Equipment maintenance method and device, electronic equipment and computer program product

Abstract

The invention provides an equipment maintenance method and device, electronic equipment and a computer program product. The method comprises the following steps: acquiring detection data and health assessment results of each maintenance point; the health assessment result is obtained according to the detection data of each maintenance point and a health assessment model; inputting the detection data and the health assessment result into a comprehensive performance assessment model so as to determine a performance assessment result of the equipment according to the comprehensive performance assessment model; and inputting the performance evaluation result and the attribute data of the equipment into a maintenance decision model so as to determine a maintenance strategy of the equipment according to the maintenance decision model. The maintenance strategy of the equipment is automatically generated through multi-model fusion, on one hand, the maintenance skill requirement of a maintenance engineer can be reduced, the maintenance efficiency can be improved, the maintenance period can be shortened, and the maintenance process standard of the equipment can be unified; and on the other hand, human errors can be avoided, so that the overhaul effectiveness and reliability can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of equipment maintenance, and in particular to an equipment maintenance method, device, electronic equipment, and computer program product. Background Art

[0002] Equipment or parts require various types of inspection tools for inspection and maintenance during the discrete manufacturing process and the subsequent operation and maintenance service process. However, the entire maintenance process has the following pain points: (1) The maintenance process is complex and the maintenance cycle is long; (2) Maintenance records are mainly "manual paper records and manual entry after inspection", which have poor reliability and human errors; (3) Equipment maintenance relies on the experience of engineers, maintenance methods are diverse, and maintenance results vary. Summary of the invention

[0003] The technical problem to be solved by the present disclosure is to overcome the above-mentioned defects in the prior art and to provide an equipment maintenance method, device, electronic equipment, and computer program product.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] In a first aspect, a method for equipment maintenance is provided, wherein the equipment comprises a plurality of maintenance points, and the method for equipment maintenance comprises:

[0006] Obtaining the test data and health assessment results of each maintenance point; the health assessment results are obtained based on the test data and health assessment model of each maintenance point;

[0007] Inputting the detection data and the health assessment result into a comprehensive performance assessment model to determine a performance assessment result of the device according to the comprehensive performance assessment model;

[0008] Inputting the performance evaluation results and attribute data of the equipment into a maintenance decision model to determine a maintenance strategy for the equipment according to the maintenance decision model;

[0009] The maintenance decision model is obtained by training a neural network based on a first training sample, and the first training sample includes performance evaluation results and attribute data marked with maintenance strategies; the health assessment model is obtained by relearning based on a second training sample, and the second training sample is collected at each maintenance point after the equipment is maintained based on the maintenance strategy and the second training sample includes detection data marked with health assessment results; the model parameters of the comprehensive performance evaluation model are determined based on the model parameters of the health assessment model of all maintenance points.

[0010] Optionally, the model parameters of the comprehensive performance evaluation model are obtained by performing distributed learning on the health evaluation models of all maintenance points.

[0011] Optionally, the acquired detection data and health assessment results are compressed and encrypted data;

[0012] Inputting the detection data and the health assessment results into a comprehensive performance assessment model includes:

[0013] The detection data and health assessment results obtained by decompressing and decrypting the compressed encrypted data are input into the comprehensive performance evaluation model.

[0014] Optionally, the health assessment model is relearned according to the detection data of the maintenance point, the health assessment result of the maintenance point after the equipment maintenance based on the maintenance strategy, and the performance assessment result of the equipment after the maintenance based on the maintenance strategy;

[0015] Optionally, the comprehensive performance evaluation model is relearned according to the maintenance strategy, the operating data of the equipment, and the performance evaluation result of the equipment after maintenance based on the maintenance strategy; optionally, the health evaluation result is a weighted result of the first evaluation result and the second evaluation result; or, the health evaluation result is determined from the first evaluation result and the second evaluation result based on a voting mechanism; the first evaluation result is the output result of the health evaluation model, and the second evaluation result is determined according to the detection data and the decision tree;

[0016] And / or, the attribute data includes at least one of the following parameters: spare parts information, maintenance engineer information, equipment mechanism knowledge, maintenance expert knowledge, performance evaluation results of the equipment after maintenance based on the maintenance strategy, etc.

