Intelligent operation and maintenance management method for energy storage power station

Abstract

The invention provides an intelligent operation and maintenance management method for an energy storage power station, and relates to the technical field of equipment operation and maintenance, and the method comprises the steps: 1, obtaining the operation data of all equipment in the energy storage power station in real time, and outputting the initial data; 2, feature extraction is conducted on the initial data, an operation feature set is constructed and obtained, a corresponding preprocessing method is matched to conduct preprocessing on the initial data, and to-be-analyzed data is obtained; 3, performing anomaly analysis and prediction analysis on the to-be-analyzed data, and outputting an operation and maintenance analysis result based on an anomaly analysis result and a prediction analysis result; and 4, matching an operation and maintenance method in the operation and maintenance method database to perform operation and maintenance on corresponding equipment in the energy storage power station, and synchronously generating an operation and maintenance log. The system can monitor and analyze the operation data of each device in the energy storage power station in real time, predict the operation state of each device, and realize early discovery and early solving of operation and maintenance problems such as device faults, thereby improving the operation and maintenance efficiency of the energy storage power station, and ensuring the stable operation of the power station.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment operation and maintenance, and in particular to an intelligent operation and maintenance management method for an energy storage power station. Background Art

[0002] Energy storage power stations can store electricity and release it when needed, which can effectively solve the imbalance of electricity in time and space. The application of energy storage power station technology runs through all aspects of power generation, transmission, distribution, and power consumption in the power system, realizing peak shaving and valley filling in the power system, smoothing and tracking plan processing of renewable energy power generation fluctuations, efficient system frequency regulation, and increasing power supply reliability. In order to ensure the normal operation of energy storage power stations, they need to be operated and maintained.

[0003] Most of the existing operation and maintenance methods can only repair equipment through manual operation and maintenance after the equipment fails. This traditional manual operation and maintenance method not only has low operation and maintenance efficiency, but also is based on manual inspections and manual analysis. It is restricted by limited human resources and data mining and analysis methods, and is difficult to meet the increasing operation and maintenance needs. It not only generates a large amount of manual operation and maintenance expenses, but may also lead to the lack, omission or delay of risk point identification and troubleshooting of abnormal batteries, abnormal components, etc. due to manual reasons, thereby causing greater safety risks and maintenance and replacement cost losses.

[0004] Therefore, the present invention provides an intelligent operation and maintenance management method for an energy storage power station. Summary of the invention

[0005] The present invention provides an intelligent operation and maintenance management method for an energy storage power station, which is used to monitor and analyze the operating data of each device in the energy storage power station in real time, and predict the operating status of each device, so as to achieve early detection and early resolution of operation and maintenance problems such as equipment failure, thereby improving the operation and maintenance efficiency of the energy storage power station and ensuring the stable operation of the power station.

[0006] The present invention provides an intelligent operation and maintenance management method for an energy storage power station, comprising:

[0007] Step 1: Obtain the operating data of each device in the energy storage power station in real time through the preset monitoring equipment and output the initial data;

[0008] Step 2: extracting features from the initial data to construct an operating feature set, and preprocessing the initial data based on the operating feature set by matching a corresponding preprocessing method to obtain data to be analyzed;

[0009] Step 3: Perform an abnormality analysis on the data to be analyzed in combination with a preset abnormality database, and at the same time, perform a predictive analysis on the data to be analyzed based on a preset predictive model, and output an operation and maintenance analysis result based on the abnormality analysis result and the predictive analysis result;

[0010] Step 4: Based on the operation and maintenance analysis results, a corresponding operation and maintenance method is matched in the operation and maintenance method database, and the corresponding equipment in the energy storage power station is operated and maintained based on the operation and maintenance method, and an operation and maintenance log is generated simultaneously.

[0011] Preferably, step 1 includes:

[0012] The real-time operation data of each device is obtained by setting up preset monitoring devices at multiple preset points in the energy storage power station;

[0013] The real-time operation data corresponding to each device in the energy storage power station is summarized and the initial data is output.

[0014] Preferably, step 2 includes:

[0015] Extracting features from the initial data using a preset feature extraction method, and constructing an operating feature set based on the extracted features;

[0016] A matching preprocessing method is selected from a preprocessing method database based on the operation feature set, and the initial data is preprocessed based on the preprocessing method to obtain data to be analyzed.

