Equipment Maintenance Ability Evaluation Method, Device and Electronic Equipment

Abstract

The present application provides a method, apparatus, and electronic device for evaluating equipment maintenance capabilities. The method includes: obtaining historical data, reference operation data, and target operation data of the equipment to be evaluated, where the reference operation data is the most recent operation data of the equipment to be evaluated, and the target operation data is the initial equipment data when the equipment to be evaluated has not been operated; inputting the historical data into a data analysis model to obtain an analysis result set, where the analysis result set includes analysis results corresponding to various historical data, and analysis views corresponding to each analysis result; inputting the reference operation data and the target operation data into a reference evaluation model to obtain a reference evaluation result; and inputting the analysis result set and the reference evaluation result into a target evaluation model to obtain a target evaluation result, which not only improves the robustness and adaptability of the algorithm in complex scenarios, but also improves the accuracy of the evaluation result.

Description

Technical Field

[0001] This application relates to the technical field of equipment maintenance support capability evaluation, and particularly relates to a method, device and electronic equipment for evaluating equipment maintenance capability. Background Art

[0002] Evaluating equipment maintenance capability is an important means to ensure the long-term reliable operation of equipment, reduce maintenance costs and improve maintenance efficiency. The methods for evaluating equipment maintenance capability adopted in related technologies usually include the following important methods:

[0003] 1. Maintenance capability evaluation based on reliability analysis. This method usually relies on technologies such as failure mode analysis (FMEA) and fault tree analysis (FTA), and evaluates the maintenance capability by calculating the reliability indicators of the system (such as mean time between failures MTBF, mean time to repair MTTR, etc.). This method usually assumes that the failure modes of the equipment are known and unchanged, but there may be variable failure modes in actual operation, such as mechanical failures (wear, deformation), electrical failures (short circuit, component aging), environmental adaptability failures (temperature, humidity effects), and human factor failures (operation errors, improper maintenance), resulting in insufficient adaptability of the model.

[0004] 2. Maintenance evaluation technology based on condition monitoring and early warning. Li et al. (2018) proposed a maintenance evaluation method based on condition monitoring technology, which is widely used especially in industrial equipment. By real-time monitoring the working state of the equipment and analyzing the equipment using sensor data such as vibration, temperature, and pressure, its maintenance capability is evaluated. This method has high costs for sensor installation and data processing; the complexity of the equipment and environmental factors may affect the accuracy of the data.

[0005] 3. Maintenance capability evaluation based on multi-objective optimization. Zhang et al. (2019) proposed a multi-objective optimization method, which comprehensively evaluates multiple factors such as maintenance time, cost, and reliability as optimization objectives. This method takes into account that the evaluation of maintenance capability is not only to evaluate the reliability of the equipment, but also to comprehensively consider factors such as resources and time. This method needs to handle the conflict problem between multiple objectives, and the optimization algorithm has a high computational complexity and requires strong computational resource support.

[0006] In view of the above-mentioned multiple problems existing in the methods for evaluating equipment maintenance capability adopted in related technologies, no effective technical solution has been proposed yet. Summary of the Invention

[0007] In view of the above-mentioned multiple technical problems existing in the prior art, the embodiments of this application provide a method, device and electronic equipment for evaluating equipment maintenance capability to solve at least one of the above-mentioned technical problems.

[0008] According to one aspect of the embodiments of the present application, there is provided a method for evaluating equipment maintenance capabilities, including: obtaining historical data, reference operation data, and target operation data of the equipment to be evaluated, where the reference operation data is the most recent operation data of the equipment to be evaluated, and the target operation data is the initial equipment data when the equipment to be evaluated has not been operated; inputting the historical data into a data analysis model to obtain an analysis result set, where the analysis result set includes analysis results corresponding to various historical data respectively, and an analysis view corresponding to each analysis result; inputting the reference operation data and the target operation data into a reference evaluation model to obtain a reference evaluation result, where the reference evaluation model is a model for initially evaluating the equipment maintenance capabilities of the equipment to be evaluated; inputting the analysis result set and the reference evaluation result into a target evaluation model to obtain a target evaluation result, where the target evaluation model is a model for finally evaluating the equipment maintenance capabilities of the equipment to be evaluated.

[0009] According to another aspect of the embodiments of the present application, there is also provided an apparatus for evaluating equipment maintenance capabilities, including: an acquisition unit for obtaining historical data, reference operation data, and target operation data of the equipment to be evaluated, where the reference operation data is the most recent operation data of the equipment to be evaluated, and the target operation data is the initial equipment data when the equipment to be evaluated has not been operated; a first input unit for inputting the historical data into a data analysis model to obtain an analysis result set, where the analysis result set includes analysis results corresponding to various historical data respectively, and an analysis view corresponding to each analysis result; a second input unit for inputting the reference operation data and the target operation data into a reference evaluation model to obtain a reference evaluation result, where the reference evaluation model is a model for initially evaluating the equipment maintenance capabilities of the equipment to be evaluated; a third input unit for inputting the analysis result set and the reference evaluation result into a target evaluation model to obtain a target evaluation result, where the target evaluation model is a model for finally evaluating the equipment maintenance capabilities of the equipment to be evaluated.

[0010] Optionally, the above first input unit includes a classification subunit for classifying the historical data to obtain multiple reference historical data, and each reference historical data corresponds to a data category; an analysis subunit for respectively performing data analysis on the corresponding reference historical data by using the data analysis logic corresponding to the data category of each reference historical data to obtain the analysis result corresponding to each reference historical data; and a drawing subunit for respectively drawing the analysis view corresponding to each reference historical data according to the analysis process of each reference historical data.

[0011] Optionally, the above-mentioned second input unit includes a preprocessing subunit for preprocessing the reference operation data to obtain preprocessed data; a first extraction subunit for performing sparse feature extraction on the preprocessed data using the initial weights to obtain sparse key features; a first generation subunit for generating a first maintenance decision-making plan corresponding to the equipment to be evaluated according to the sparse key features and the target operation data; and a first evaluation subunit for evaluating the first maintenance decision-making plan through a bionic immune optimization mechanism, and selecting and optimizing the first maintenance decision-making plan to obtain a reference evaluation result.

[0012] Optionally, the above-mentioned first evaluation subunit includes a simulated maintenance module for performing simulated maintenance on the equipment to be evaluated according to the first maintenance decision-making plan to obtain a simulated maintenance result; an adjustment and update module for adjusting the initial weights according to the simulated maintenance result, and iteratively updating the first maintenance decision-making plan according to the sparse key features extracted using the updated initial weights, to obtain the initial maintenance decision-making plan included in the reference evaluation result and the initial evaluation result corresponding to the initial maintenance decision-making plan.