[0017] In a second aspect, an equipment maintenance system is provided for equipment maintenance, wherein the equipment includes a plurality of maintenance points, and the equipment maintenance system includes:

[0018] The edge device is used to obtain the detection data of each maintenance point, input the detection data into the health assessment model, and determine the health assessment result according to the output result of the health assessment model;

[0019] The edge device also sends the detection data and the health assessment result to a server;

[0020] The server is used to input the detection data and the health assessment result into a comprehensive performance assessment model to determine the performance assessment result of the device according to the comprehensive performance assessment model; and input the performance assessment result and attribute data of the device into a maintenance decision model to determine the maintenance strategy of the device according to the maintenance decision model;

[0021] The maintenance decision model is obtained by training a neural network based on a first training sample, and the first training sample includes performance evaluation results and attribute data marked with maintenance strategies; the health assessment model is obtained by relearning based on a second training sample, and the second training sample is collected at each maintenance point after the equipment is maintained based on the maintenance strategy and the second training sample includes detection data marked with health assessment results; the model parameters of the comprehensive performance evaluation model are determined based on the model parameters of the health assessment model of all maintenance points.

[0022] Optionally, when the detection data is input into a health assessment model, the edge device is specifically used to:

[0023] The detection data is subjected to empirical mode decomposition (EMD), the intrinsic mode function (IMF1) component obtained after the decomposition is calculated with the detection data to obtain a correlation coefficient as a feature vector, the feature vector is transformed into a two-dimensional image through a Gram angular field, and the two-dimensional image is input into the health assessment model.

[0024] Optionally, the health assessment result is a weighted result of the first assessment result and the second assessment result; or, the health assessment result is determined from the first assessment result and the second assessment result based on a voting mechanism; the first assessment result is the output result of the health assessment model, and the second assessment result is determined based on the detection data and the decision tree.

[0025] Optionally, before sending the detection data and the health assessment result to the server, the edge device further compresses and encrypts the detection data and the health assessment result to obtain compressed encrypted data, and sends the compressed encrypted data to the server;

[0026] Before the server inputs the detection data and the health assessment results into the comprehensive performance assessment model, it decompresses and decrypts the compressed encrypted data and inputs the obtained detection data and health assessment results into the comprehensive performance assessment model.

[0027] Optionally, the edge device further relearns the health assessment model based on the detection data of the maintenance point, the health assessment result of the maintenance point after the equipment maintenance based on the maintenance strategy, and the performance assessment result of the equipment after the maintenance based on the maintenance strategy;

[0028] Optionally, the server further relearns the comprehensive performance evaluation model according to the maintenance strategy, the operating data of the equipment, and the performance evaluation result of the equipment after maintenance based on the maintenance strategy.

[0029] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein when the processor executes the computer program, the equipment maintenance method described in any one of the first aspects is implemented.

[0030] In a fourth aspect, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the device maintenance method described in any one of the first aspects is implemented.

[0031] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.

[0032] The positive progressive effect of the present disclosure is that the present disclosure automatically generates maintenance strategies for equipment through multi-model fusion, and there is no need for maintenance engineers to manually analyze equipment performance and formulate maintenance strategies from detection data with different data types and inconsistent data formats. On the one hand, it can reduce the maintenance skill requirements of maintenance engineers, improve maintenance efficiency, shorten maintenance cycles, and unify equipment maintenance process standards; on the other hand, it can avoid the introduction of human errors, thereby improving maintenance effectiveness and reliability, thereby improving the quality and service life of equipment maintenance, and reducing long-term operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A flow chart of a device maintenance method provided for an exemplary embodiment of the present disclosure;

[0034] Figure 2 A structural schematic diagram of an equipment maintenance system provided by an exemplary embodiment of the present disclosure;

[0035] Figure 3 The present invention provides a schematic structural diagram of an electronic device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0036] The present disclosure is further described below by way of examples, but the present disclosure is not limited to the scope of the examples.

