, and the replacement cost is/> .

Preventive replacement: if in Momentary single spare part degradation monitoring data/> Does not exceed the preset failure threshold/> , calculated/> The remaining life at time is/> . Let the remaining life threshold of preventive replacement be/> , if/> , in order to prevent the component from failing within the next detection interval, preventive replacement of the component is performed, and the replacement cost is/> , and/> .

In this embodiment, both cost and availability are considered to obtain optimal spare parts acquisition and component replacement time. make is the current monitoring time,/> is the next monitoring time;/> ,/> respectively/> Predicted spare parts acquisition time and component replacement time at all times;/> is the long-term average cost function,/> is the long-term average availability function; the decision variable is/> and/> , the objective function is [/> ,/> ], that is, a multi-objective joint decision-making model that considers both cost and availability, and the constraints are/> . Since spare parts acquisition is something that will happen in the future,/> It is a basic requirement,/> This is because corresponding component replacement activities can only be carried out when spare parts exist. According to the renewal-reward theory, a life cycle of a component is represented by the time span between two replacements, including failure replacement and preventive replacement. Based on this, the multi-objective joint decision-making model is expressed as:

(1)

In this strategy, the joint decision-making problem of spare parts acquisition and parts replacement that considers both cost and availability based on forecast information includes the determination of the cost function, the availability function, and how to optimize the spare parts acquisition and parts replacement time. Before building a decision-making model, first of all, for the entire life cycle of single spare parts inventory system components, spare parts need to be obtained regardless of failure replacement or preventive replacement, as follows:

Based on these two acquisition schemes, two situations that may occur during the entire life cycle of the component are analyzed, as shown in Figure 3.

Situation 1: Pre-acquisition failure replacement: The parts have been obtained according to the predicted acquisition time and have arrived, but they fail before the predicted replacement time. The existing spare parts need to be replaced immediately, that is, the parts must be replaced immediately. There is a sudden failure between inventory maintenance costs of inventory spare parts and inventory costs of obtaining spare parts;

Situation 2: Pre-acquisition preventive replacement: The parts have been obtained according to the predicted acquisition time and have arrived. If the original parts have not failed before the predicted replacement time, the existing spare parts will be replaced according to the predicted replacement time, that is, the parts will be replaced after the predicted replacement time. Between sudden failures, there are inventory maintenance costs of inventory spare parts and inventory costs of obtaining spare parts.

In this embodiment, under the framework of PHM technology, to formulate a maintenance decision-making plan for degraded components, the prerequisite is to obtain the probability density function of the RUL of the degraded components. Existing research has proven that the traditional deep learning model using dropout is equivalent to the corresponding Bayesian deep learning model based on variational inference. It can characterize the uncertainty of the prediction result through the randomness of the weight and give the interval for predicting RUL. It is estimated that the PDF of the RUL prediction results under the deep learning model is as follows:

(2)

in, PDF indicating the remaining life of the component, /> Indicates the total number of samples,/> Indicates the number of samples,/> Represents the probability density function PDF of a sampling result based on the optimal variational distribution,/> Indicates the predicted component RUL result,/> Represents the input single spare part degradation monitoring data,/> Represents the training set input data,/> Indicates the corresponding RUL label of the training set,/> Represents a sampling result based on the optimal variational distribution.

After the RUL prediction network is trained through the training set, the corresponding RUL prediction results can be obtained by inputting new single-component degradation monitoring data. In order to facilitate subsequent expression, the probability density function of predicted RUL represented by formula (2) is simplified to , where,/> for/> Time predicted RUL,/> is the input monitoring data. From the probability density function of RUL, the cumulative distribution function of RUL can be further obtained/> and reliability function/> They are as follows:

(3)

(4)

in, Represents the cumulative distribution function,/> Represents the reliability function,/> Express/> RUL results predicted at time,/> Represents the input of new single spare part degradation monitoring data,/> Represents the cumulative distribution density function,/> Indicates time,/> represents the differential symbol.

The joint decision-making model proposed by the present invention is based on the analysis of RUL prediction information of multi-element degraded components. Therefore, as long as the PDF of the remaining life is obtained, the corresponding cumulative distribution function and reliability function can be further derived, and on this basis, a reasonable maintenance plan can be formulated strategies to ensure safe and stable operation of components.