[0013] Optionally, the above-mentioned adjustment and update module includes an initialization sub-module for initializing the population parameters of the sparse key features in a hierarchical manner of equipment data to obtain an elite layer and a common layer, where the elite layer is initialized using the spatial uniform distribution method and the common layer is initialized using the Gaussian perturbation method; a first determination sub-module for determining the initial fitness values and pheromone concentration matrices of each individual in the elite layer and the common layer; and a second determination sub-module for determining the discoverer, follower, and guard according to multiple initial fitness values and multiple pheromone concentration matrices, and performing position updates.

[0014] Optionally, the above-mentioned second determination sub-module includes a first determination sub-module for determining the individuals corresponding to multiple target pheromone concentration matrices that meet the preset conditions in the multiple pheromone concentration matrices as discoverers, and updating the positions of the corresponding discoverers according to the pheromone concentration matrices corresponding to each discoverer; a second determination sub-module for updating the follower positions according to multiple initial fitness values using an adaptive neighborhood search dynamic update strategy; a third determination sub-module for iteratively updating multiple initial fitness values using the Cauchy mutation and reverse learning strategies to obtain multiple target fitness values; and a selection sub-module for selecting a part of the individuals corresponding to the multiple target fitness values as guards in proportion, and updating the positions of the guards using the Cauchy perturbation and reverse learning strategies.

[0015] Optionally, the above-mentioned third input unit includes a second extraction subunit, configured to perform sparse feature extraction on multiple analysis results in the analysis result set and the reference evaluation result by using a reference weight to obtain target sparse features; a second generation subunit, configured to generate a second maintenance decision-making plan corresponding to the equipment to be evaluated according to the target sparse features; a second evaluation subunit, configured to evaluate the second maintenance decision-making plan through a bionic immune optimization mechanism, and perform simulated maintenance on the equipment to be evaluated according to the evaluation result of the second maintenance decision-making plan to obtain a target simulated maintenance result; an adjustment and update subunit, configured to adjust the reference weight according to the target simulated maintenance result, and iteratively update the second maintenance decision-making plan according to the target sparse features extracted by using the updated reference weight to obtain the target maintenance decision-making plan included in the target evaluation result and the target evaluation result corresponding to the target maintenance decision-making plan.

[0016] Optionally, the above-mentioned equipment maintenance ability evaluation device further includes a determination unit, configured to determine a set of post-maintenance indexes of the equipment to be evaluated according to the target evaluation result after obtaining the target evaluation result; a first generation unit, configured to generate a first prompt message when the set of post-maintenance indexes meets a preset index condition, where the first prompt message is used to prompt to perform maintenance on the equipment to be evaluated by using the target maintenance decision-making plan; a second generation unit, configured to generate a second prompt message when the set of post-maintenance indexes does not meet the preset index condition, where the second prompt message is used to prompt the initial equipment information of the equipment to be evaluated and to prompt to replace the equipment to be evaluated.

[0017] According to another aspect of the embodiments of the present application, there is provided a computer-readable storage medium storing computer instructions for causing a computer to execute the above-mentioned equipment maintenance ability evaluation method.

[0018] According to another aspect of the embodiments of the present application, there is further provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to cause the at least one processor to execute the above-mentioned equipment maintenance ability evaluation method.

[0019] Compared with the prior art, the technical solution provided by the embodiments of the present application may include the following beneficial effects:

[0020] Through the above-mentioned equipment maintenance ability evaluation method of the present application, not only the robustness and adaptability of the algorithm in complex scenarios are improved, but also the accuracy of the evaluation result is improved. Description of the Drawings

[0021] To more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0022] Figure 1 It is a schematic diagram of the hardware environment of an optional equipment maintenance ability evaluation method according to an embodiment of the present invention;

[0023] Figure 2 It is a flowchart of an optional equipment maintenance ability evaluation method according to an embodiment of the present invention;

[0024] Figure 3 It is a schematic diagram of an optional equipment maintenance ability evaluation method according to an embodiment of the present invention;

[0025] Figure 4 It is a schematic structural diagram of an optional equipment maintenance ability evaluation device according to an embodiment of the present invention;

[0026] Figure 5 It is a schematic structural diagram of an optional electronic device according to an embodiment of the present invention. Specific Embodiments

[0027] In order to enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0028] Before introducing the technical solutions of the present application, some terms involved in the present application will be explained. The following relevant explanations can be combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application. The embodiments of the present application include at least some of the following contents.

[0029] Equipment maintenance ability refers to the comprehensive ability of equipment to maintain its normal operation in different operation stages and working environments, including the maintenance resources, skills, and time required. It measures the ability of the equipment to resume normal operation when a failure occurs, and usually includes multiple dimensions such as fault diagnosis, maintenance efficiency, maintenance quality, and availability.

[0030] Maintenance prediction refers to predicting possible future failures of equipment or the time for maintenance based on equipment operation data (such as sensor data, historical failure records, etc.). This usually relies on technologies such as machine learning and data analysis to discover potential failure problems in advance through the analysis of historical operation data of the equipment.

[0031] It should be noted that the terms "first", "second", etc. in the specification, claims and above-mentioned drawings of this application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0032] It should be noted that, without conflict, the embodiments and features in the embodiments of this application can be combined with each other. The following will describe this application in detail with reference to the drawings and in combination with the embodiments.

[0033] To solve at least one problem existing in the prior art, the embodiments of this application provide a method, device and electronic device for evaluating equipment maintenance capabilities. As an optional implementation, the above method for evaluating equipment maintenance capabilities can be applied, but is not limited to, in an equipment maintenance capability evaluation system composed of a terminal device 102 and a server 104 as shown in Figure 1 the figure. As shown in Figure 1 the figure, the terminal device 102 is connected to the server 104 through a network 110. The above network 110 may include, but is not limited to: wired networks, wireless networks. Among them, the wired network includes: local area network, metropolitan area network and wide area network, and the wireless network includes: Bluetooth, WIFI and other networks for realizing wireless communication. The above terminal device 102 may include, but is not limited to, at least one of the following: mobile phone (such as Android mobile phone, iOS mobile phone, etc.), laptop computer, tablet computer, handheld computer, MID (Mobile Internet Devices), PAD, desktop computer, smart TV, etc.