[0037] Prefixes such as "first" and "second" are used in the embodiments of the present disclosure only to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of prefixes such as ordinal numbers to distinguish description objects in the embodiments of the present disclosure does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and no unnecessary limitation should be constituted due to the use of such prefixes. In addition, in the description of the present embodiment, unless otherwise specified, the meaning of "plurality" is two or more.

[0038] In the embodiments of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of the device data involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0039] Figure 1 This is a flowchart of an equipment maintenance method provided by an exemplary embodiment of the present disclosure. The equipment maintenance method is used to perform maintenance on equipment. The equipment includes multiple maintenance points. The number and deployment location of the maintenance points can be set according to actual conditions. Each maintenance point is deployed with a check fixture, which is used to collect detection data of the maintenance point. The equipment can be of various types and various application scenarios.

[0040] See also Figure 1 , the equipment maintenance method comprises the following steps:

[0041] Step 101: Obtain the test data and health assessment results of each maintenance point; the health assessment results are obtained based on the test data and health assessment model of each maintenance point.

[0042] In one implementation, each inspection point corresponds to a health assessment model, and the inspection data of the inspection point is input into the health assessment model to obtain the health assessment result of the inspection point.

[0043] In one implementation, the same type of maintenance points correspond to one health assessment model, and the detection data of the same type of maintenance points are determined by the same health assessment model to determine the health assessment results. For example, all maintenance points on the same pipeline are determined by health assessment model A, and the maintenance points on the engine are determined by health assessment model B.

[0044] The inspection data of the inspection point is collected by the inspection fixtures deployed at the inspection point. The types of inspection fixtures deployed at the inspection point can include but are not limited to calipers, micrometers, dial indicators, micrometers, megohmmeters, temperature guns, hardness gauges, depth gauges, multimeters, thickness gauges, vibration gauges, ammeters, etc. The data types collected by each inspection fixture include binary, decimal, hexadecimal data, text, images, videos, etc.

[0045] Corresponding to the deployed inspection fixture, the inspection data collected by the inspection fixture includes at least one of the following parameters: length, shape error, position error, insulation resistance, temperature, hardness, depth, voltage, current, thickness, vibration, etc. In addition to the data collected by the inspection fixture, the inspection data can also include key information such as the inspection point location, inspection sequence, inspection fixture type, inspection time, and inspection personnel.

[0046] It is understandable that the types of inspection tools deployed at different types of inspection points are not completely the same, and the parameters contained in the inspection data collected at each inspection point are not completely the same.

[0047] The detection data of each detection point includes multiple types of data, and the data format is not unified (structured data and unstructured data are mixed). In order to facilitate subsequent processing, in one embodiment, the detection data can be preprocessed. The preprocessing may include but is not limited to pattern recognition, data cleaning, normalization, formatting, etc. The preprocessed detection data is subsequently used to repair the equipment.

[0048] Step 102: Input the detection data and health assessment results into a comprehensive performance assessment model to determine the performance assessment results of the equipment according to the comprehensive performance assessment model.

[0049] The comprehensive performance evaluation model is used to evaluate the performance of the equipment. Its input parameters are the test data and health evaluation results, and its output parameters are the performance evaluation results of the equipment. The performance evaluation results may include but are not limited to at least one of the following parameters: health level, remaining life, performance degradation value, etc.

[0050] In one embodiment, the comprehensive performance evaluation model can learn the correlation of the impact of each maintenance point on the equipment operation status, and determine the weight of each maintenance point according to the correlation, and comprehensively determine the performance evaluation result of the equipment according to the weight and the health evaluation result of each maintenance point. It should be noted that the weight is not a fixed value, and is dynamically adjusted according to the actual operation of the equipment.

[0051] Step 103: Input the performance evaluation results and attribute data of the equipment into the maintenance decision model to determine the maintenance strategy of the equipment according to the maintenance decision model.

[0052] The attribute data includes at least one of the following parameters: spare parts information, maintenance engineer information, equipment mechanism knowledge, maintenance expert knowledge, performance evaluation results of the equipment after maintenance based on the maintenance strategy, etc.