In this embodiment, according to the update-reward theory, the long-term average cost model and the long-term average availability model of the single spare parts support system are respectively established.

In this embodiment, based on the above four situations that may occur during the entire life cycle of the component and the engineering reality, the following assumptions are made for the multi-objective joint decision-making model proposed by the present invention:

(1) Component condition monitoring is periodic, and the cost of a single condition monitoring is recorded as , the preventive replacement cost is recorded as , the failure replacement cost is/> , and/> ;

(2) The condition monitoring behavior of original service spare parts will not affect the remaining life of the component, and the storage degradation of redundant spare parts in the warehouse is negligible. After preventive replacement or failure replacement, the components will be restored to the original state. The time to complete the invalid replacement activity is , the time to complete the preventive replacement activities is/> ;

(3) There is a period of time between the acquisition of spare parts and the arrival of spare parts. It is generally considered that the length of this time is fixed, recorded as L;

(4) The cost of obtaining spare parts is , the inventory maintenance cost per unit time is recorded as/> ;

(5) Inventory cost can be recorded as , the inventory time is recorded as/> , the replacement cost is recorded as/> , which includes failure replacement costs and preventive replacement costs.

In this embodiment, based on the above model assumptions and the update-reward theory, the long-term average cost model can be expressed as:

(5)

in, Represents the expected cycle cost,/> Indicates the expected cycle length,/> and/> They are the spare parts acquisition and parts replacement time that need to be optimized respectively.

The first step is to derive the expected cycle cost , which can be expressed as:

(6)

Total cost of spare parts acquisition : Because the number of spare parts that need to be obtained in a life cycle is 1, the total cost of obtaining spare parts is/> .

Inventory parts total sustaining cost : In both cases 1 and 2, there are long-term inventory parts maintenance costs in the early stage. The total inventory time is divided into two parts. Among them, the inventory maintenance time of spare parts is from the start of operation to the current monitoring time/> , similarly, the inventory maintenance time for obtaining spare parts is/> , calculated as follows:

(7)

To get the total cost of inventory , first we need to derive the corresponding inventory time/> , there is inventory time in both case 1 and case 2.

(8)

Therefore, the inventory cost is . Replacement cost/> Including preventive replacement costs and failure replacement costs, it can be expressed as:

(9)

To sum up, we get the expected cycle cost for:

(10)

The second step is to derive the expected cycle length . Among the two situations that may occur throughout the life cycle of the component, Situation 1 The spare part has arrived, but the component fails before the predicted preventive replacement time, so the failure replacement is performed immediately; Scenario 2 The spare part has also arrived, and the component fails before the predicted preventive replacement time. If it does not fail within the replacement time, preventive replacement will be performed at the predicted replacement time. Based on the above analysis, the expected cycle length/> It can be expressed as:

(11)

Combining formula (10) and formula (11), we can get:

(12)

In this embodiment, availability is an important indicator for measuring the actual operating performance of a component. It is used to describe the probability or time occupancy expectation of the component being able to operate normally within a certain inspection period. The structure is as follows:

(13)

in, Indicates the expected running time of the component,/> Represents the expected cycle length, which is the sum of component running time and downtime. Among the four situations that may occur throughout the life cycle of a component, Scenario 1 and Scenario 2 have downtime caused by stockouts and replacement activities because the spare parts have not yet arrived when the component fails. Case 1 involves downtime caused by ineffective replacement, and case 2 involves downtime caused by preventive replacement.

The first step is to derive the expected running time , which can be expressed as:

(14)

Combining formula (11) and formula (10), the long-term average availability model can be obtained as:

(15)

in, Represents the long-run average cost model,/> and/> Respectively represent the spare parts acquisition and component replacement time that need to be optimized,/> Represents the expected cycle cost,/> Indicates the expected cycle length,/> Represents the total cost of obtaining spare parts,/> Indicates the first/> time,/> Indicates the length of time between spare parts acquisition and spare part arrival, /> Represents the inventory maintenance cost per unit time, Represents the reliability function,/> Express/> RUL results predicted at time,/> Represents the time to complete the ineffective replacement activity,/> Indicates the time to complete preventive replacement activities,/> represents the cumulative distribution function of component replacement time, Represents the cumulative distribution function,/> Represents preventive replacement costs,/> Represents the cumulative distribution density function,/> Represents the replacement cost of failure,/> Indicates the cost of single monitoring,/> Indicates the cost of obtaining spare parts,/> Represents the total sustaining cost of inventory parts,/> Indicates inventory cost,/> Indicates replacement cost,/> Indicates the monitoring time point,/> represents the preventive replacement cost, Represents the long-term average availability model,/> Indicates the expected running time of the component,/> Represents the distribution function of acquisition time.