[0034] The above terminal device 102 is also provided with a display 106, a processor 108, and a memory 112. The display 106 can be used to display target evaluation results, etc. The processor 108 can be used to process the collected historical data, reference operation data, and target operation data. The memory 112 can be used to store historical data, reference operation data, target operation data, and target evaluation results. It can be understood that when the above terminal device 102 receives a user's equipment maintenance ability evaluation request, the terminal device 102 can obtain historical data, reference operation data, and target operation data through scanning or other means, and send the obtained historical data, reference operation data, and target operation data carried in the equipment maintenance ability evaluation request to the server 104 through the network 110. The server 104 can generate a target evaluation result based on the obtained historical data, reference operation data, and target operation data, so that the user can perform subsequent operations on the equipment to be evaluated according to the target evaluation result.

[0035] The above server 104 can be a single server, a server cluster composed of multiple servers, or a cloud server. The above server 104 includes a database 114 and a processing engine 116. Among them, the above database 114 can be used to store historical data, reference operation data, target operation data, target evaluation results, etc. The above processing engine 116 is used to process the above historical data, reference operation data, and target operation data.

[0036] According to one aspect of the embodiments of the present invention, the above equipment maintenance ability evaluation system can also perform the following steps: First, the terminal device 102 executes S102 as Figure 1 shown; then, the server 104 executes S104 to S112 as Figure 1 shown. In the above embodiments of the present invention, by adopting the above equipment maintenance ability evaluation method, not only the robustness and adaptability of the algorithm in complex scenarios are improved, but also the accuracy of the evaluation result is increased.

[0037] The above is only an example, and no limitation is made thereto in this embodiment.

[0038] As an alternative implementation manner, please refer to Figure 2, which shows a flowchart of the equipment maintenance ability evaluation method provided by an embodiment of the present application. The execution subject of each step of this method can be the terminal device or server introduced above. In the following method embodiments, for the convenience of description, only the execution subject of each step being "computer device" is used for illustration. This method may include at least one of the following steps (S202 to S208): S202, obtaining the historical data, reference operation data, and target operation data of the equipment to be evaluated, where the reference operation data is the most recent operation data of the equipment to be evaluated, and the target operation data is the initial equipment data when the equipment to be evaluated has never been operated; S204, inputting the historical data into the data analysis model to obtain an analysis result set, where the analysis result set includes the analysis results corresponding to various historical data and the analysis views corresponding to each analysis result; S206, inputting the reference operation data and the target operation data into the reference evaluation model to obtain a reference evaluation result, where the reference evaluation model is a model for initially evaluating the equipment maintenance ability of the equipment to be evaluated; S208, inputting the analysis result set and the reference evaluation result into the target evaluation model to obtain a target evaluation result, where the target evaluation model is a model for finally evaluating the equipment maintenance ability of the equipment to be evaluated.

[0039] It should be noted that the historical data in S202 above can be, but is not limited to, understood as the historical operation data of the equipment to be evaluated. The reference operation data is the most recent operation data of the equipment to be evaluated before the current equipment maintenance ability evaluation. The reference operation data includes operation time (for example, the average life of the equipment is two hundred hours, but the most recent operation only ran for one hundred hours before malfunctioning), component failure rate, training environment (the usage environment of the equipment), and component management level, etc. The target operation data is the initial equipment data when the equipment to be evaluated has never been operated after being manufactured. The initial equipment data includes some initial parameters, equipment type, and the average service life corresponding to the equipment type after the equipment to be evaluated is manufactured. The historical data, reference operation data, and target operation data are all multi-dimensional data, and the specific dimensions are not limited in this application.

[0040] The data analysis model in S204, the reference evaluation model in S206, and the target evaluation model in S208 above can be, but are not limited to, understood as three sub-models included in the equipment maintenance ability evaluation model. By inputting the above historical data, reference operation data, and target operation data into the equipment maintenance ability evaluation model, the three sub-models in the equipment maintenance ability evaluation model operate on the input data respectively, so that the equipment maintenance ability evaluation model outputs the final target evaluation result.

[0041] The data analysis model in S204 is used to perform data analysis on various historical data respectively to obtain the analysis results of each type of historical data. Specific data analysis includes, for example: analyzing the changes in the running time, the number of failures, the failure rate, and the failure modes (such as mechanical failures (wear, deformation), electrical failures (short circuit, component aging), environmental adaptability failures (temperature, humidity effects), human factor failures (operation errors, improper maintenance), etc.) of the equipment to be evaluated during historical operation.

[0042] The specific analysis process of the above data analysis model includes: S1, classifying the historical data to obtain various reference historical data, and each type of reference historical data corresponds to a data category; S2, respectively using the data analysis logics corresponding to the data categories of each type of reference historical data to perform data analysis on the corresponding reference historical data to obtain the analysis results corresponding to each type of reference historical data; S3, respectively drawing the analysis views corresponding to each type of reference historical data according to the analysis processes of each type of reference historical data.

[0043] The operation in S2 above can be understood, but is not limited to, that: each data category of the reference historical data corresponds to a data analysis logic. Determine the data analysis logic corresponding to each type of reference historical data according to the data category of each type of reference historical data, and then use the data analysis logic corresponding to each type of reference historical data to perform data analysis to obtain the analysis results of the corresponding reference historical data. The process of analyzing the reference historical data specifically includes: extracting sparse features from each type of reference historical data according to the data dimensions of the reference historical data to obtain the historical data features corresponding to each type of reference historical data; performing data analysis on the historical data features to obtain the analysis results corresponding to each type of reference historical data. Then, according to the analysis processes of each type of reference historical data, draw the analysis views corresponding to each type of reference historical data. For example: map the initial running duration of the equipment to be evaluated and the running duration after each repair to obtain a line chart of the running changes of the equipment to be evaluated; another example: count the failure modes of the equipment to be evaluated each time and the number of occurrences of each failure mode to obtain a bar chart of the types and quantities of failure modes, etc. Generate the analysis views corresponding to each type of reference historical data while obtaining the analysis results, which is convenient for users to more intuitively see the places where the equipment to be evaluated is prone to failures, and at the same time convenient for feedback to the manufacturer of the equipment to be evaluated to optimize the production of the equipment to be evaluated.

