Pumped storage equipment maintenance method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a pumped storage equipment maintenance method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring equipment data of pumped storage equipment; inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment; inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state comprises a target health degree and a target importance degree; according to the target health degree and the target importance degree, the control level of the pumped storage equipment is determined; querying a maintenance strategy library according to the management and control level to obtain a maintenance strategy corresponding to the management and control level, and taking the maintenance strategy as a maintenance strategy of the pumped storage equipment; and performing corresponding maintenance processing on the pumped storage equipment according to the maintenance strategy of the pumped storage equipment. By adopting the method, the maintenance effect of the pumped storage equipment can be improved.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for pumped storage equipment maintenance. Background Art

[0002] In the field of power grid technology, in order to ensure the normal operation of pumped storage equipment, it is necessary to conduct regular maintenance on the pumped storage equipment through relevant maintenance strategies.

[0003] In traditional technology, when pumped storage equipment is repaired, maintenance strategies are generally formulated based on manual experience; however, when formulating maintenance strategies, relying solely on manual experience can easily lead to limitations in the formulated maintenance strategies, resulting in poor maintenance effects on pumped storage equipment. Summary of the invention

[0004] Based on this, it is necessary to provide a pumped-storage equipment maintenance method, device, computer equipment, computer-readable storage medium and computer program product that can improve the maintenance effect of pumped-storage equipment in response to the above-mentioned technical problems.

[0005] In a first aspect, the present application provides a method for overhauling a pumped storage equipment, comprising:

[0006] Obtain equipment data for pumped storage equipment;

[0007] Inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment;

[0008] Inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes a target health degree and a target importance degree;

[0009] Determining a control level of the pumped-storage equipment according to the target health and the target importance;

[0010] Querying a maintenance strategy library according to the control level to obtain a maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment;

[0011] According to the maintenance strategy of the pumped-storage equipment, the pumped-storage equipment is subjected to corresponding maintenance processing.

[0012] In one of the embodiments, the pre-trained fault analysis model includes a fault mode analysis model, a fault impact analysis model and a maintenance mode analysis model;

[0013] The step of inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment comprises:

[0014] Filtering out, from the equipment data, first equipment data related to failure mode analysis, second equipment data related to failure impact analysis, and third equipment data related to maintenance mode analysis;

[0015] Inputting the first equipment data into the failure mode analysis model to obtain a failure mode analysis result of the pumped storage equipment;

[0016] Inputting the second equipment data into the fault impact analysis model to obtain a fault impact analysis result of the pumped storage equipment;

[0017] Inputting the third equipment data into the maintenance mode analysis model to obtain a maintenance mode analysis result of the pumped-storage equipment;

[0018] The fault mode analysis result, the fault impact analysis result and the maintenance mode analysis result are fused to obtain the fault analysis result of the pumped storage equipment.

[0019] In one of the embodiments, the pre-trained device status recognition model includes a device health recognition model and a device importance recognition model;

[0020] The step of inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment comprises:

[0021] From the fault analysis results, screen out a first fault analysis result related to health analysis and a second fault analysis result related to importance analysis;

[0022] Performing feature extraction processing on the first fault analysis result to obtain a first feature vector corresponding to the first fault analysis result, and performing feature extraction processing on the second fault analysis result to obtain a second feature vector corresponding to the second fault analysis result;

[0023] Inputting the first feature vector into the equipment health recognition model to obtain the target health of the pumped-storage equipment, and inputting the second feature vector into the equipment importance recognition model to obtain the target importance of the pumped-storage equipment;

[0024] The target health and target importance are combined to obtain the equipment state of the pumped-storage equipment.

[0025] In one embodiment, the step of inputting the first feature vector into the equipment health recognition model to obtain the target health of the pumped-storage equipment includes:

[0026] Inputting the first characteristic vector into the equipment health recognition model to obtain a state quantity evaluation result of each characteristic state quantity of the pumped-storage equipment;

[0027] Filtering out the state quantity evaluation results of key characteristic state quantities from the state quantity evaluation results of the characteristic state quantities;

[0028] According to the weight of the key characteristic state quantity, the state quantity evaluation result of the key characteristic state quantity is fused to obtain the target evaluation result of the pumped storage equipment;

[0029] A health degree corresponding to the target evaluation result is obtained as a target health degree of the pumped-storage equipment.

[0030] In one embodiment, the step of inputting the second feature vector into the equipment importance recognition model to obtain the target importance of the pumped-storage equipment includes:

[0031] Inputting the second feature vector into the equipment importance identification model to obtain the event level of the pumped-storage equipment under each preset safety impact dimension;

[0032] Filtering out the event levels of the pumped-storage equipment under the key safety impact dimensions from the event levels of the pumped-storage equipment under each preset safety impact dimension;

[0033] From the event levels of the pumped storage equipment under the key safety impact dimension, the highest event level is selected as the target event level;

[0034] The importance corresponding to the target event level is obtained as the target importance of the pumped-storage equipment.

[0035] In one embodiment, determining the control level of the pumped storage equipment according to the target health and the target importance includes:

[0036] Obtaining an equipment risk matrix; the equipment risk matrix is ​​used to represent the corresponding relationship between health, importance and control level;

[0037] The equipment risk matrix is ​​queried to obtain a control level corresponding to the target health and the target importance as the control level of the pumped-storage equipment.

[0038] In a second aspect, the present application also provides a pumped storage equipment maintenance device, comprising:

[0039] A data acquisition module, used to acquire equipment data of the pumped storage equipment;

[0040] A fault analysis module, used for inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment;

[0041] A state recognition module, used for inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes a target health degree and a target importance degree;

[0042] A level determination module, used to determine the control level of the pumped storage equipment according to the target health and the target importance;

[0043] A strategy query module, used to query the maintenance strategy library according to the control level, and obtain the maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment;

[0044] The equipment maintenance module is used to perform corresponding maintenance processing on the pumped-storage equipment according to the maintenance strategy of the pumped-storage equipment.

[0045] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0046] Obtain equipment data for pumped storage equipment;

[0047] Inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment;

[0048] Inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes a target health degree and a target importance degree;

[0049] Determining a control level of the pumped-storage equipment according to the target health and the target importance;

[0050] Querying a maintenance strategy library according to the control level to obtain a maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment;

[0051] According to the maintenance strategy of the pumped-storage equipment, the pumped-storage equipment is subjected to corresponding maintenance processing.