[0017] Preferably, selecting a matching preprocessing method from a method database based on the operating feature set includes:

[0018] Based on the operation feature set and in combination with a preset feature-factor comparison table, a first matching factor corresponding to each feature in the operation feature set is obtained;

[0019] Based on all the first matching factors corresponding to the operation feature set, a corresponding preprocessing method is selected from a method database.

[0020] Preferably, step 3 includes:

[0021] Compare and analyze the data to be analyzed with the corresponding data in the preset abnormality database to obtain an abnormality analysis result;

[0022] At the same time, based on the operation feature set, a matching preset prediction model is selected in the model database, and the data to be analyzed is input into the preset prediction model for prediction analysis to obtain a prediction analysis result;

[0023] The abnormal analysis results and the prediction analysis results are comprehensively analyzed to obtain the operation and maintenance analysis results.

[0024] Preferably, the data to be analyzed is compared with the corresponding data in the preset abnormality database to obtain an abnormality analysis result, including:

[0025] Inputting the operation feature set into a preset abnormality database for feature matching, extracting abnormal data features that meet a first threshold condition from the preset abnormality database, and constructing a first feature set based on all the extracted abnormal data features;

[0026] Based on the first feature set, corresponding source data is selected from a preset abnormal database to construct first data;

[0027] Acquire a mapping relationship between the operation feature set and each data feature in the first feature set, and construct a feature relationship graph;

[0028] Based on the characteristic relationship diagram, the first data under each mapping relationship is compared and analyzed with the data to be analyzed to obtain a first comparison result corresponding to each mapping relationship;

[0029] Summarizing and analyzing the first comparison results under all mapping relationships in the feature relationship graph to obtain an abnormality analysis result;

[0030] The abnormal analysis result is input into a preset result analysis model for analysis, and the abnormal state matching the abnormal analysis result is obtained by combining with a preset result-state comparison table, and a result-state analysis table is output;

[0031] Input the result-state analysis table into a state database for identification and determination, and determine the state type to which each abnormal state in the result-state analysis table belongs and the risk level under the corresponding type;

[0032] Based on the result-status analysis table and the status type and risk level corresponding to each abnormal status in the table, output the status real-time determination result;

[0033] Based on the real-time status determination result, the corresponding alarm instruction is matched in the early warning database and an alarm signal is sent.

[0034] Preferably, selecting a matching preset prediction model from a model database based on the operating feature set, and inputting the data to be analyzed into the preset prediction model for prediction analysis to obtain a prediction analysis result includes:

[0035] Selecting a matching first prediction model in a model database based on the operation feature set, and selecting historical operation and maintenance data whose similarity satisfies a second threshold condition in a history database based on the operation feature set;

[0036] Combining the format adaptation information corresponding to the first prediction model, converting the format of the historical operation and maintenance data to obtain data to be divided;

[0037] In combination with the obtained prediction performance requirement information of the energy storage power station for the preset prediction model, a matching data set partitioning method is selected in the method database, and the data to be partitioned is partitioned based on the data set partitioning method to obtain an operation and maintenance training set, an operation and maintenance verification set, and an operation and maintenance test set respectively;

[0038] Inputting the operation and maintenance training set and the operation and maintenance verification set into the first prediction model respectively, training and optimizing the first prediction model respectively, to obtain a second prediction model;

[0039] Inputting the operation and maintenance test set into the second prediction model for performance testing, and determining the second prediction model whose test result meets the prediction performance requirement information as the preset prediction model;

[0040] The data to be analyzed is input into the preset prediction model for prediction analysis to obtain prediction analysis results.

[0041] Preferably, step 4 includes:

[0042] Analyze the operation and maintenance analysis results, and determine all operation and maintenance types contained in the operation and maintenance analysis results based on the analyzed content, wherein the operation and maintenance types include: equipment maintenance, fault diagnosis and repair, energy efficiency optimization, and data monitoring management;

[0043] Classifying the operation and maintenance analysis results based on the operation and maintenance type to obtain sub-results corresponding to each category;

[0044] Combined with the preset result-factor matching table, obtaining the screening factor corresponding to each of the sub-results;

[0045] Based on the screening factors under each category, selecting an operation and maintenance method matching each of the sub-results from the operation and maintenance method database;

[0046] At the same time, obtaining the equipment information corresponding to each of the sub-results and the pre-maintenance data of the corresponding equipment, matching the operation and maintenance method with the equipment corresponding to the corresponding sub-result, and outputting the matching results that meet the third threshold condition as a method-equipment comparison table;