[0044] The reference evaluation model in S206 above can be, but is not limited to, understood as being used to generate an initial maintenance decision-making plan for the reference evaluation result and the initial evaluation result corresponding to the initial maintenance decision-making plan. That is, in order to generate the first relatively optimal maintenance plan, the initial evaluation result is used to indicate the initial maintenance data and the initial maintenance result corresponding to the initial maintenance decision-making plan. The initial maintenance data includes the initial maintenance duration, the initial maintenance location, the initial maintenance cost, etc. The initial maintenance result is used to indicate the availability of the equipment to be evaluated after the equipment to be evaluated is maintained using the initial maintenance decision-making plan. The target evaluation model in S208 can be, but is not limited to, understood as an optimization of the first relatively optimal maintenance plan to obtain the final maintenance plan. That is, the target maintenance decision-making plan included in the target evaluation result, and the target evaluation result corresponding to the target maintenance decision-making plan. The target evaluation result is used to indicate the target maintenance data and the target maintenance result corresponding to the target maintenance decision-making plan. The target maintenance data includes the target maintenance duration, the target maintenance location, the target maintenance cost, etc. The target maintenance result is used to indicate the availability of the equipment to be evaluated after the equipment to be evaluated is maintained using the target maintenance decision-making plan.

[0045] It should be noted that the data input into the target evaluation model in S208, in addition to the analysis result set and the reference evaluation result, also includes the maintenance goal and the constraint condition. The maintenance goal and the constraint condition are data input or selected by the user. The maintenance goal, for example: maximizing the equipment operation efficiency or minimizing the maintenance cost, etc. The constraint condition, for example: ensuring the availability of the equipment, the ability of the technical personnel, the availability of the spare parts, etc.

[0046] Through the above implementation manner of the present application, various types of data and various models are used to process and analyze the data, which improves the robustness and adaptability of the algorithm in complex scenarios. At the same time, by using two evaluation models to generate and optimize the evaluation results for different input data, the accuracy of the evaluation results is improved.

[0047] As an optional implementation manner, inputting the reference operation data and the target operation data into the reference evaluation model to obtain the reference evaluation result includes: S1, performing data preprocessing on the reference operation data to obtain preprocessed data; S2, extracting sparse key features from the preprocessed data using the initial weight to obtain sparse key features; S3, generating a first maintenance decision-making plan corresponding to the equipment to be evaluated according to the sparse key features and the target operation data; S4, evaluating the first maintenance decision-making plan through the bionic immune optimization mechanism, and selecting and optimizing the first maintenance decision-making plan to obtain the reference evaluation result.

[0048] The data preprocessing in S1 above is used to improve the effectiveness and accuracy of the reference operation data. The initial weight is the feature extraction weight set in advance. The operation of performing sparse feature extraction on the preprocessed data is specifically to perform sparse feature extraction on the preprocessed data according to the data dimension of the preprocessed data. It should be noted that in this application, the sparse feature extraction operations on the reference historical data, the preprocessed data, and the subsequent reference evaluation results (the sparse feature extraction operation on the subsequent reference evaluation results is also an extraction operation based on the data dimension of the reference evaluation results) can be implemented by SparseMLP (Sparse Multi-Layer Perceptron) with the dimension extraction technology added.

[0049] The operation in S3 above can be, but is not limited to, understood as the initial maintenance decision-making plan generated by the reference evaluation model based on the sparse key features and the target operation data. It should be noted that the first maintenance decision-making plan includes multiple initial plans (that is, a set of initial plans is generated). The operation in S4 can be, but is not limited to, understood as an operation to optimize the initial maintenance decision-making plan. The operations in S3 to S4 above can be implemented by the improved SSA algorithm (Sparrow Search Algorithm) incorporated in the reference evaluation model. The improved SSA algorithm in this application can be called the RSSA algorithm (Robust Sparrow Search Algorithm). RSSA aims to enhance the robustness and stability of the standard sparrow search algorithm when dealing with complex problems. By adjusting the algorithm, it can better adapt to dynamic environments or situations with high noise, thereby improving the performance of the algorithm in practical applications. The RSSA algorithm simulates the antibody generation and selection mechanism in the biological immune system and optimizes the decision-making ability of complex systems through the learning and memory process. Through the adaptive and learning mechanism, the memory bank mechanism, diversity preservation, and antibody evolution and elimination, the optimal maintenance decision-making plan is dynamically generated in the equipment maintenance ability evaluation. SparseMLP is an improved neural network model that uses sparse connections to reduce computational complexity and is suitable for equipment maintenance evaluation dealing with high-dimensional and large-scale data. Through the SparseMLP model, in-depth analysis of the historical maintenance data of the equipment can be carried out, and reasonable decisions can be generated based on data-driven. Combining RSSA and SparseMLP to construct an efficient equipment maintenance ability evaluation model can give full play to the respective advantages of the two and achieve a more intelligent and accurate decision-making process.

[0050] Through the above implementation manners of this application, SparseMLP (Sparse Multi-Layer Perceptron) with the dimension extraction technology added is used to perform sparse feature extraction on the reference historical data, providing efficient and concise input for the subsequent reference evaluation model. By using SparseMLP (Sparse Multi-Layer Perceptron) with the dimension extraction technology added to perform sparse feature extraction on the preprocessed data, and subsequent sparse feature extraction on the reference evaluation results, concise and accurate optimization feature data can be provided for the optimization of the maintenance decision-making plan.

[0051] As an alternative implementation, the above-mentioned evaluation of the first maintenance decision-making plan through the bionic immune optimization mechanism and the optimization of the first maintenance decision-making plan to obtain the reference evaluation result include: S1, performing simulated maintenance on the equipment to be evaluated according to the first maintenance decision-making plan to obtain the simulated maintenance result; S2, adjusting the initial weight according to the simulated maintenance result, and iteratively updating the first maintenance decision-making plan according to the sparse key features extracted by using the updated initial weight, to obtain the initial maintenance decision-making plan included in the reference evaluation result and the initial evaluation result corresponding to the initial maintenance decision-making plan.

[0052] The simulated maintenance result in the above S1 can be, but is not limited to, understood as the data after simulated maintenance obtained by performing simulated maintenance on the equipment to be evaluated according to the first maintenance decision-making plan. The operation in S2 can be, but is not limited to, understood as adjusting the extraction weight for sparse feature extraction of the preprocessed data according to the simulated maintenance result, and then performing sparse feature extraction on the preprocessed data by using the adjusted initial weight to obtain new sparse key features, and then repeating the foregoing operation to iteratively update the first maintenance decision-making plan until the fitness function converges, to obtain the initial maintenance decision-making plan and the initial evaluation result corresponding to the initial maintenance decision-making plan.

[0053] Through the above implementation of the present application, by using the above method, the obtained initial maintenance decision-making plan can be accurate enough, thereby further improving the accuracy of the subsequent obtained target maintenance decision-making plan.