[0052] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0053] Obtain equipment data for pumped storage equipment;

[0054] Inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment;

[0055] Inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes a target health degree and a target importance degree;

[0056] Determining a control level of the pumped-storage equipment according to the target health and the target importance;

[0057] Querying a maintenance strategy library according to the control level to obtain a maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment;

[0058] According to the maintenance strategy of the pumped-storage equipment, the pumped-storage equipment is subjected to corresponding maintenance processing.

[0059] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0060] Obtain equipment data for pumped storage equipment;

[0061] Inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment;

[0062] Inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes a target health degree and a target importance degree;

[0063] Determining a control level of the pumped-storage equipment according to the target health and the target importance;

[0064] Querying a maintenance strategy library according to the control level to obtain a maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment;

[0065] According to the maintenance strategy of the pumped-storage equipment, the pumped-storage equipment is subjected to corresponding maintenance processing.

[0066] The above-mentioned pumped-storage equipment maintenance method, device, computer equipment, computer-readable storage medium and computer program product first obtain the equipment data of the pumped-storage equipment, and input the equipment data into a pre-trained fault analysis model to obtain the fault analysis result of the pumped-storage equipment, and then input the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped-storage equipment; the equipment state includes a target health and a target importance; then, according to the target health and the target importance, the control level of the pumped-storage equipment is determined, and the maintenance strategy library is queried according to the control level to obtain the maintenance strategy corresponding to the control level as the maintenance strategy of the pumped-storage equipment, and finally, according to the maintenance strategy of the pumped-storage equipment, the pumped-storage equipment is subjected to corresponding maintenance processing. In this way, when the pumped-storage equipment is inspected and maintained, the equipment data, fault analysis results, equipment status (i.e., target health and target importance) and control level of the pumped-storage equipment are comprehensively considered, and the maintenance strategy of the pumped-storage equipment is determined based on this. This is conducive to improving the accuracy of determining the maintenance strategy, thereby facilitating accurate maintenance of the pumped-storage equipment, thereby improving the maintenance effect of the pumped-storage equipment, and at the same time avoiding the limitations of the maintenance strategy formulated manually, which leads to the defect of poor maintenance effect on the pumped-storage equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 A schematic diagram of a process flow of a method for repairing pumped storage equipment in one embodiment;

[0069] Figure 2 is a schematic diagram of a typical fault tree in an embodiment;

[0070] Figure 3 A flowchart of RCM (Reliability Centered Maintenance) analysis steps in one embodiment;

[0071] Figure 4 A flowchart of the steps of inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment in one embodiment;

[0072] Figure 5 A schematic diagram of a process flow of a device status evaluation step in an embodiment;

[0073] Figure 6 is a schematic diagram of a device risk matrix in one embodiment;

[0074] Figure 7 A schematic flow chart of a method for repairing pumped storage equipment in another embodiment;

[0075] Figure 8 A flowchart of a maintenance strategy processing method based on RCM in one embodiment;

[0076] Fig. 9 A structural block diagram of a pumped storage equipment maintenance device in one embodiment;

[0077] Fig.10 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0079] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0080] Figure 1 is a flow chart of a method for repairing a pumped storage device according to an exemplary embodiment. Figure 1 As shown, the pumped storage equipment maintenance method is applied to the server; it is understandable that the method can also be applied to the terminal, and can also be applied to the system including the terminal and the server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be but not limited to various personal computers, laptops, smart phones and tablets; the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In this exemplary embodiment, the method includes the following steps S101 to S106. Among them:

[0081] Step S101, obtaining equipment data of the pumped storage equipment.

[0082] Among them, pumped storage equipment is a comprehensive system, mainly including water pump turbines, generator motors and a series of auxiliary equipment and control systems.

[0083] Among them, equipment data refers to data related to pumped storage equipment, such as current equipment parameters, historical equipment parameters, historical maintenance information, etc.

[0084] Exemplarily, in response to a maintenance request for a pumped-storage equipment, the server obtains the initial equipment data of the pumped-storage equipment from a local database storing initial equipment data of multiple pumped-storage equipment, and preprocesses the initial equipment data, such as converting the format, to obtain preprocessed equipment data as the equipment data of the pumped-storage equipment.

[0085] Step S102, inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment.

[0086] Among them, the pre-trained fault analysis model refers to a model used to output fault analysis results of pumped storage equipment, such as a neural network model, a deep learning model, etc.

[0087] Among them, the fault analysis results refer to the analysis results related to the pumped storage equipment fault, including at least the fault mode analysis results, the fault impact analysis results, and the maintenance mode analysis results.

[0088] Exemplarily, the server inputs each device data into a pre-trained weight prediction model (such as a neural network model, an attention mechanism model, etc.) to obtain the weight corresponding to each device data; from each device data, the device data whose corresponding weight is greater than the preset weight is screened out as the key device data; then, the key device data is input into the pre-trained fault analysis model, and the fault analysis model is used to perform fault analysis on the key device data to obtain the fault analysis results of the pumped storage equipment.

[0089] Furthermore, the pre-trained fault analysis model is trained in the following manner: the server obtains sample equipment data of the sample pumped-storage equipment, and inputs the sample equipment data of the sample pumped-storage equipment into the fault analysis model to be trained, to obtain the predicted fault analysis results of the sample pumped-storage equipment; the actual fault analysis results of the sample pumped-storage equipment are obtained, and based on the difference between the predicted fault analysis results and the actual fault analysis results of the sample pumped-storage equipment, the fault analysis model to be trained is iteratively trained to obtain a trained fault analysis model as the pre-trained fault analysis model.

[0090] Step S103, inputting the fault analysis result into a pre-trained equipment status recognition model to obtain the equipment status of the pumped storage equipment; the equipment status includes a target health degree and a target importance degree.

[0091] Among them, the pre-trained equipment state recognition model refers to a model used to output the equipment state of the pumped-storage equipment, such as a neural network model, a deep learning model, etc.

[0092] Among them, the equipment status is used to characterize the status information of the pumped storage equipment, specifically including the target health and target importance.

[0093] Among them, the target health is used to characterize the equipment health level of the pumped storage equipment, including normal, caution, abnormal, and severe.

[0094] Among them, the target importance is used to characterize the equipment importance level of pumped storage equipment, including general, attention, important, and critical.