[0053] Spare parts information includes all parts information related to the equipment, including spare item information and spare parts information. Maintenance engineer information includes maintenance engineer experience, contact information, etc. Equipment mechanism knowledge includes the basic principles and methods used by the equipment in actual operation. Spare parts information and equipment mechanism knowledge can be obtained from the equipment manufacturer. Maintenance expert knowledge can be the experience summary of each maintenance engineer after each maintenance of the equipment.

[0054] The maintenance strategy obtained in step 103 can be provided to the maintenance engineer of the equipment, so that the maintenance engineer can perform maintenance on the equipment in sequence according to the received maintenance strategy. The maintenance strategy can include, but is not limited to, the maintenance sequence of maintenance points, the maintenance method of each maintenance point, and the performance evaluation result of the equipment after maintenance based on the maintenance strategy. The performance evaluation result output by the maintenance decision model can be used as a reference for the next maintenance.

[0055] The maintenance decision model in step 103 is obtained by training a neural network according to a first training sample, and the first training sample includes performance evaluation results and attribute data labeled with maintenance strategies.

[0056] After the health assessment model in step 101 is trained or put into use, it is further learned based on the second training sample to optimize the health assessment model. The second training sample is collected at each maintenance point after the equipment is maintained based on the maintenance strategy and the second training sample includes detection data marked with health assessment results.

[0057] In one embodiment, before being put into use, the health assessment model is obtained by training the neural network based on the third training sample. The third training sample can be the historical operation data and / or experimental data of the maintenance point of the equipment. The historical operation data and / or experimental data include detection data marked with health assessment results.

[0058] In one embodiment, the model parameters of the comprehensive performance evaluation model are determined based on the model parameters of the health evaluation model of all maintenance points. Before being put into use, the comprehensive performance evaluation model is determined based on the model parameters of the health evaluation model obtained by training the neural network according to the third training sample. After being put into use, the model parameters of the comprehensive performance evaluation model are determined based on the model parameters of the optimized health evaluation model.

[0059] In one embodiment, the model parameters of the comprehensive performance evaluation model are obtained by performing distributed learning on the health evaluation models of all maintenance points. Preferably, the model parameters of the comprehensive performance evaluation model are obtained by performing federated learning on the health evaluation models of all maintenance points.

[0060] In this embodiment, the maintenance strategy of the equipment is automatically generated through the fusion of multiple models. There is no need for maintenance engineers to manually analyze the equipment performance and formulate maintenance strategies from the detection data with different data types and inconsistent data formats. On the one hand, it can reduce the maintenance skill requirements of maintenance engineers, improve maintenance efficiency, shorten the maintenance cycle, and unify the maintenance process standards of equipment; on the other hand, it can avoid the introduction of human errors, thereby improving the effectiveness and reliability of maintenance, thereby improving the quality and service life of equipment maintenance, and reducing long-term operating costs.

[0061] In one embodiment, the health assessment result is a weighted result of the first assessment result and the second assessment result; or, the health assessment result is determined from the first assessment result and the second assessment result based on a voting mechanism; the first assessment result is the output result of the health assessment model, and the second assessment result is determined based on the detection data and the decision tree.

[0062] The following is an introduction to the construction method of the decision tree: the detection data is Fourier transformed, and the feature extraction of the Fourier transformed data is performed to obtain the features related to the health status of each detection point. The C45 method is used to make decisions, and the features with the best decision effect are selected to generate a decision tree. It should be noted that the type and number of features extracted from each detection point are determined according to the type and format of the detection data of the maintenance point, and the features extracted from each detection point are not exactly the same.

[0063] In this embodiment, the health assessment result is determined simultaneously by a decision tree model and a neural network model, and the outputs of the two are combined (for example, by weighted averaging or a voting mechanism) to obtain a final health assessment result, thereby improving the accuracy of the health assessment result.