[0054] As an alternative implementation, the above-mentioned iterative update of the first maintenance decision-making plan according to the sparse key features extracted by using the updated initial weight includes: S1, initializing the population parameters of the sparse key features by using the method of equipment data stratification to obtain an elite layer and a normal layer, wherein the elite layer is initialized by using the spatial uniform distribution method, and the normal layer is initialized by using the Gaussian perturbation method; S2, determining the initial fitness value and pheromone concentration matrix of each individual in the elite layer and the normal layer; S3, determining the discoverer, follower and vigilant according to the multiple initial fitness values and multiple pheromone concentration matrices, and performing position update.

[0055] The operation in S1 can be, but is not limited to, understood as initializing the population parameters through the method of equipment data stratification, dividing the population (sparse key features) into an elite layer and a normal layer, and initializing the elite layer by using the spatial uniform distribution method: evenly scattering points within the Voronoi region of the anchor node to ensure that the initial positions cover the possible region of the global optimal solution. The normal layer is initialized by using the Gaussian perturbation method: taking the position of the anchor node as the center point, within the radius Generate an initial solution randomly within a range. Then determine the initial fitness value and pheromone concentration matrix of each individual in the elite layer and the ordinary layer. It can be understood that the method for determining the initial fitness value and pheromone concentration matrix can adopt existing methods, and the present application does not make limitations. The operations in S3 can be, but are not limited to, understood as determining the discoverer, follower, and vigilant according to multiple initial fitness values and multiple pheromone concentration matrices, and respectively updating the positions of the discoverer, follower, and vigilant. It can be understood that each individual corresponds to an initial solution in the first maintenance decision plan, and the process of updating the individual position is the process of iteratively updating each initial solution in the first maintenance decision plan.

[0056] Through the above implementation manners of the present application, by adopting hierarchical initialization and pheromone guidance, the robustness of the algorithm in complex scenarios is improved.

[0057] As an optional implementation manner, the above determining the discoverer, follower, and vigilant according to multiple initial fitness values and multiple pheromone concentration matrices, and performing position update includes: S1, determining the individuals corresponding to multiple target pheromone concentration matrices that meet the preset conditions in the multiple pheromone concentration matrices as the discoverers, and updating the positions of the corresponding discoverers according to the pheromone concentration matrix corresponding to each discoverer; S2, adopting a dynamic update strategy of adaptive neighborhood search to update the follower positions according to multiple initial fitness values; S3, adopting the Cauchy mutation and reverse learning strategy to iteratively update multiple initial fitness values to obtain multiple target fitness values; S4, selecting a part of the individuals corresponding to multiple target fitness values as the vigilant according to a proportion, and adopting the Cauchy perturbation and reverse learning strategy to update the positions of the vigilant.

[0058] It should be noted that the process of updating the individual positions in the above S1 to S4 is the process of iteratively updating each initial solution in the first maintenance decision plan. The preset condition in S1 is a condition set in advance. The operation in S1 can be, but is not limited to, understood as determining the individuals corresponding to the matrices with better pheromone concentration in the multiple pheromone concentration matrices as the discoverers, and then updating the positions of the discoverers according to the pheromone concentration, that is, introducing the pheromone concentration matrix can know the movement of the discoverers, and the pheromone concentration is dynamically updated through the RSSI stability of the anchor nodes, and the pheromone concentration in the area with high stability increases. Specifically, the following formula can be used to update the position of the discoverer:

[0059]

[0060] In the formula is the updated position of the discoverer, represents at the th iteration, the rd sparrow individual at the dimensional position information; is the pheromone influence factor, which decreases with the number of iterations, is the pheromone concentration matrix corresponding to the th sparrow individual, and

[0061] The followers prey after the discoverers and search near the positions of the discoverers. To avoid the algorithm falling into local optimum and improve the global search ability, a dynamic update strategy of adaptive neighborhood search is adopted in the process of updating the positions of the followers. Specifically, the following formula can be used to update the positions of the followers:

[0062]

[0063] where, is the updated position of the follower, is the optimal solution in the current th iteration, represents the neighborhood search step size, which can be understood as, but is not limited to, the amplitude of the follower moving towards the optimal solution, represents randomly selecting an individual from the sparrow population, represents at the th iteration, the dimensional position information of the th sparrow individual, is the average fitness value of all sparrow individuals (in this application, the individual is short for the sparrow individual), is the fitness value of the th sparrow individual, is the amplitude of random perturbation, Figure 3 is a probability density function,

[0064] S302, obtain sparse key features, and the sparse key features are the features extracted by using the updated latest initial weights;

[0065] S304, initialize the population parameters in the way of equipment data stratification, that is, initialize the population individuals, and each individual can be understood as, but is not limited to, a maintenance plan in the first maintenance decision plan;

[0066] S306, calculate the fitness value of each individual, that is, calculate the initial fitness values of all individuals included in the elite layer and the ordinary layer;

[0067] S308. Determine multiple individuals with relatively better pheromone concentration matrices as discoverers;

[0068] S310. Update the positions of followers by means of adaptive neighborhood search;

[0069] S312. Select a part of them as guards in proportion, specifically: sort the fitness values corresponding to all individuals from high to low; determine multiple individuals with higher fitness values as guards according to a preset proportion, and then execute S312-1 to S312-5 to update the positions of the guards;

[0070] S312-1. Randomly select an unvisited target individual from the guards, that is, randomly select an unvisited guard individual from the guards as the target individual;

[0071] S312-2. Update the position of the target individual by using the reverse learning strategy to obtain the first updated position. It should be noted that the specific methods of using the reverse learning strategy and using Cauchy perturbation to update the position of the target individual in S312-3 can adopt existing methods, which are not limited in this application; it should be noted that when using the reverse learning strategy to update the position of the target individual, the first fitness value of the target individual at the first updated position will be calculated, and at the same time, the initial fitness value of the target individual will be obtained. If the first fitness value is greater than or equal to the initial fitness value, the position of the target individual will be updated by using the reverse learning strategy, and the updated position will be determined as the first updated position. If the first fitness value is less than the initial fitness value, the initial position of the target individual will be determined as the first updated position;

[0072] S312-3. Update the position of the updated target individual by using Cauchy perturbation to obtain the second updated position; it can be understood, but not limited to, that for the target individual whose position is updated by using the reverse learning strategy, Cauchy perturbation is used for secondary position update;

[0073] S312-4. Calculate the new fitness value (also called the second fitness value) of the target individual at the second updated position;

[0074] S312-5. Compare the new fitness value with the initial fitness value of the target individual (if the position is updated in S312-2, it is the first fitness value) to determine whether to update the position, specifically: if the new fitness value is greater than or equal to the initial fitness value, update the position of the target individual; if the new fitness value is less than the initial fitness value, do not update the position of the target individual.