[0047] Based on the mapping relationship between each operation and maintenance method and equipment in the method-equipment comparison table, the corresponding equipment in the energy storage power station is operated and maintained, and the operating parameters of the corresponding equipment after operation and maintenance are obtained and output as post-maintenance data;

[0048] Based on the pre-maintenance data and the post-maintenance data, and in combination with preset effect evaluation indicators, the operation and maintenance effect under each of the sub-results is evaluated, and an operation and maintenance effect evaluation result is output;

[0049] Performing status determination on the pre-dimension data and post-dimension data corresponding to each device in the method-device table to obtain a first status and a second status respectively;

[0050] Compare and analyze the first state and the second state corresponding to each device to obtain a state analysis result of each device, and summarize them to obtain a device state change table;

[0051]

[0052] Among them, Sj represents the state analysis result of the jth device; State 1j (live) indicates the first state level corresponding to the real-time data in the pre-dimensional data under the j-th device; State 2j (live) indicates the second status level corresponding to the real-time data in the post-dimension data under the j-th device; State represents the standard coefficient of the state level change corresponding to the jth device; Lj1 represents the first adjustment coefficient corresponding to the real-time state change of the jth device determined based on real-time data before and after operation and maintenance; State 1j (pre) represents the first state level corresponding to the predicted data in the pre-dimensional data under the j-th device; State 2j (pre) represents the second state level corresponding to the predicted data in the post-maintenance data of the j-th device; Lj2 represents the second adjustment coefficient corresponding to the predicted state change of the j-th device determined according to the predicted data before and after operation and maintenance;

[0053] Based on the operation and maintenance effect evaluation results and the equipment status change table, and in combination with the corresponding timestamp, an operation and maintenance log is generated.

[0054] Beneficial effects of the present invention: The present invention obtains data through real-time monitoring equipment, extracts and preprocesses features, combines abnormal analysis and predictive analysis, outputs operation and maintenance suggestions and executes corresponding operation and maintenance methods, and finally records operation and maintenance logs, thereby realizing intelligent operation and maintenance management of energy storage power stations. The present invention can realize real-time monitoring and analysis, discover equipment problems in advance and accurately predict the operating status of each device in the energy storage power station, thereby facilitating the optimization of operation and maintenance strategies, reducing operation and maintenance costs, and thus improving system reliability and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1It is a flow chart of an intelligent operation and maintenance management method of an energy storage power station provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] like Figure 1 As shown, an embodiment of the present invention provides an intelligent operation and maintenance management method for an energy storage power station, comprising:

[0059] Step 1: Obtain the operating data of each device in the energy storage power station in real time through the preset monitoring equipment and output the initial data;

[0060] Step 2: Extract features from the initial data, construct an operating feature set, and preprocess the initial data based on the operating feature set by matching the corresponding preprocessing method to obtain the data to be analyzed;

[0061] Step 3: Perform an anomaly analysis on the data to be analyzed in combination with the preset anomaly database. At the same time, perform predictive analysis on the data to be analyzed based on the preset predictive model, and output the operation and maintenance analysis results based on the anomaly analysis results and the predictive analysis results;

[0062] Step 4: Based on the operation and maintenance analysis results, the corresponding operation and maintenance method is matched in the operation and maintenance method database, and the corresponding equipment in the energy storage power station is operated and maintained based on the operation and maintenance method, and the operation and maintenance log is generated simultaneously.

[0063] In this embodiment, the preset monitoring device: a device used to monitor the operating status of each device inside the energy storage power station, used to obtain the operating data of the device in real time, such as a sensor, an intelligent monitoring device, a data collector, etc.;

[0064] In this embodiment, operating data: data obtained in real time from a preset monitoring device, reflecting the real-time operating status of each device in the energy storage power station, for example, battery charging and discharging status, temperature, voltage, current and other data;

[0065] In this embodiment, initial data: raw data extracted from the operating data, serving as a starting point for subsequent analysis and processing, for example, raw battery temperature and voltage data obtained from a monitoring device;

[0066] In this embodiment, the operating feature set is a feature set extracted from the initial data, which is used to describe the operating characteristics of the device, for example, statistical characteristics such as the mean, maximum, and minimum values ​​of the battery temperature;

[0067] In this embodiment, preprocessing method: a data preprocessing method performed on initial data, used to prepare data for subsequent analysis, for example, data cleaning, missing value processing, standardization and other data processing methods;