[0064] In one embodiment, steps 101 to 103 are performed by a server, which may be, but is not limited to, a cloud server. The cloud server is deployed with a comprehensive performance evaluation model and a maintenance decision model. Each or each type of maintenance point (edge ​​side) is provided with an edge device, and the edge device is deployed with a health assessment model. The edge device can obtain the detection data (multimodal data / heterogeneous data) collected by various inspection tools, and input the multimodal data into the health assessment model to determine the health assessment result of the maintenance point according to the health assessment model. Each edge device also sends the health assessment results and detection data to the cloud server, so that the cloud server executes steps 101 to 103 to generate a maintenance strategy.

[0065] Edge devices can be implemented using multimodal data perception terminals, which can collect and format various types of data during the inspection of one or more inspection tools in real time and synchronously, and can forward data to the server using different communication methods such as Bluetooth, wireless communication technology WiFi, wireless communication technology Lora, and wireless network technology Zigbee according to different scenarios. Through multimodal data perception terminals, real-time synchronous collection and multi-dimensional forwarding of heterogeneous data of various types of inspection tools are achieved, the inspection process is digitized, the inspection efficiency and reliability are improved, and labor costs are reduced.

[0066] It should be noted that each multimodal data sensing terminal can synchronously collect the detection data of each detection point, or can detect and collect the detection data of each maintenance point in sequence according to the detection process package, and the present embodiment does not specifically limit this. Among them, the detection process package includes key information such as the detection point location, detection sequence, detection tool type, detection time, and detection personnel.

[0067] In this embodiment, the maintenance of equipment is achieved through the fusion of multi-models coordinated by the cloud edge side. The cloud server relies on multi-dimensional data, and obtains efficient and reliable maintenance methods and strategies through the data intercommunication and collaborative application of comprehensive performance evaluation models and maintenance decision models. The application of this embodiment can not only ensure the consistency, reliability and effectiveness of equipment maintenance, but also reduce the cost of equipment maintenance for enterprises.

[0068] When the model is put into use, it can also be relearned to ensure the prediction accuracy of the model.

[0069] In one embodiment, the health assessment model is relearned based on the detection data of the maintenance point, the health assessment result of the maintenance point after the equipment is repaired based on the maintenance strategy, and the performance assessment result of the equipment after the maintenance based on the maintenance strategy. Specifically, the relearning of the health assessment model is implemented by the edge detection device of each detection point, and the health assessment result of the maintenance point is subsequently determined based on the relearned health assessment model.

[0070] In one embodiment, the comprehensive performance evaluation model is relearned based on the maintenance strategy, the operation data of the equipment, and the performance evaluation results of the equipment after maintenance based on the maintenance strategy. Specifically, the relearning of the comprehensive performance evaluation model is implemented by the server. Figure 2 The server includes a cloud learning layer, an evaluation layer, and a decision layer. The server aggregates all received health assessment results and test data, distributes and processes them, and distributes one copy to the evaluation layer, which analyzes the current performance assessment results of the equipment through a comprehensive performance assessment model based on the correlation of each test point. One copy is forwarded to the decision layer, which integrates spare parts information, maintenance engineer information, equipment mechanism knowledge, maintenance expert knowledge, and the performance assessment results calculated by the evaluation layer and inputs them into the maintenance decision model to generate the equipment maintenance strategy and feed it back to the edge side. One copy is distributed to the cloud learning layer, which decouples the data received by the server and / or the data output by the model (including the health assessment results of the equipment after maintenance based on the maintenance strategy, and the historical health assessment results of the equipment before maintenance based on the maintenance strategy) and combines the operating data with the performance assessment results of the equipment after maintenance based on the maintenance strategy, completes the re-learning of the comprehensive performance assessment model, and then performs maintenance on the equipment based on the re-learned comprehensive performance assessment model. Among them, the data received by the server includes the detection data of each maintenance point of the equipment after maintenance based on the maintenance strategy and / or the historical detection data of each maintenance point of the equipment before maintenance based on the maintenance strategy; the operation data of the equipment includes the operation data of the equipment after maintenance based on the maintenance strategy and / or the operation data of the equipment before maintenance based on the maintenance strategy, and the operation data of the equipment includes the collection of maintenance data of all maintenance points and / or data collected through the equipment's data acquisition system, operation interface, etc.