[0095] Exemplarily, the server performs feature extraction processing on the fault analysis result to obtain an initial feature vector of the fault analysis result (such as a shallow feature vector), performs feature extraction processing on the initial feature vector again to obtain an intermediate feature vector of the fault analysis result (such as a deep feature vector), and fuses the initial feature vector and the intermediate feature vector to obtain a target feature vector of the fault analysis result; then, the target feature vector is input into a pre-trained equipment state recognition model, and the target feature vector is subjected to health recognition processing through the equipment state recognition model to obtain the predicted health probability of the pumped-storage equipment at different health levels, and the health level with the largest corresponding health prediction probability is selected from the different health levels as the target health level of the pumped-storage equipment; at the same time, the target feature vector is subjected to importance recognition processing through the equipment state recognition model to obtain the predicted importance probability of the pumped-storage equipment at different importance levels, and the importance level with the largest corresponding importance prediction probability is selected from the different importance levels as the target importance level of the pumped-storage equipment; finally, the target health level and the target importance are combined to obtain the equipment state of the pumped-storage equipment.

[0096] Furthermore, the pre-trained equipment state recognition model is trained in the following manner: the server obtains the sample fault analysis results of the sample pumped-storage equipment, and inputs the sample fault analysis results of the sample pumped-storage equipment into the equipment state recognition model to be trained to obtain the predicted equipment state of the sample pumped-storage equipment; the actual equipment state of the sample pumped-storage equipment is obtained, and according to the difference between the predicted equipment state and the actual equipment state of the sample pumped-storage equipment, the equipment state recognition model to be trained is iteratively trained to obtain the trained equipment state recognition model as the pre-trained equipment state recognition model.

[0097] Step S104: determining the management and control level of the pumped-storage equipment according to the target health and target importance.

[0098] Among them, the control level is used to characterize the equipment level of pumped storage equipment, which can be divided into Level I, Level II, Level III and Level IV from high to low, and is determined comprehensively based on the target health and target importance; for example, different health and importance correspond to different control levels.

[0099] Exemplarily, the server obtains the correspondence between health, importance and control level, and queries the correspondence between health, importance and control level based on the target health and target importance, and obtains the control level corresponding to the target health and target importance as the control level of the pumped storage equipment.

[0100] Step S105, querying the maintenance strategy library according to the control level, obtaining the maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment.

[0101] Among them, maintenance strategy refers to the strategy for maintenance of pumped storage equipment, including maintenance type, maintenance items, maintenance time, maintenance cycle, maintenance requirements, etc.

[0102] Among them, the maintenance strategy library stores maintenance strategies corresponding to multiple control levels.

[0103] Among them, different control levels correspond to different maintenance strategies; for example, for the same pumped storage equipment, if the control level is level I, the first maintenance strategy is selected; if the control level is level II, the second maintenance strategy is selected.

[0104] Exemplarily, the server generates a maintenance strategy query request carrying the control level according to the control level, and sends the maintenance strategy query request to the maintenance strategy library to obtain the maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment.

[0105] For example, the principles of differentiated maintenance strategies for pumped-storage equipment at different control levels are as follows: a) Level I control equipment: shorten the maintenance cycle and carry out maintenance in a timely manner (within 1 year); b) Level II and III control equipment: carry out maintenance according to the maintenance cycle determined by RCM analysis, that is, the benchmark cycle; c) Level IV control equipment: for equipment evaluated as normal, the maintenance cycle can be appropriately extended according to the law of equipment failure, and the extension time shall not exceed its benchmark cycle.

[0106] Step S106: performing corresponding maintenance processing on the pumped-storage equipment according to the maintenance strategy of the pumped-storage equipment.

[0107] Exemplarily, the server generates a maintenance instruction for the pumped-storage equipment according to the maintenance strategy of the pumped-storage equipment, and sends the maintenance instruction for the pumped-storage equipment to the maintenance equipment; the maintenance equipment performs corresponding maintenance processing on the pumped-storage equipment according to the maintenance instruction.

[0108] Furthermore, after the pumped storage equipment is overhauled, the maintenance equipment collects multiple indicator data of the pumped storage equipment after the overhaul, such as reliability index, maintenance cost, equipment status, etc., and sends the multiple indicator data to the server. The server predicts the actual effect of the pumped storage equipment maintenance based on the multiple indicator data, and iteratively improves the maintenance strategy of the pumped storage equipment based on the actual effect of the pumped storage equipment maintenance.

[0109] In the above-mentioned pumped-storage equipment maintenance method, the equipment data of the pumped-storage equipment is first obtained, and the equipment data is input into a pre-trained fault analysis model to obtain a fault analysis result of the pumped-storage equipment, and then the fault analysis result is input into a pre-trained equipment state recognition model to obtain the equipment state of the pumped-storage equipment; the equipment state includes a target health and a target importance; then, according to the target health and the target importance, the control level of the pumped-storage equipment is determined, and according to the control level, a maintenance strategy library is queried to obtain a maintenance strategy corresponding to the control level as the maintenance strategy of the pumped-storage equipment, and finally, according to the maintenance strategy of the pumped-storage equipment, the pumped-storage equipment is subjected to corresponding maintenance processing. In this way, when the pumped-storage equipment is inspected and maintained, the equipment data, fault analysis results, equipment status (i.e., target health and target importance) and control level of the pumped-storage equipment are comprehensively considered, and the maintenance strategy of the pumped-storage equipment is determined based on this. This is conducive to improving the accuracy of determining the maintenance strategy, thereby facilitating accurate maintenance of the pumped-storage equipment, thereby improving the maintenance effect of the pumped-storage equipment, and at the same time avoiding the limitations of the maintenance strategy formulated manually, which leads to the defect of poor maintenance effect on the pumped-storage equipment.

[0110] In an exemplary embodiment, the pre-trained fault analysis model includes a fault mode analysis model, a fault impact analysis model and a maintenance mode analysis model; then, the above step S102, inputting the equipment data into the pre-trained fault analysis model to obtain the fault analysis result of the pumped-storage equipment, specifically includes the following contents: from the equipment data, screening out the first equipment data related to the fault mode analysis, the second equipment data related to the fault impact analysis, and the third equipment data related to the maintenance mode analysis; inputting the first equipment data into the fault mode analysis model to obtain the fault mode analysis result of the pumped-storage equipment; inputting the second equipment data into the fault impact analysis model to obtain the fault impact analysis result of the pumped-storage equipment; inputting the third equipment data into the maintenance mode analysis model to obtain the maintenance mode analysis result of the pumped-storage equipment; the fault mode analysis result, the fault impact analysis result and the maintenance mode analysis result are fused to obtain the fault analysis result of the pumped-storage equipment.