[0075] It should be noted that the position updates in S312-2 and S312-3 are not real position updates. Instead, they are simulated position updates through the reverse learning strategy and Cauchy perturbation, aiming to obtain the first updated position and the second updated position when the assumption is updated. The real updates are all carried out after comparing the corresponding fitness values.

[0076] S314, determine whether the sentinels have been traversed. If not, execute S312-1; if so, execute S316.

[0077] S316, determine whether the iteration termination condition is met. If the iteration termination condition is met, end the iterative update; if not, continue the iteration and execute S306. The iteration termination conditions include the maximum number of iterations or the convergence of the fitness function. It should be noted that when the iteration termination condition is not met and S306 is continued for iteration, the operation of calculating the fitness value of each individual in S306 at this time is based on the fitness value calculated according to the updated position of each individual. After ending the iterative update, the initial maintenance decision plan included in the reference evaluation result and the initial evaluation result corresponding to the initial maintenance decision plan can be output.

[0078] Through the above implementation manners of the present application, dynamic sparse structure adjustment is performed. Hierarchical initialization, pheromone guidance (corresponding operations on the pheromone concentration matrix), dynamic neighborhood search, Cauchy compilation, and reverse learning are introduced into RSSA, and an environment-aware intelligent search framework is constructed. The present application deeply integrates the RSSA signal characteristics into the algorithm optimization process, realizing the full-link collaborative optimization from data preprocessing to search strategy. Focus on the mapping relationship between the pheromone concentration and environmental parameters, and adopt a dynamic adjustment strategy to further improve the robustness of the algorithm in complex scenarios.

[0079] As an optional implementation manner, inputting the analysis result set and the reference evaluation result into the target evaluation model to obtain the target evaluation result includes: S1, performing sparse feature extraction on multiple analysis results in the analysis result set and the reference evaluation result by using the reference weight to obtain the target sparse feature; S2, generating a second maintenance decision plan for the equipment to be evaluated according to the target sparse feature; S3, evaluating the second maintenance decision plan through the bionic immune optimization mechanism, and performing simulated maintenance on the equipment to be evaluated according to the evaluation result of the second maintenance decision plan to obtain the target simulated maintenance result; S4, adjusting the reference weight according to the target simulated maintenance result, and iteratively updating the second maintenance decision plan according to the target sparse feature extracted by using the updated reference weight to obtain the target maintenance decision plan included in the target evaluation result and the target evaluation result corresponding to the target maintenance decision plan.

[0080] It should be noted that the above initial weights are the feature extraction weights preset for the reference evaluation model, and the above reference weights are the feature extraction weights preset for the target evaluation model, and the initial weights are different from the reference weights. At the same time, the data input into the target evaluation model, in addition to the analysis result set and the reference evaluation result, also includes the maintenance target and the constraint conditions, and the operation in S2 is specifically to generate the second maintenance decision-making plan according to the target sparse features, the maintenance target and the constraint conditions.

[0081] It should be noted that the specific process of updating the second maintenance decision-making plan from S3 to S4 until finally obtaining the target maintenance plan included in the target evaluation result and the target evaluation result corresponding to the target maintenance plan is essentially the same as the specific process of updating the first maintenance decision-making plan until finally obtaining the initial maintenance decision-making plan included in the reference evaluation result and the initial evaluation result corresponding to the initial maintenance decision-making plan. Specifically, it can be understood by referring to the update process of the first maintenance decision-making plan. The reference evaluation model and the target evaluation model continuously adjust the feature extraction weights of SparseMLP according to the feedback data after the simulated maintenance execution. RSSA updates the maintenance plan through multiple iterations and finally finds the best maintenance strategy. In RSSA, the algorithm optimizes the model training process by introducing an implicit sparse structure. It reduces the computational complexity by automatically identifying and adjusting the important connections in the network while maintaining the model's expressive power. This method utilizes the sparsity of the parameters in the model, making the unimportant connections be automatically ignored during training, thereby improving the training speed and inference efficiency. SparseMLP, on the basis of the traditional multi-layer perceptron, combines sparsity and diversity, and reduces the number of model parameters by introducing a sparse connection strategy, thereby reducing the risk of overfitting. SparseMLP utilizes the connection sparsity between different layers, enabling each layer to adaptively select the most important features for processing, thus improving the model's performance on specific tasks. Combining these two technologies, RSSA and SparseMLP form an efficient deep learning framework. During the training process, RSSA provides a mechanism for dynamically adjusting the network sparse structure, enabling SparseMLP to more flexibly handle the characteristics of different data sets. This combination not only improves the efficiency of the model but also effectively enhances the generalization ability of the model, making it perform better in various application scenarios. Through this method, researchers can construct more competitive deep learning models, especially in applications with limited computing resources or high requirements for real-time performance.

[0082] Through the above implementation manners of the present application, after the reference evaluation model generates the reference evaluation result, the target evaluation model is introduced for secondary evaluation. Thus, due to the differences in the data mainly targeted by the two models and the initial preset weights (the initial weight corresponding to the reference evaluation model and the reference weight corresponding to the target evaluation model), the target evaluation model re-updates and evaluates the reference evaluation result, further improving the accuracy of the evaluation result. At the same time, the combination of RSSA and SparseMLP and its application to equipment maintenance ability evaluation can not only adaptively adjust the network structure to achieve higher computing efficiency and better performance, but also dynamically adjust the features and parameters used in the evaluation process, thereby improving the accuracy and efficiency of the evaluation and enabling real-time feedback. Equipment maintenance ability evaluation usually requires real-time analysis of a large amount of data. Especially during the operation of the equipment, real-time maintenance prediction and ability evaluation are crucial for reducing downtime. After combining ISSA and SparseMLP, the model can improve the real-time performance of the evaluation while maintaining high computing efficiency. The sparse connections and dynamic structure adjustment make the network more efficient during the computing process, enabling it to quickly provide accurate evaluation results and timely support for maintenance decisions. By reducing unnecessary computations and automatically optimizing the structure, the evaluation system can process sensor data and historical maintenance data more quickly, achieving real-time evaluation and dynamic adjustment.

[0083] As an optional implementation manner, after obtaining the target evaluation result, it further includes: S1, determining the set of post-maintenance indexes of the equipment to be evaluated according to the target evaluation result; S2, generating a first prompt message when the set of post-maintenance indexes meets the preset index condition, where the first prompt message is used to prompt to perform maintenance on the equipment to be evaluated using the target maintenance decision plan; S3, generating a second prompt message when the set of post-maintenance indexes does not meet the preset index condition, where the second prompt message is used to prompt the initial equipment information of the equipment to be evaluated and to prompt to replace the equipment to be evaluated.