In this embodiment, the long-term average cost model and the long-term average availability model are constructed respectively above. On this basis, a multi-objective optimization model is constructed to minimize the cost while meeting the availability requirements, thereby determining the optimal spare parts acquisition and components. Replacement timing.

(16)

at any monitoring time point , the established multi-objective joint decision-making model is shown in Equation (16). In order to solve the multi-objective optimization problem, the present invention constructs a decision boundary. Specifically, let the long-term average availability/> Meet the performance requirements of components in actual engineering/> requirements, and on this basis minimize the long-term average cost/> , then the multi-objective joint decision-making model can be rewritten as shown in formula (17).

(17)

In formula (17), It is the availability threshold, usually set according to the actual operation requirements of the project. In order to ensure that the parts meet the availability requirements, the spare parts acquisition and parts replacement time must be within a certain value range, that is/> . Given this, optimal spare parts acquisition and component replacement times/> It can be calculated by formula (18), that is, minimizing costs while meeting availability requirements, and determining the optimal spare parts acquisition and component replacement time.

(18)

in, Represent the optimal spare parts acquisition and component replacement time respectively,/> represent the lowest and highest points of acquisition time and replacement time, respectively.

The present invention will be further described below.

Aiming at the RUL prediction information of aero-engines, this invention studies a joint decision-making research method for spare parts acquisition and replacement under the guarantee of a single spare part, and verifies the effectiveness of the system proposed by this invention.

The present invention is based on the prediction results of a certain deep learning model. Figure 4 shows the remaining life PDF of engine No. 76 in the test set. It can be seen from this that the prediction effect of the RUL prediction method is good and has certain practical reference value.

Based on the above prediction results, the parameters in the multi-objective joint decision-making model are set as follows: Yuan, Yuan,/> Yuan,/> Yuan,/> Yuan,/> Yuan, availability threshold/> . Engine No. 76 has a total of CM data of 176 monitoring cycles. It is assumed that the interval between each monitoring cycle is 3 hours and the spare parts lead time is 72 hours. Table 1 shows the optimal spare parts acquisition and component replacement time obtained by the single cost decision-making model and the joint decision-making model proposed by the present invention at different monitoring time points. The specific results are as follows. Table 1 shows the decision-making of the two methods at different monitoring time points. result:

Table 1

As part operation time accumulates, degradation characteristics increase, and as the part approaches failure, its availability gradually decreases. In Table 1, the optimal spare parts acquisition and replacement time obtained through the multi-objective joint decision-making model is in advance of the single cost decision-making model, which is consistent with the actual operation and maintenance of the components.

In order to further illustrate the feasibility of the method of the present invention, taking the 135th monitoring cycle at the current monitoring time point, that is, the 405th hour, as an example, the long-term average cost is plotted in Figure 5 (a) and Figure 5 (b) respectively. and long-term average availability Functional images that change with spare parts acquisition and replacement time. Figure 6(a) and Figure 6(b) give a two-dimensional schematic diagram of the decision-making results of a single cost model. It can be seen from Figure 6 that at the 405th hour of the current monitoring time point, if only the lowest cost is the decision-making goal, the optimal spare parts acquisition and component replacement times are the 442nd and 480th hours respectively.