[0071] In other implementations, the server feeds back the performance evaluation results of the equipment after maintenance to the multimodal sensing terminal of each detection point, so that the multimodal sensing terminal can re-learn, train and update the health assessment model of each detection point in combination with the detection data.

[0072] In one embodiment, in order to ensure the security of data transmission, the multimodal sensing terminal compresses and encrypts the health assessment results and the detection data to obtain compressed encrypted data, and sends the compressed encrypted data to the server. The server decompresses and decrypts the compressed encrypted data to obtain the detection data and health assessment results, and inputs them into the comprehensive performance evaluation model.

[0073] The compression may be, but not limited to, a coding conversion method, and the encryption algorithm may be, but not limited to, a symmetric encryption algorithm or an asymmetric encryption algorithm.

[0074] The embodiment of the present disclosure also provides an equipment maintenance system for equipment maintenance. Figure 2 ,The equipment contains multiple maintenance points, and the equipment maintenance system includes: edge devices and servers.

[0075] The edge device is used to obtain the inspection data of each maintenance point, input the inspection data into the health assessment model, determine the health assessment result according to the output result of the health assessment model, and send the inspection data and health assessment result to the server.

[0076] Edge devices can be implemented using multimodal data perception terminals, which can collect and format various types of data during the detection of one or more inspection tools in real time, and can forward data to the server using different communication methods such as Bluetooth, WiFi, Lora, and Zigbee according to different scenarios. Through multimodal data perception terminals, real-time synchronous collection and multi-dimensional forwarding of heterogeneous data of various types of inspection tools are achieved, the detection process is digitized, the detection efficiency and reliability are improved, and labor costs are reduced.

[0077] The server is used to input the detection data and health assessment results into the comprehensive performance evaluation model to determine the performance evaluation results of the equipment according to the comprehensive performance evaluation model; and input the performance evaluation results and attribute data of the equipment into the maintenance decision model to determine the maintenance strategy of the equipment according to the maintenance decision model.

[0078] Among them, the maintenance decision model is obtained by training the neural network according to the first training sample, and the first training sample includes performance evaluation results and attribute data marked with maintenance strategies; the health assessment model is obtained by re-learning according to the second training sample, and the second training sample is collected at each maintenance point after the equipment is maintained based on the maintenance strategy and the second training sample includes detection data marked with health assessment results; the model parameters of the comprehensive performance evaluation model are determined according to the model parameters of the health assessment model of all maintenance points.

[0079] The specific implementation method of the edge device and server in this embodiment when performing equipment maintenance is the same as the specific implementation method of the edge device and server mentioned in the above-mentioned equipment maintenance method embodiment. The specific implementation method of the edge device and server when performing equipment maintenance can be found in any of the above-mentioned embodiments and will not be repeated here.

[0080] In this embodiment, the maintenance strategy of the equipment is automatically generated through the multi-model fusion coordinated by the cloud-edge side. There is no need for maintenance engineers to manually analyze the equipment performance and formulate maintenance strategies from the detection data with different data types and inconsistent data formats. On the one hand, it can reduce the maintenance skill requirements of maintenance engineers, improve maintenance efficiency, shorten maintenance cycles, and unify the maintenance process standards of equipment; on the other hand, it can avoid the introduction of human errors, thereby improving the effectiveness and reliability of maintenance, thereby improving the quality and service life of equipment maintenance, and reducing long-term operating costs.

[0081] In one embodiment, when the detection data is input into the health assessment model to obtain the health assessment result, the edge device is specifically used to:

[0082] The detection data is subjected to EMD decomposition, and the IMF1 component obtained after the decomposition is calculated with the detection data (for example, every 500-1000 data points) to obtain the correlation coefficient as the feature vector. The feature vector is transformed into a two-dimensional image through the Gram angular field, and the two-dimensional image is input into the health assessment model, so that the health assessment model outputs the health assessment result of the maintenance point.