[0084] The set of post-maintenance indexes in the above S1 includes: the service life of the equipment after maintenance (the time that the equipment to be evaluated can still be used normally assuming that the target maintenance decision plan in the target evaluation result is truly used to perform maintenance on the equipment to be evaluated), the failure probability of the equipment after maintenance (the probability that the equipment to be evaluated fails assuming that maintenance is truly performed), the cost required for maintenance, the time required for maintenance, etc.

[0085] Multiple post - repair indices in S2 respectively correspond to a preset index condition. That the set of post - repair indices in S2 meets the preset index condition can be understood, but is not limited to, that in the set of post - repair indices, the number of post - repair indices that meet the corresponding preset index condition is greater than the preset number. The preset index condition, preset number, etc. are all pre - set. The first prompt message can be understood, but is not limited to, that the plan for repairing the equipment to be evaluated is better than the plan for replacing a new equipment of the same type as the equipment to be evaluated. Therefore, it prompts to repair the equipment to be evaluated, while the second prompt message can be understood, but is not limited to, that the plan for replacing a new equipment of the same type as the equipment to be evaluated is better than the plan for repairing the equipment to be evaluated. Therefore, it prompts to replace the equipment to be evaluated, and at the same time, it prompts the initial equipment information (equipment model, size, etc.) to facilitate the user to accurately find the equipment with the same initial equipment information as the equipment to be evaluated when replacing the equipment to be evaluated.

[0086] Through the above - mentioned embodiments of the present application, after obtaining the target evaluation result, it further prompts the user for subsequent operations, improving the practicality of the target evaluation result in subsequent processing. That is, the solution in the present application is to evaluate the maintenance ability of unoperated equipment. The equipment maintenance ability evaluation in the present application not only indicates the difficulty of repairing the equipment, repair time, repair cost, etc., but also evaluates whether the equipment can still be used after repair. If it is evaluated that the equipment cannot be used after repair and can only be replaced with a new one, then the equipment will not be repaired. That is to say, after the equipment has a problem, this solution will conduct a comprehensive evaluation of the difficulty of repairing the equipment, repair time, etc., and the subsequent measure for the equipment is to repair or replace.

[0087] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0088] According to another aspect of the embodiments of the present invention, there is also provided an equipment maintenance ability evaluation device for implementing the above - mentioned equipment maintenance ability evaluation method, as Figure 4 shown. The device includes:

[0089] An acquisition unit 402 is configured to acquire historical data, reference operation data, and target operation data of the equipment to be evaluated, where the reference operation data is the most recent operation data of the equipment to be evaluated, and the target operation data is the initial equipment data when the equipment to be evaluated has not been operated; a first input unit 404 is configured to input the historical data into a data analysis model to obtain an analysis result set, where the analysis result set includes analysis results corresponding to various historical data and analysis views corresponding to each analysis result; a second input unit 406 is configured to input the reference operation data and the target operation data into a reference evaluation model to obtain a reference evaluation result, where the reference evaluation model is a model for initially evaluating the equipment maintenance ability of the equipment to be evaluated; a third input unit 408 is configured to input the analysis result set and the reference evaluation result into a target evaluation model to obtain a target evaluation result, where the target evaluation model is a model for finally evaluating the equipment maintenance ability of the equipment to be evaluated.

[0090] The specific manners of the operations performed by the units in the above device embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0091] According to another aspect of the embodiments of the present invention, an electronic device for implementing the above equipment maintenance ability evaluation method is further provided. The electronic device may be a Figure 5 terminal device or a server as shown. As Figure 5 shown, the electronic device includes: at least one processor 504; and a memory 502 communicatively connected to the at least one processor 504; where the memory 502 stores a computer program executable by the at least one processor 504, and the computer program is executed by the at least one processor 504 to enable the at least one processor 504 to execute the steps in any one of the above method embodiments. In this embodiment, the above electronic device may be at least one network device among multiple network devices in a computer network. The above processor may be configured to execute each step of the above equipment maintenance ability evaluation method through the computer program.

[0092] Optionally, those of ordinary skill in the art can understand that Figure 5 the structure shown is only schematic, and the electronic device may also be other terminal devices. Figure 5It does not limit the structure of the above-mentioned electronic device. The memory 502 can be used to store software programs and modules, such as the program instructions / modules corresponding to the equipment maintenance ability evaluation method and device in the embodiments of the present invention. The processor 504 executes various functional applications and data processing by running the software programs and modules stored in the memory 502, that is, implements the above-mentioned equipment maintenance ability evaluation method. The memory 502 may include a high-speed random access memory, and may also include a non-volatile memory. In some examples, the memory 502 may further include a memory remotely set relative to the processor 504, and these remote memories can be connected to the terminal through a network. The memory 502 can specifically but not limitedly be used to store various data information involved in this application. Such as Figure 5 As shown, the above-mentioned memory 502 may but not limitedly include the acquisition unit 402, the first input unit 404, the second input unit 406, and the third input unit 408 in the above-mentioned equipment maintenance ability evaluation device. In addition, it may also include but not limitedly other module units in the above-mentioned equipment maintenance ability evaluation device, which will not be elaborated in this example.

[0093] Optionally, the above-mentioned transmission device 506 is used to receive or send data via a network. In one example, the transmission device 506 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers through a network cable, so as to communicate with the Internet or a local area network. In one example, the transmission device 506 is a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet wirelessly. The above-mentioned electronic device further includes: a display 508, and a connection bus 510, which is used to connect each module component in the above-mentioned electronic device.

[0094] According to one aspect of the present application, a computer program product is provided. The computer program product includes computer programs / instructions, and the computer programs / instructions include program codes for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit, it executes various functions provided by the embodiments of the present application.

[0095] The serial numbers of the above-mentioned embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0096] According to one aspect of the present application, a computer-readable storage medium is provided. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to execute the above-mentioned equipment maintenance ability evaluation method. The computer-readable storage medium can be set to store a computer program for executing each step in the above-mentioned equipment maintenance ability evaluation method.