[0111] Among them, the pre-trained fault analysis model includes three models, namely, the fault mode analysis model, the fault impact analysis model and the maintenance mode analysis model.

[0112] The first equipment data refers to equipment data related to fault mode analysis; the second equipment data refers to equipment data related to fault impact analysis; and the third equipment data refers to equipment data related to maintenance mode analysis.

[0113] Among them, the failure mode analysis results are used to characterize the failure modes of pumped storage equipment, such as cracks, shaft collar scratches, rust, deformation, bolt loosening, seal aging, etc.

[0114] Among them, the failure mode analysis model is a model used to output the failure mode analysis results of pumped storage equipment, which can be a neural network model, a deep learning model, a large model, etc. In actual scenarios, the failure mode analysis model refers to a fault tree model, such as Figure 2 The typical fault tree shown in Figure 1 is used as an example to illustrate that a typical fault tree model is established according to the three-level division of equipment, components and elements. Based on the three-level fault tree (component / element name + root cause + fault characteristics + fault mechanism), all events of functional failure of the equipment are analyzed. A total of 8 major components, 44 components and 200 failure modes of the pump-turbine system of the pumped storage unit are analyzed, and 13 major components, 69 components and 503 failure modes of the generator motor system are analyzed.

[0115] Among them, the fault impact analysis results are used to characterize the fault impact results of pumped storage equipment, such as personal safety impact results, power grid safety impact results, equipment safety impact results, etc.

[0116] Among them, the fault impact analysis model is a model used to output the fault impact analysis results of pumped storage equipment, which can be a neural network model, a deep learning model, a large model, etc. In actual scenarios, the fault impact analysis determines the impact of equipment failure on personal, power grid, and equipment risks from three levels: components, parts, and equipment. After analysis, there are 4 types of failure modes that cause personal impact in pumped storage units (1 general accident and 3 level 4 events); 130 types of power grid impacts (6 level 1 events, 22 level 2 events, 10 level 3 events, 24 level 4 events, 60 level 5 events, and 8 level 6 events); 27 types of equipment impacts (1 level 1 event, 13 level 2 events, 2 level 3 events, 5 level 4 events, and 6 level 6 events).

[0117] Among them, the maintenance mode analysis results are used to characterize the maintenance mode of pumped storage equipment, including maintenance task type, maintenance items, maintenance cycle, etc.

[0118] Among them, the maintenance mode analysis model is a model used to output the maintenance mode analysis results of pumped storage equipment, which can be a neural network model, a deep learning model, a large model, etc. In actual scenarios, for each fault mode, select the appropriate maintenance task type, maintenance item, and maintenance reference cycle. 49 fault modes of pumped storage units adopt condition maintenance, 574 fault modes adopt periodic maintenance, and 80 fault modes adopt post-maintenance.

[0119] Exemplarily, according to the data filtering instruction, the server filters out the equipment data related to the fault mode analysis from the equipment data as the first equipment data; filters out the equipment data related to the fault impact analysis from the equipment data as the second equipment data; filters out the equipment data related to the maintenance mode analysis from the equipment data as the third equipment data; performs feature extraction processing on the first equipment data, the second equipment data and the third equipment data respectively to obtain the feature vector corresponding to the first equipment data, the feature vector corresponding to the second equipment data and the feature vector corresponding to the third equipment data; then, inputs the feature vector corresponding to the first equipment data into the fault mode analysis model for fault mode analysis processing to obtain the fault mode analysis result of the pumped-storage equipment; inputs the feature vector corresponding to the second equipment data into the fault impact analysis model for fault impact analysis processing to obtain the fault impact analysis result of the pumped-storage equipment; inputs the feature vector corresponding to the third equipment data into the maintenance mode analysis model for maintenance mode analysis processing to obtain the maintenance mode analysis result of the pumped-storage equipment; finally, the fault mode analysis result, the fault impact analysis result and the maintenance mode analysis result are combined to obtain the fault analysis result of the pumped-storage equipment.

[0120] For example, see Figure 3 The server performs fault mode analysis on the pumped-storage equipment to obtain the fault mode of the pumped-storage equipment; performs fault impact analysis on the pumped-storage equipment to obtain fault impact analysis results of the pumped-storage equipment, such as personal safety impact results, power grid safety impact results, equipment safety impact results, etc.; performs maintenance mode analysis on the pumped-storage equipment to obtain maintenance mode analysis results of the pumped-storage equipment, such as task type, maintenance item, maintenance cycle, etc.

[0121] In this embodiment, multiple analysis models are used to perform targeted analysis and processing on the pumped-storage equipment, and the obtained fault mode analysis results, fault impact analysis results and maintenance mode analysis results are combined to obtain the fault analysis results of the pumped-storage equipment, which is beneficial to improving the accuracy of determining the fault analysis results.

[0122] In an exemplary embodiment, Figure 4As shown, the pre-trained equipment status recognition model includes an equipment health recognition model and an equipment importance recognition model; then, the above step S103, inputting the fault analysis result into the pre-trained equipment status recognition model to obtain the equipment status of the pumped storage equipment, specifically includes the following steps S401 to S404. Among them:

[0123] Step S401 , screening out, from the fault analysis results, a first fault analysis result related to health analysis and a second fault analysis result related to importance analysis.

[0124] Step S402: performing feature extraction processing on the first fault analysis result to obtain a first feature vector corresponding to the first fault analysis result, and performing feature extraction processing on the second fault analysis result to obtain a second feature vector corresponding to the second fault analysis result.

[0125] Step S403, inputting the first eigenvector into the equipment health recognition model to obtain the target health of the pumped-storage equipment, and inputting the second eigenvector into the equipment importance recognition model to obtain the target importance of the pumped-storage equipment.

[0126] Step S404: Combine the target health and the target importance to obtain the equipment status of the pumped storage equipment.

[0127] The pre-trained device status recognition model includes a device health recognition model and a device importance recognition model.

[0128] The equipment health recognition model is a model for outputting the health of the pumped storage equipment, which may refer to a neural network model, a deep learning model, etc. The equipment importance recognition model is a model for outputting the importance of the pumped storage equipment, which may refer to a neural network model, a deep learning model, etc.