[0083] In one embodiment, the health assessment result is a weighted result of the first assessment result and the second assessment result; or, the health assessment result is determined from the first assessment result and the second assessment result based on a voting mechanism; the first assessment result is the output result of the health assessment model, and the second assessment result is determined based on the detection data and the decision tree.

[0084] Before the edge device sends the detection data and the health assessment result to the server, the edge device also compresses and encrypts the detection data and the health assessment result to obtain compressed encrypted data, and sends the compressed encrypted data to the server;

[0085] Before the server inputs the detection data and the health assessment results into the comprehensive performance assessment model, it decompresses and decrypts the compressed encrypted data and inputs the obtained detection data and health assessment results into the comprehensive performance assessment model.

[0086] In this embodiment, data security measures can be improved to prevent data from being tampered with during the forwarding process.

[0087] Optionally, the edge device further relearns the health assessment model based on the detection data of the maintenance point, the health assessment result of the maintenance point after the equipment maintenance based on the maintenance strategy, and the performance assessment result of the equipment after the maintenance based on the maintenance strategy;

[0088] Optionally, the server further relearns the comprehensive performance evaluation model according to the maintenance strategy, the operating data of the equipment, and the performance evaluation result of the equipment after maintenance based on the maintenance strategy.

[0089] As for the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The system embodiment described above is only illustrative, wherein the units described as separate components may or may not be physically separated, and the components as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the disclosed solution.

[0090] Figure 3 This is a structural diagram of an electronic device shown in an example embodiment of the present disclosure, the electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor, and when the processor executes the computer program, the equipment maintenance method described in any of the above embodiments is implemented. Figure 3 The electronic device 30 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0091] like Figure 3 As shown, the electronic device 30 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including the memory 32 and the processor 31).

[0092] The bus 33 includes a data bus, an address bus, and a control bus.

[0093] The memory 32 may include a volatile memory, such as a random access memory (RAM) 321 and / or a cache memory 322 , and may further include a read-only memory (ROM) 323 .

[0094] The memory 32 may also include a program tool 325 (or utility) having a set (at least one) of program modules 324, such program modules 324 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0095] The processor 31 executes various functional applications and data processing by running the computer program stored in the memory 32, such as the equipment maintenance method provided in any of the above embodiments.

[0096] The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 35. Furthermore, the electronic device 30 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 36. As shown, the network adapter 36 communicates with other modules of the electronic device 30 via a bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0097] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0098] The embodiments of the present disclosure further provide a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the device maintenance method provided by any of the above embodiments is implemented.

[0099] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0100] The embodiment of the present disclosure further provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the device maintenance method described in any one of the above items is implemented.

[0101] Among them, the program code for executing the computer program product of the present disclosure can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on the remote device.

[0102] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, but these changes and modifications all fall within the protection scope of the present disclosure.

Claims

1. A method for equipment maintenance, characterized in that: The equipment includes multiple maintenance points, and the equipment maintenance method includes: Obtaining the test data and health assessment results of each maintenance point; the health assessment results are obtained based on the test data and health assessment model of each maintenance point; Inputting the detection data and the health assessment result into a comprehensive performance assessment model to determine a performance assessment result of the device according to the comprehensive performance assessment model; Inputting the performance evaluation results and attribute data of the equipment into a maintenance decision model to determine a maintenance strategy for the equipment according to the maintenance decision model; Among them, the maintenance decision model is obtained by training the neural network according to the first training sample, and the first training sample includes performance evaluation results and attribute data marked with maintenance strategies; the health assessment model is obtained by re-learning according to the second training sample, and the second training sample is collected at each maintenance point and the second training sample includes detection data marked with health assessment results; the model parameters of the comprehensive performance evaluation model are determined according to the model parameters of the health assessment models of all maintenance points.

2. The equipment maintenance method according to claim 1, characterized in that: The model parameters of the comprehensive performance evaluation model are obtained by performing distributed learning on the health evaluation models of all maintenance points.