The model proposed by the present invention takes cost and availability as decision-making objectives at the same time, and constructs a multi-objective decision-making function based on prediction information, thereby obtaining optimal spare parts acquisition and component replacement time. In order to make the decision-making results more intuitive, taking the 405th hour of the current monitoring time as an example, it is assumed that the threshold requirement of long-term average availability is From the above, the value ranges of spare parts acquisition time and replacement time can be obtained respectively, and the minimum value of cost can be found within this range. Figure 7 shows the joint decision-making results. It can be seen from Figure 7 that the optimal spare parts acquisition and replacement times obtained by the multi-objective joint decision-making model are the 439th hour and the 477th hour respectively. This is because the degradation of components will increase with the As the running time accumulates, its availability will gradually decrease. To meet availability threshold requirements, spare parts acquisition and replacement times are pre-determined. Compared with the decision-making model that aims at the lowest single cost, the optimal spare parts acquisition time and replacement time obtained by the model proposed in the present invention are both in advance, which can help operation and maintenance personnel arrange spare parts acquisition and replacement activities in a timely manner to ensure that components can be completed efficiently and stably. Task.

In view of the realistic demand for maintenance costs and actual use efficiency of components guaranteed by single spare parts in the actual operation and maintenance process, the present invention proposes a maintenance decision-making method based on predictive information. The main work is as follows:

Based on the predicted remaining life information, the present invention proposes a single-inventory spare parts single-component operation system, taking into account high availability and low cost, a joint decision-making method for spare parts acquisition and spare parts replacement. Taking cost and availability as decision-making objectives at the same time, taking spare parts acquisition and component replacement time as decision variables, and handling the trade-off between cost and availability through the constructed decision boundary, minimizing the cost while meeting availability requirements, and obtaining multi-degraded parts. Optimum spare parts acquisition and component replacement time. Finally, the effectiveness of the proposed method was verified through the aero-engine data set. The obtained decision-making results can help operation and maintenance personnel arrange spare parts and replacement activities scientifically and reasonably, which has certain engineering application value.

Example 2

As shown in Figure 8, the present invention provides a health management method considering single spare parts guarantee, which includes the following steps:

S1. Based on the monitored single spare part degradation data, calculate the PDF of the remaining life of the component, and calculate the cumulative distribution function and reliability function of the remaining life of the component;

S2. Based on the cumulative distribution function, reliability function and prediction results, and according to the actual working environment requirements for component availability, determine the optimal spare parts acquisition and component replacement time through a multi-objective joint decision-making model;

S3. Based on the determination result, determine whether the optimal spare parts acquisition time is before the next status monitoring time. If so, install maintenance activities based on the predicted optimal spare parts acquisition and component replacement time. Otherwise, obtain the status monitoring data at the next monitoring time. And return to step S1.

This invention provides a joint decision-making research method for spare parts acquisition and replacement under the guarantee of a single spare part. Health management of engineering equipment will inevitably face the actual situation of too long downtime and inability to guarantee system availability. Based on the predicted remaining balance of aero-engines Life distribution, a multi-objective joint decision-making model that considers both cost and availability is established to maximize component availability while minimizing maintenance costs, and then determine the optimal spare parts acquisition and component replacement time to solve the health management of engineering equipment It is inevitable to face the problems of long downtime and inability to guarantee system availability in order to obtain better health management results.

Claims (9)

1. A health management system considering single spare part guarantee, comprising:

the first processing subsystem is used for calculating PDF of the residual service life of the component according to the monitored degradation data of the single spare part, and calculating a cumulative distribution function and a reliability function of the residual service life of the component based on the PDF;

the second processing subsystem is used for determining optimal spare part acquisition and part replacement time through a multi-objective joint decision model according to the requirements of the actual working environment on the availability of the parts based on the cumulative distribution function, the reliability function and the prediction result;

and the third processing subsystem is used for judging whether the optimal spare part acquisition time is before the next state monitoring moment according to the determination result, if so, installing and maintaining activities according to the predicted optimal spare part acquisition and part replacement time, otherwise, acquiring state monitoring data of the next monitoring moment, and returning to the first processing subsystem.

2. The health management system considering single spare part assurance according to claim 1, wherein the expression of PDF of the remaining life of the part is as follows:

wherein,PDF, which represents the remaining life of the component, N represents the total number of samples, N represents the number of samples,probability density function PDF representing one-time sampling result based on optimal variation distribution, < >>Representing predicted component RUL results, +.>Representing the input single spare part degradation monitoring data, X represents the training set input data, Y represents the corresponding RUL label of the training set, and +.>Representing a result of one sampling based on the optimal variation distribution.