[0097] Those skilled in the art can understand that to implement all or part of the processes in the above method embodiments, it can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The above description is only a preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the art in the technical field, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. An equipment maintenance ability evaluation method, characterized in that Including: Obtain the historical data, reference operation data, and target operation data of the equipment to be evaluated. Among them, the reference operation data is the most recent operation data of the equipment to be evaluated, and the target operation data is the initial equipment data when the equipment to be evaluated has not been operated; Input the historical data into the data analysis model to obtain an analysis result set, where the analysis result set includes analysis results corresponding to various historical data, and analysis views corresponding to each analysis result; Input the reference operation data and the target operation data into the reference evaluation model to obtain a reference evaluation result, including: performing data preprocessing on the reference operation data to obtain preprocessed data; extracting sparse key features from the preprocessed data using the initial weight to obtain sparse key features; generating a first maintenance decision plan corresponding to the equipment to be evaluated based on the sparse key features and the target operation data; evaluating the first maintenance decision plan through a bionic immune optimization mechanism, and selecting and optimizing the first maintenance decision plan to obtain a reference evaluation result, where the reference evaluation model is a model for initially evaluating the equipment maintenance ability of the equipment to be evaluated; Input the analysis result set and the reference evaluation result into the target evaluation model to obtain a target evaluation result, including: extracting sparse key features from multiple analysis results in the analysis result set and the reference evaluation result using the reference weight to obtain target sparse features; generating a second maintenance decision plan corresponding to the equipment to be evaluated based on the target sparse features; evaluating the second maintenance decision plan through a bionic immune optimization mechanism, and performing simulated maintenance on the equipment to be evaluated according to the evaluation result of the second maintenance decision plan to obtain a target simulated maintenance result; adjusting the reference weight according to the target simulated maintenance result, and iteratively updating the second maintenance decision plan according to the target sparse features extracted using the updated reference weight to obtain the target maintenance decision plan included in the target evaluation result and the target evaluation result corresponding to the target maintenance decision plan, where the target evaluation model is a model for finally evaluating the equipment maintenance ability of the equipment to be evaluated.

2. The method according to claim 1, wherein Input the historical data into the data analysis model to obtain an analysis result set, including: Classify the historical data to obtain various reference historical data, and each reference historical data corresponds to a data category; Respectively perform data analysis on the corresponding reference historical data using the data analysis logic corresponding to the data category of each reference historical data to obtain the analysis result corresponding to each reference historical data; Respectively draw the analysis view corresponding to each reference historical data according to the analysis process of each reference historical data.

3. The method according to claim 1, wherein Evaluate the first maintenance decision plan through a bionic immune optimization mechanism, and optimize the first maintenance decision plan to obtain a reference evaluation result, including: Perform simulated maintenance on the equipment to be evaluated according to the first maintenance decision plan to obtain a simulated maintenance result; Adjust the initial weights according to the simulated maintenance results, and iteratively update the first maintenance decision plan based on the sparse key features extracted using the updated initial weights, to obtain the initial maintenance decision plan included in the reference evaluation result and the initial evaluation result corresponding to the initial maintenance decision plan.

4. The method according to claim 3, wherein Iteratively updating the first maintenance decision plan based on the sparse key features extracted using the updated initial weights includes: Initializing the population parameters of the sparse key features in a way of equipment data stratification to obtain an elite layer and a normal layer, where the elite layer is initialized using the spatial uniform distribution method, and the normal layer is initialized using the Gaussian perturbation method; Determine the initial fitness value and pheromone concentration matrix of each individual in the elite layer and the normal layer; Determine the discoverer, follower, and watcher according to the multiple initial fitness values and the multiple pheromone concentration matrices, and perform position updates.

5. The method according to claim 4, wherein Determining the discoverer, follower, and watcher according to the multiple initial fitness values and the multiple pheromone concentration matrices, and performing position updates includes: Determine the individuals corresponding to the multiple target pheromone concentration matrices that meet the preset conditions in the multiple pheromone concentration matrices as the discoverers, and update the positions of the corresponding discoverers according to the pheromone concentration matrix corresponding to each discoverer; Adopt a dynamic update strategy of adaptive neighborhood search to update the follower positions according to the multiple initial fitness values; Iteratively update the multiple initial fitness values using the Cauchy mutation and reverse learning strategies to obtain multiple target fitness values; Select a part of the individuals corresponding to the multiple target fitness values as watchers in proportion, and update the positions of the watchers using the Cauchy perturbation and reverse learning strategies.

6. The method according to claim 1, characterized in that, After obtaining the target evaluation result, it further includes: Determine the post-maintenance index set of the equipment to be evaluated according to the target evaluation result; Generate a first prompt message when the post-maintenance index set meets the preset index conditions, where the first prompt message is used to prompt to perform maintenance on the equipment to be evaluated using the target maintenance decision plan; Generate a second prompt message when the post-maintenance index set does not meet the preset index conditions, where the second prompt message is used to prompt the initial equipment information of the equipment to be evaluated, and to prompt to perform equipment replacement on the equipment to be evaluated.

7. An equipment maintenance ability evaluation device, characterized in that, Includes: An acquisition unit for acquiring the historical data, reference operation data, and target operation data of the equipment to be evaluated, where the reference operation data is the most recent operation data of the equipment to be evaluated, and the target operation data is the initial equipment data when the equipment to be evaluated has not been operated; A first input unit for inputting the historical data into a data analysis model to obtain an analysis result set, where the analysis result set includes the analysis results corresponding to various historical data, and the analysis view corresponding to each analysis result; A second input unit, configured to input the reference operation data and the target operation data into a reference evaluation model to obtain a reference evaluation result, including: performing data preprocessing on the reference operation data to obtain preprocessed data; extracting sparse key features from the preprocessed data by using initial weights to obtain sparse key features; generating a first maintenance decision plan corresponding to the equipment to be evaluated according to the sparse key features and the target operation data; evaluating the first maintenance decision plan through a bionic immune optimization mechanism, and selecting and optimizing the first maintenance decision plan to obtain a reference evaluation result, wherein the reference evaluation model is a model for initially evaluating the equipment maintenance ability of the equipment to be evaluated; A third input unit, configured to input the analysis result set and the reference evaluation result into a target evaluation model to obtain a target evaluation result, including: extracting sparse features from multiple analysis results in the analysis result set and the reference evaluation result by using reference weights to obtain target sparse features; generating a second maintenance decision plan corresponding to the equipment to be evaluated according to the target sparse features; evaluating the second maintenance decision plan through a bionic immune optimization mechanism, and performing simulated maintenance on the equipment to be evaluated according to the evaluation result of the second maintenance decision plan to obtain a target simulated maintenance result; adjusting the reference weights according to the target simulated maintenance result, and iteratively updating the second maintenance decision plan according to the target sparse features extracted by using the updated reference weights to obtain the target maintenance decision plan included in the target evaluation result and the target evaluation result corresponding to the target maintenance decision plan, wherein the target evaluation model is a model for finally evaluating the equipment maintenance ability of the equipment to be evaluated.

8. An electronic device, characterized in that, The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the equipment maintenance ability evaluation method according to any one of claims 1-6.

Images