[0129] Among them, the first fault analysis result refers to the fault analysis result related to the health analysis, specifically refers to the fault mode analysis result, such as the operation inspection, maintenance, overhaul, preventive test, power-on test and other data involved in the fault mode analysis result.

[0130] Among them, the second fault analysis result refers to the fault analysis result related to the importance analysis, specifically the fault impact analysis result, such as personal safety impact result, power grid safety impact result, equipment safety impact result, etc.

[0131] Exemplarily, according to the data identification instruction, the server identifies the fault analysis result related to the health analysis from the fault analysis results as the first fault analysis result, and identifies the fault analysis result related to the importance analysis from the fault analysis results as the second fault analysis result; then, the first fault analysis result is subjected to feature extraction processing through the feature extraction model corresponding to the first fault analysis result to obtain the corresponding feature vector as the first feature vector corresponding to the first fault analysis result; at the same time, the second fault analysis result is subjected to feature extraction processing through the feature extraction model corresponding to the second fault analysis result to obtain the corresponding feature vector as the second feature vector corresponding to the second fault analysis result; then, the first feature vector is input into the equipment health identification model for health identification processing to obtain the target health of the pumped-storage equipment, and the second feature vector is input into the equipment importance identification model for importance identification processing to obtain the target importance of the pumped-storage equipment; finally, the target health and the target importance are combined to obtain the equipment status of the pumped-storage equipment.

[0132] For example, see Figure 5 , the server evaluates the health and importance of the pumped storage equipment, and obtains the health evaluation level and importance evaluation level of the pumped storage equipment; among them, the health evaluation is based on the data of operation inspection, maintenance, overhaul, preventive test and live test (online monitoring) involved in the failure mode, and scores the characteristic state quantities reflecting the health of the equipment in the failure mode, so as to determine the health level of the equipment, such as normal, attention, abnormal, and serious. In addition, the importance evaluation configuration management rules for each type of equipment are established, and the impact of the failure on power personnel, power safety, pumped storage equipment, etc. that may be caused by the equipment failure is evaluated. Finally, the highest level of the failure impact is taken as the importance level of the equipment, such as general, attention, important, and critical.

[0133] In this embodiment, the target health of the pumped-storage equipment is determined through the equipment health recognition model, and the target importance of the pumped-storage equipment is determined through the equipment importance recognition model. Finally, the target health and the target importance are combined to obtain the equipment status of the pumped-storage equipment. This can make the final equipment status more accurate, thereby improving the accuracy of determining the equipment status.

[0134] In an exemplary embodiment, the above step S403, inputting the first eigenvector into the equipment health recognition model to obtain the target health of the pumped-storage equipment, specifically includes the following contents: inputting the first eigenvector into the equipment health recognition model to obtain the state quantity evaluation results of each characteristic state quantity of the pumped-storage equipment; screening out the state quantity evaluation results of key characteristic state quantities from the state quantity evaluation results of each characteristic state quantity; fusing the state quantity evaluation results of the key characteristic state quantities according to the weights of the key characteristic state quantities to obtain the target evaluation results of the pumped-storage equipment; obtaining the health corresponding to the target evaluation result as the target health of the pumped-storage equipment.

[0135] Among them, characteristic state quantities refer to state quantities that reflect the health status of pumped storage equipment, such as corrosion inspection, nondestructive testing, cavitation depth, etc.

[0136] The state quantity evaluation result refers to the state quantity score.

[0137] Among them, the key characteristic state quantity refers to the characteristic state quantity whose corresponding weight is greater than the preset weight.

[0138] Among them, the target evaluation result of the pumped-storage equipment refers to the health score of the pumped-storage equipment; different health scores correspond to different health levels, for example, a higher health score corresponds to a normal health level.

[0139] Exemplarily, the server inputs the first feature vector into the equipment health identification model, processes the first feature vector through the equipment health identification model, and obtains the state quantity evaluation results and weights of each characteristic state quantity of the pumped-storage equipment; then, from each characteristic state quantity, the characteristic state quantity whose corresponding weight is greater than the preset weight is screened out as the key characteristic state quantity; then, according to the weight of the key characteristic state quantity, the state quantity evaluation results of the key characteristic state quantity are summed to obtain the target evaluation result of the pumped-storage equipment, for example, the target evaluation result = the weight of the key characteristic state quantity A1 × the state quantity evaluation result of the key characteristic state quantity A1 + the weight of the key characteristic state quantity A2 × the state quantity evaluation result of the key characteristic state quantity A2 + the weight of the key characteristic state quantity A3 × the state quantity evaluation result of the key characteristic state quantity A3; finally, according to the target evaluation result, the correspondence between the evaluation result and the health degree is queried to obtain the health degree corresponding to the target evaluation result as the target health degree of the pumped-storage equipment.

[0140] In this embodiment, when determining the target health of the pumped-storage equipment, the weights of the key characteristic state quantities and the state quantity evaluation results of the key characteristic state quantities are comprehensively considered, which is conducive to improving the accuracy of determining the target health.

[0141] In an exemplary embodiment, the above step S403, inputting the second eigenvector into the equipment importance identification model to obtain the target importance of the pumped-storage equipment, specifically includes the following contents: inputting the second eigenvector into the equipment importance identification model to obtain the event level of the pumped-storage equipment under each preset safety impact dimension; from the event levels of the pumped-storage equipment under each preset safety impact dimension, screening out the event level of the pumped-storage equipment under the key safety impact dimension; from the event levels of the pumped-storage equipment under the key safety impact dimension, screening out the highest level of event level as the target event level; obtaining the importance corresponding to the target event level as the target importance of the pumped-storage equipment.

[0142] Among them, the preset safety impact dimensions refer to personal safety impact dimensions, power grid safety impact dimensions, and equipment safety impact dimensions.

[0143] Among them, the event level of pumped-storage equipment under the dimension of impact on personal safety specifically refers to the highest event level among multiple event levels of pumped-storage equipment under the dimension of impact on personal safety.

[0144] Among them, the event level of pumped-storage equipment under the grid security impact dimension specifically refers to the highest event level among multiple event levels of pumped-storage equipment under the grid security impact dimension.

[0145] Among them, the event level of the pumped-storage equipment under the equipment safety impact dimension specifically refers to the highest event level among multiple event levels of the pumped-storage equipment under the equipment safety impact dimension.

[0146] Among them, the key security impact dimension refers to the preset security impact dimension whose corresponding weight is greater than the preset weight.

[0147] Among them, the target event level refers to the highest event level among the event levels of pumped storage equipment under the key safety impact dimension.