3. The health management system considering single spare part security as claimed in claim 1, wherein the expression of the cumulative distribution function is as follows:

the expression of the reliability function is as follows:

R(l _k |X _1:k )＝1-F(l _k |X _1:k )

wherein F (l) _k |X _1:k ) Represents a cumulative distribution function, R (l _k |X _1:k ) Representing a reliability function, l _k Representing t _k Time-of-day prediction RUL results, X _1:k New preparation for representing inputPart degradation monitoring data, f (τ|X _1:k ) Represents a cumulative distribution density function, τ represents time, and d represents a differential sign.

4. The health management system under consideration of single spare parts assurance of claim 1, wherein the second processing subsystem comprises:

the first calculation module is used for respectively constructing a long-term average cost model and a long-term average availability model based on the cumulative distribution function and the reliability function;

the second calculation module is used for constructing a multi-objective joint decision model based on the long-term average cost model and the long-term average availability model;

the third calculation module is used for obtaining a prediction result of the residual service life of the single spare part;

and the fourth calculation module is used for determining optimal spare part acquisition and part replacement time through a multi-target joint decision model according to the requirement of the actual working environment on the availability of the parts based on the prediction result of the residual service life of the single spare part.

5. The health management system considering single spare part support according to claim 4, wherein the expression of the long-term average cost model is as follows:

wherein C is _k (t _o ,t _p ) Representing a long-term average cost model, t _o And t _p Respectively are provided withRepresenting spare part acquisition and part replacement time requiring optimization, EU representing desired cycle cost, EV representing desired cycle length, C _o Representing the total cost of acquisition of spare parts, t _k Represents the t _k The time, L, represents the length of time between spare part acquisition and spare part arrival, C _h Representing inventory maintenance cost per unit time, R (l) _k |X _1:k ) Representing a reliability function, l _k Representing t _k Time-of-day prediction RUL results, T _f Indicating the time to complete the failed replacement activity, T _p Indicating the time to complete the preventive replacement activity, F (t _p |X _1:k ) Cumulative distribution function representing component replacement time, F (l _k |X _1:k ) Representing cumulative distribution function, C _p Represents the cost of preventive replacement, f (l) _k |X _1:k ) Representing cumulative distribution density function, C _f Representing cost of replacement by failure, C _m Represents the cost of single monitoring, C _o Representing spare part acquisition cost, C _c Representing total maintenance costs of inventory components, C _H Representing inventory costs, C _R Represents replacement cost, k represents monitoring time point, C _p Representing the cost of preventive replacement.

6. The health management system considering single spare part support according to claim 5, wherein the expression of the long-term average availability model is as follows:

wherein A is _k (t _o ,t _p ) Represents a long-term average availability model, EO represents a component expected run time, F (t) _o +L|X _1:k ) A distribution function representing the acquisition time.

7. The health management system considering single spare part security of claim 6, wherein the expression of the multi-objective joint decision model is as follows:

maximizing C _k (t _o ,t _p )

Satisfy A _k (t _o ,t _p )≥ζ

t _k ≤t _o ≤t _p

Where ζ represents the availability threshold.

8. The health management system considering single spare part assurance according to claim 7, wherein the expression of the optimal spare part acquisition and part replacement time is as follows:

t _o ^* ,t _p ^* ＝argminC _k (t _o ,t _p )

t _o ,t _p ∈[t _min ,t _max ]

wherein t is _o ^* ,t _p ^* Respectively representing optimal spare part acquisition and part replacement time, t _min ,t _max The lowest point and the highest point of the acquisition time and the replacement time are respectively represented.

9. A health management method for executing the health management system under consideration of single spare part assurance according to any one of claims 1 to 8, characterized by comprising the steps of:

s1, calculating PDF of the residual service life of the part according to monitored single spare part degradation data, and calculating a cumulative distribution function and a reliability function of the residual service life of the part;

s2, determining optimal spare part acquisition and part replacement time through a multi-objective joint decision model according to the requirements of the actual working environment on the availability of the parts based on the cumulative distribution function, the reliability function and the prediction result;

and S3, judging whether the optimal spare part acquisition time is before the next state monitoring time according to the determination result, if so, installing maintenance activities according to the predicted optimal spare part acquisition and part replacement time, otherwise, acquiring state monitoring data of the next monitoring time, and returning to the step S1.