3. The equipment maintenance method according to claim 1, characterized in that: The acquired test data and health assessment results are compressed and encrypted data; Inputting the detection data and the health assessment results into a comprehensive performance assessment model includes: The detection data and health assessment results obtained by decompressing and decrypting the compressed encrypted data are input into the comprehensive performance evaluation model.

4. The equipment maintenance method according to any one of claims 1 to 3, characterized in that: Also includes: Relearning the health assessment model according to the detection data of the maintenance point, the health assessment result of the maintenance point after the equipment maintenance based on the maintenance strategy, and the performance assessment result of the equipment after the maintenance based on the maintenance strategy; and / or, relearning the comprehensive performance evaluation model according to the maintenance strategy, the operating data of the equipment, and the performance evaluation result of the equipment after maintenance based on the maintenance strategy; And / or, the health assessment result is a weighted result of the first assessment result and the second assessment result; or, the health assessment result is determined from the first assessment result and the second assessment result based on a voting mechanism; The first evaluation result is an output result of the health evaluation model, and the second evaluation result is determined according to the test data and the decision tree; And / or, the attribute data includes at least one of the following parameters: spare parts information, maintenance engineer information, equipment mechanism knowledge, maintenance expert knowledge, performance evaluation results of the equipment after maintenance based on the maintenance strategy, etc.

5. An equipment maintenance system, characterized in that: Used for equipment maintenance, the equipment includes multiple maintenance points, and the equipment maintenance system includes: The edge device is used to obtain the detection data of each maintenance point, input the detection data into the health assessment model, and determine the health assessment result according to the output result of the health assessment model; The edge device also sends the detection data and the health assessment result to a server; The server is used to input the detection data and the health assessment result into a comprehensive performance assessment model to determine the performance assessment result of the device according to the comprehensive performance assessment model; and input the performance assessment result and attribute data of the device into a maintenance decision model to determine the maintenance strategy of the device according to the maintenance decision model; Among them, the maintenance decision model is obtained by training the neural network according to the first training sample, and the first training sample includes performance evaluation results and attribute data marked with maintenance strategies; the health assessment model is obtained by re-learning according to the second training sample, and the second training sample is collected at each maintenance point and the second training sample includes detection data marked with health assessment results; the model parameters of the comprehensive performance evaluation model are determined according to the model parameters of the health assessment models of all maintenance points.

6. The equipment maintenance system according to claim 5, characterized in that: When the detection data is input into the health assessment model, the edge device is specifically used to: The detection data is subjected to EMD decomposition, the IMF1 component obtained after the decomposition is calculated with the detection data to obtain a correlation coefficient as a feature vector, the feature vector is transformed into a two-dimensional image through a Gram angular field, and the two-dimensional image is input into the health assessment model.

7. The equipment maintenance system according to claim 5, characterized in that: The health assessment result is a weighted result of the first assessment result and the second assessment result; or, the health assessment result is determined from the first assessment result and the second assessment result based on a voting mechanism; The first evaluation result is the output result of the health evaluation model, and the second evaluation result is determined according to the detection data and the decision tree.

8. The equipment maintenance system according to claim 5, characterized in that: Before the edge device sends the detection data and the health assessment result to the server, the edge device also compresses and encrypts the detection data and the health assessment result to obtain compressed encrypted data, and sends the compressed encrypted data to the server; Before the server inputs the detection data and the health assessment result into the comprehensive performance assessment model, the server decompresses and decrypts the compressed encrypted data and inputs the obtained detection data and health assessment result into the comprehensive performance assessment model; And / or, the edge device further relearns the health assessment model based on the detection data of the maintenance point, the health assessment result of the maintenance point after the equipment maintenance based on the maintenance strategy, and the performance assessment result of the equipment after the maintenance based on the maintenance strategy; And / or, the server further relearns the comprehensive performance evaluation model according to the maintenance strategy, the operating data of the equipment, and the performance evaluation result of the equipment after maintenance based on the maintenance strategy.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, characterized in that: When the processor executes the computer program, the equipment maintenance method according to any one of claims 1 to 4 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the equipment maintenance method according to any one of claims 1 to 4 is implemented.

Images