[0148] Among them, different target event levels correspond to different importances. For example, a target event with a higher level has a corresponding importance of critical.

[0149] Exemplarily, the server inputs the second eigenvector into the equipment importance identification model, processes the second eigenvector through the equipment importance identification model, and obtains the event level of the pumped-storage equipment under each preset safety impact dimension, and the weight corresponding to each preset safety impact dimension; then, from each preset safety impact dimension, the preset safety impact dimension whose corresponding weight is greater than the preset weight is screened out as the key safety impact dimension; then, from the event level of the pumped-storage equipment under the key safety impact dimension, the highest level of event level is screened out as the target event level; finally, according to the target event level, the correspondence between the event level and the importance is queried to obtain the importance corresponding to the target event level as the target importance of the pumped-storage equipment.

[0150] In this embodiment, when determining the target importance of the pumped-storage equipment, the target event level in the event level of the pumped-storage equipment under the key safety impact dimension is comprehensively considered, which is conducive to improving the accuracy of determining the target importance.

[0151] In an exemplary embodiment, the above step S104 determines the control level of the pumped-storage equipment according to the target health and target importance, and specifically includes the following contents: obtaining the equipment risk matrix; the equipment risk matrix is ​​used to represent the correspondence between health, importance and control level; querying the equipment risk matrix to obtain the control level corresponding to the target health and target importance as the control level of the pumped-storage equipment.

[0152] The rows of the equipment risk matrix represent importance, and the columns of the equipment risk matrix represent health. Figure 6 .

[0153] The corresponding relationship between health, importance and control level is used to indicate that different health and importance correspond to different control levels. Figure 6 , assuming that the health level is serious and the importance is critical, the corresponding control level is Level I.

[0154] For example, the server obtains Figure 6 The equipment risk matrix shown in the figure is queried according to the target health and target importance to obtain the control level corresponding to the target health and target importance as the control level of the pumped storage equipment.

[0155] In this embodiment, when determining the control level of the pumped-storage equipment, the target health and target importance of the pumped-storage equipment are comprehensively considered, which is conducive to improving the accuracy of determining the control level; at the same time, by querying the equipment risk matrix, it is conducive to quickly determining the control level of the pumped-storage equipment.

[0156] In an exemplary embodiment, Figure 7 As shown, another pumped storage equipment maintenance method is provided, which is described by taking the method applied to a server as an example, and includes the following steps S701 to S712. Among them:

[0157] Step S701, obtaining equipment data of the pumped storage equipment.

[0158] Step S702: Filter out first equipment data related to fault mode analysis, second equipment data related to fault impact analysis, and third equipment data related to maintenance mode analysis from the equipment data.

[0159] Step S703, input the first equipment data into the fault mode analysis model to obtain the fault mode analysis result of the pumped-storage equipment; input the second equipment data into the fault impact analysis model to obtain the fault impact analysis result of the pumped-storage equipment; input the third equipment data into the maintenance mode analysis model to obtain the maintenance mode analysis result of the pumped-storage equipment.

[0160] Step S704, fusing the fault mode analysis result, the fault impact analysis result and the maintenance mode analysis result to obtain the fault analysis result of the pumped storage equipment.

[0161] Step S705 , screening out the first fault analysis result related to health analysis and the second fault analysis result related to importance analysis from the fault analysis results.

[0162] Step S706 , performing feature extraction processing on the first fault analysis result to obtain a first feature vector corresponding to the first fault analysis result, and performing feature extraction processing on the second fault analysis result to obtain a second feature vector corresponding to the second fault analysis result.

[0163] Step S707, inputting the first eigenvector into the equipment health recognition model to obtain the target health of the pumped-storage equipment, and inputting the second eigenvector into the equipment importance recognition model to obtain the target importance of the pumped-storage equipment.

[0164] Step S708: Combine the target health and the target importance to obtain the equipment status of the pumped storage equipment.

[0165] Step S709, obtaining the equipment risk matrix; the equipment risk matrix is ​​used to represent the corresponding relationship between health, importance and control level.

[0166] Step S710, query the equipment risk matrix to obtain the control level corresponding to the target health and target importance as the control level of the pumped storage equipment.

[0167] Step S711, query the maintenance strategy library according to the control level, and obtain the maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment.

[0168] Step S712: Perform corresponding maintenance processing on the pumped-storage equipment according to the maintenance strategy of the pumped-storage equipment.

[0169] In the above-mentioned pumped-storage equipment maintenance method, when the pumped-storage equipment is being maintained, the equipment data, fault analysis results, equipment status (i.e., target health and target importance) and control level of the pumped-storage equipment are comprehensively considered, and the maintenance strategy of the pumped-storage equipment is determined based on this, which is beneficial to improving the accuracy of determining the maintenance strategy, thereby facilitating accurate maintenance of the pumped-storage equipment, thereby improving the maintenance effect of the pumped-storage equipment, and at the same time avoiding the limitations of the maintenance strategy formulated manually, resulting in the defect of poor maintenance effect on the pumped-storage equipment.

[0170] In an exemplary embodiment, in order to more clearly illustrate the pumped storage equipment maintenance method provided in the embodiment of the present application, the pumped storage equipment maintenance method is specifically described below with a specific embodiment. In an exemplary embodiment, Figure 8 As shown, the present application also proposes a maintenance strategy processing method based on RCM, which first performs RCM analysis, then performs equipment status evaluation, then determines the maintenance strategy, and finally implements and evaluates the maintenance strategy; specifically, it includes the following contents:

[0171] (1) Obtain equipment data of pumped storage equipment, perform RCM analysis (such as failure mode analysis, failure impact analysis, and maintenance mode analysis) on the equipment data, and obtain relevant analysis results. This part can be achieved through a neural network model.

[0172] (2) Based on the relevant analysis results, evaluate the equipment status, including health and importance. This part can be achieved through a large model.

[0173] (3) According to the health and importance evaluation results, the equipment control level is determined according to the equipment risk matrix. The equipment control level is divided into "Level I, Level II, Level III and Level IV" from high to low. According to the RCM maintenance strategy library and control level, the corresponding maintenance decision is determined, which includes the maintenance items, maintenance time and maintenance cycle of each unit.

[0174] (4) Evaluate the actual effect of maintenance strategy implementation by reliability index, maintenance cost and equipment status. This part can be achieved through neural network model.

[0175] (5) According to the actual effect of the maintenance strategy implementation, reliability improvements are made and the maintenance strategy is iteratively improved.

[0176] The steps for performing RCM analysis on equipment data specifically include the following:

[0177] (1) Acquire equipment data of the pumped storage equipment, and select first data related to fault mode analysis, second data related to fault impact analysis, and third data related to maintenance mode analysis from the equipment data.

[0178] (2) Using the large model, based on the first data, a fault mode analysis is performed to obtain the fault mode analysis results. For example, a typical fault tree model is established according to the three-level division of equipment, components, and elements, and all events of functional failure of the equipment are analyzed based on the three-level fault tree analysis (component / element name + root cause + fault characteristics + fault mechanism). A total of 8 major components, 44 components, and 200 fault modes of the pump-turbine system of the pumped storage unit were analyzed, and 13 major components, 69 components, and 503 fault modes of the generator motor system were analyzed.

[0179] (3) Using the large model, based on the second data, a fault impact analysis is performed to obtain the results of the fault impact analysis. For example, the fault impact analysis determines the impact of equipment failure on personal, power grid, and equipment risks from the three levels of components, parts, and equipment. After analysis, there are 4 types of fault modes that cause personal impact in the pumped storage unit (1 general accident and 3 level 4 events); 130 types of power grid impacts (6 level 1 events, 22 level 2 events, 10 level 3 events, 24 level 4 events, 60 level 5 events, and 8 level 6 events); 27 types of equipment impacts (1 level 1 event, 13 level 2 events, 2 level 3 events, 5 level 4 events, and 6 level 6 events).

[0180] (4) Using the big model and based on the third data, the maintenance mode analysis is performed to obtain the maintenance mode analysis results. For example, for each fault mode, the appropriate maintenance task type, maintenance item, and maintenance reference period are selected. For example, 49 fault modes of pumped storage units adopt condition maintenance, 574 fault modes adopt periodic maintenance, and 80 fault modes adopt post-maintenance.

[0181] (5) Obtain the final fault analysis result based on the fault mode analysis results, fault impact analysis results and maintenance mode analysis results.

[0182] Among them, according to the relevant analysis results, the steps for evaluating the equipment status include the following:

[0183] (1) Obtain the failure mode analysis results and failure impact analysis results of pumped storage equipment.

[0184] (2) Based on the data of operation inspection, maintenance, overhaul, preventive test and live test (online monitoring) involved in the failure mode analysis results, score each characteristic state quantity reflecting the health of the equipment in the failure mode to obtain the score of each characteristic state quantity. This part can be implemented through a neural network model, for example, inputting the data features of the relevant data into the model, and outputting the score of each characteristic state quantity through the model.

[0185] (3) According to the weight of each feature state quantity, the scores of each feature state quantity are fused to obtain the target score.

[0186] (4) Based on the target score, determine the equipment health level of the equipment, such as normal, caution, abnormal, and severe.

[0187] (5) From the fault impact analysis results, extract the personal safety impact results, power grid safety impact results, and equipment safety impact results.

[0188] (6) Determine the event levels corresponding to the personal safety impact results, power grid safety impact results, and equipment safety impact results respectively.

[0189] (7) From the event levels corresponding to the personal safety impact results, the power grid safety impact results, and the equipment safety impact results, the highest level of event level is screened out as the equipment importance level of the equipment, such as general, concern, important, and critical.

[0190] (8) Obtain the weight corresponding to the equipment health level and the weight corresponding to the equipment importance level, and fuse the equipment health level and the equipment importance level according to the weight corresponding to the equipment health level and the weight corresponding to the equipment importance level to obtain the equipment status evaluation result.

[0191] The above embodiments can achieve the following technical effects: (1) improving the maintenance effect of pumped storage equipment; (2) improving the processing accuracy of maintenance strategies; (3) improving the accuracy of equipment fault analysis; (4) improving the accuracy of equipment status evaluation.

[0192] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0193] Based on the same inventive concept, the embodiment of the present application also provides a pumped storage equipment maintenance device for implementing the above-mentioned pumped storage equipment maintenance method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the pumped storage equipment maintenance device provided below can refer to the limitations of the pumped storage equipment maintenance method above, and will not be repeated here.

[0194] In an exemplary embodiment, Fig. 9 As shown, a pumped storage equipment maintenance device is provided, including: a data acquisition module 910, a fault analysis module 920, a state identification module 930, a level determination module 940, a strategy query module 950 and an equipment maintenance module 960, wherein:

[0195] The data acquisition module 910 is used to acquire equipment data of the pumped storage equipment.

[0196] The fault analysis module 920 is used to input the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment.

[0197] The state recognition module 930 is used to input the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes the target health and the target importance.

[0198] The level determination module 940 is used to determine the control level of the pumped storage equipment according to the target health and target importance.

[0199] The strategy query module 950 is used to query the maintenance strategy library according to the control level, and obtain the maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment.

[0200] The equipment maintenance module 960 is used to perform corresponding maintenance processing on the pumped-storage equipment according to the maintenance strategy of the pumped-storage equipment.

[0201] In an exemplary embodiment, the pre-trained fault analysis model includes a fault mode analysis model, a fault impact analysis model, and a maintenance mode analysis model;

[0202] The fault analysis module 920 is also used to screen out first equipment data related to fault mode analysis, second equipment data related to fault impact analysis, and third equipment data related to maintenance mode analysis from the equipment data; input the first equipment data into the fault mode analysis model to obtain the fault mode analysis result of the pumped-storage equipment; input the second equipment data into the fault impact analysis model to obtain the fault impact analysis result of the pumped-storage equipment; input the third equipment data into the maintenance mode analysis model to obtain the maintenance mode analysis result of the pumped-storage equipment; and fuse the fault mode analysis result, the fault impact analysis result, and the maintenance mode analysis result to obtain the fault analysis result of the pumped-storage equipment.

[0203] In an exemplary embodiment, the pre-trained device state recognition model includes a device health recognition model and a device importance recognition model;

[0204] The state identification module 930 is also used to screen out a first fault analysis result related to the health analysis and a second fault analysis result related to the importance analysis from the fault analysis results; perform feature extraction processing on the first fault analysis result to obtain a first feature vector corresponding to the first fault analysis result, and perform feature extraction processing on the second fault analysis result to obtain a second feature vector corresponding to the second fault analysis result; input the first feature vector into the equipment health identification model to obtain the target health of the pumped-storage equipment, and input the second feature vector into the equipment importance identification model to obtain the target importance of the pumped-storage equipment; and combine the target health and the target importance to obtain the equipment state of the pumped-storage equipment.

[0205] In an exemplary embodiment, the state identification module 930 is also used to input the first feature vector into the equipment health identification model to obtain the state quantity evaluation results of each characteristic state quantity of the pumped-storage equipment; screen out the state quantity evaluation results of the key characteristic state quantities from the state quantity evaluation results of each characteristic state quantity; fuse the state quantity evaluation results of the key characteristic state quantities according to the weights of the key characteristic state quantities to obtain the target evaluation results of the pumped-storage equipment; and obtain the health corresponding to the target evaluation result as the target health of the pumped-storage equipment.

[0206] In an exemplary embodiment, the state identification module 930 is also used to input the second feature vector into the equipment importance identification model to obtain the event level of the pumped-storage equipment under each preset safety impact dimension; from the event levels of the pumped-storage equipment under each preset safety impact dimension, the event level of the pumped-storage equipment under the key safety impact dimension is screened out; from the event levels of the pumped-storage equipment under the key safety impact dimension, the highest level of event level is screened out as the target event level; and the importance corresponding to the target event level is obtained as the target importance of the pumped-storage equipment.

[0207] In an exemplary embodiment, the level determination module 940 is also used to obtain an equipment risk matrix; the equipment risk matrix is ​​used to represent the correspondence between health, importance and control level; the equipment risk matrix is ​​queried to obtain the control level corresponding to the target health and target importance as the control level of the pumped storage equipment.

[0208] Each module in the above pumped storage equipment maintenance device can be implemented in whole or in part by software, hardware and a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0209] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as equipment data and maintenance strategies. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for maintenance of pumped storage equipment is implemented.

[0210] Those skilled in the art will understand that Fig.10The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0211] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0212] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0213] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0214] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0215] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0216] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for overhauling a pumped storage equipment, characterized in that: The method comprises: Obtain equipment data for pumped storage equipment; Inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment; Inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes a target health degree and a target importance degree; Determining a control level of the pumped-storage equipment according to the target health and the target importance; Querying a maintenance strategy library according to the control level to obtain a maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment; According to the maintenance strategy of the pumped-storage equipment, the pumped-storage equipment is subjected to corresponding maintenance processing.

2. The method according to claim 1, characterized in that The pre-trained fault analysis model includes a fault mode analysis model, a fault impact analysis model and a maintenance mode analysis model; The step of inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment comprises: Filtering out, from the equipment data, first equipment data related to failure mode analysis, second equipment data related to failure impact analysis, and third equipment data related to maintenance mode analysis; Inputting the first equipment data into the failure mode analysis model to obtain a failure mode analysis result of the pumped storage equipment; Inputting the second equipment data into the fault impact analysis model to obtain a fault impact analysis result of the pumped storage equipment; Inputting the third equipment data into the maintenance mode analysis model to obtain a maintenance mode analysis result of the pumped-storage equipment; The fault mode analysis result, the fault impact analysis result and the maintenance mode analysis result are fused to obtain the fault analysis result of the pumped storage equipment.

3. The method according to claim 1, characterized in that The pre-trained device status recognition model includes a device health recognition model and a device importance recognition model; The step of inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment comprises: From the fault analysis results, screen out a first fault analysis result related to health analysis and a second fault analysis result related to importance analysis; Performing feature extraction processing on the first fault analysis result to obtain a first feature vector corresponding to the first fault analysis result, and performing feature extraction processing on the second fault analysis result to obtain a second feature vector corresponding to the second fault analysis result; Inputting the first feature vector into the equipment health recognition model to obtain the target health of the pumped-storage equipment, and inputting the second feature vector into the equipment importance recognition model to obtain the target importance of the pumped-storage equipment; The target health and target importance are combined to obtain the equipment state of the pumped-storage equipment.

4. The method according to claim 3, characterized in that: The step of inputting the first feature vector into the equipment health recognition model to obtain the target health of the pumped-storage equipment includes: Inputting the first characteristic vector into the equipment health recognition model to obtain a state quantity evaluation result of each characteristic state quantity of the pumped-storage equipment; Filtering out the state quantity evaluation results of key characteristic state quantities from the state quantity evaluation results of the characteristic state quantities; According to the weight of the key characteristic state quantity, the state quantity evaluation result of the key characteristic state quantity is fused to obtain the target evaluation result of the pumped storage equipment; A health degree corresponding to the target evaluation result is obtained as a target health degree of the pumped-storage equipment.

5. The method according to claim 3, characterized in that: The step of inputting the second feature vector into the equipment importance recognition model to obtain the target importance of the pumped-storage equipment includes: Inputting the second feature vector into the equipment importance identification model to obtain the event level of the pumped-storage equipment under each preset safety impact dimension; Filtering out the event levels of the pumped-storage equipment under the key safety impact dimensions from the event levels of the pumped-storage equipment under each preset safety impact dimension; From the event levels of the pumped storage equipment under the key safety impact dimension, the highest event level is selected as the target event level; The importance corresponding to the target event level is obtained as the target importance of the pumped-storage equipment.

6. The method according to any one of claims 1 to 5, characterized in that: The step of determining the control level of the pumped-storage equipment according to the target health and the target importance includes: Obtaining an equipment risk matrix; the equipment risk matrix is ​​used to represent the corresponding relationship between health, importance and control level; The equipment risk matrix is ​​queried to obtain a control level corresponding to the target health and the target importance as the control level of the pumped-storage equipment.

7. A pumped storage equipment maintenance device, characterized in that: The device comprises: A data acquisition module, used to acquire equipment data of the pumped storage equipment; A fault analysis module, used for inputting the equipment data into a pre-trained fault analysis model to obtain a fault analysis result of the pumped storage equipment; A state recognition module, used for inputting the fault analysis result into a pre-trained equipment state recognition model to obtain the equipment state of the pumped storage equipment; the equipment state includes a target health degree and a target importance degree; A level determination module, used to determine the control level of the pumped storage equipment according to the target health and the target importance; A strategy query module, used to query the maintenance strategy library according to the control level, and obtain the maintenance strategy corresponding to the control level as the maintenance strategy of the pumped storage equipment; The equipment maintenance module is used to perform corresponding maintenance processing on the pumped-storage equipment according to the maintenance strategy of the pumped-storage equipment.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images