[0068] In this embodiment, data to be analyzed: data to be subjected to abnormality analysis and predictive analysis after preprocessing;

[0069] In this embodiment, the preset abnormality database: the preset abnormality database is used to store various abnormal conditions that may occur, such as battery overheating, voltage abnormality, etc.;

[0070] In this embodiment, the preset prediction model: a pre-established model for predicting the operating data and operating status of a device, for example, a battery life prediction model established based on historical data;

[0071] In this embodiment, abnormal analysis result: abnormal situation analysis result obtained after abnormality detection is performed on the data to be analyzed;

[0072] In this embodiment, the prediction analysis result is: the result obtained after predicting the data to be analyzed based on the preset prediction model;

[0073] In this embodiment, the operation and maintenance analysis result: the operation and maintenance suggestion obtained by combining the abnormal analysis result and the prediction analysis result, for example, it is recommended to cool the battery and adjust the charging strategy;

[0074] In this embodiment, the operation and maintenance method database: a database storing various operation and maintenance methods, used to match corresponding operation and maintenance methods, such as regular maintenance, remote monitoring, equipment replacement, etc.;

[0075] In this embodiment, the operation and maintenance method: specific operation and maintenance operations or measures matched according to the operation and maintenance analysis results, for example, lowering the battery temperature, adjusting the charging rate, etc.;

[0076] In this embodiment, the operation and maintenance log is a log recording the operation and maintenance operations, which is used to track and analyze the operation and maintenance process to facilitate subsequent tracing analysis.

[0077] The implementation principle and beneficial effects of this embodiment: The present invention obtains data through real-time monitoring equipment, extracts and preprocesses features, combines abnormal analysis and predictive analysis, outputs operation and maintenance suggestions and executes corresponding operation and maintenance methods, and finally records operation and maintenance logs to achieve intelligent operation and maintenance management of energy storage power stations. The present invention can achieve real-time monitoring and analysis, discover equipment problems in advance and predict future operating status, thereby optimizing operation and maintenance strategies, reducing operation and maintenance costs, and improving system reliability and safety.

[0078] An embodiment of the present invention provides an intelligent operation and maintenance management method for an energy storage power station, wherein step 1 includes:

[0079] The real-time operation data of each device is obtained by setting up preset monitoring devices at multiple preset points in the energy storage power station;

[0080] Summarize the real-time operating data of each device in the energy storage power station and output the initial data.

[0081] In this embodiment, preset points: multiple locations pre-set in the energy storage power station, used to install preset monitoring equipment to obtain real-time operating data of various devices, for example, the locations of battery packs, inverters, transformers and other equipment;

[0082] In this embodiment, real-time operation data: real-time operation data of each device obtained from a preset monitoring device set at a preset point, reflecting the actual operation status of the device, for example, battery temperature, voltage, current, charge and discharge status and other data.

[0083] The implementation principle and beneficial effects of this embodiment: The present invention sets preset monitoring devices at multiple preset points in the energy storage power station, obtains the operating data of each device in real time, summarizes these data, generates initial data, and provides a basis for subsequent analysis and operation and maintenance. It can comprehensively monitor the real-time operating status of each device in the energy storage power station, realize comprehensive data collection and aggregation, and provide accurate and reliable data support for subsequent feature extraction, abnormal analysis and predictive analysis, thereby improving the operation and maintenance efficiency and system reliability of the power station.

[0084] An intelligent operation and maintenance management method for an energy storage power station provided by an embodiment of the present invention includes, in step 2:

[0085] Extract features from the initial data using a preset feature extraction method, and construct a running feature set based on the extracted features;

[0086] A matching preprocessing method is selected from a preprocessing method database based on the running feature set, and the initial data is preprocessed based on the preprocessing method to obtain data to be analyzed.

[0087] In this embodiment, the feature extraction method is preset: this is a method preset to extract features from the initial data, which is used to describe the operating characteristics of the device;

[0088] In this embodiment, the preprocessing method database: a database storing various preprocessing methods, used to select an appropriate preprocessing method according to the operating feature set, for example, data cleaning, standardization, noise reduction and other preprocessing methods.

[0089] The implementation principle and beneficial effects of this embodiment are as follows: the initial data is feature extracted using a preset feature extraction method to construct an operation feature set. Then, a matching preprocessing method is selected from the preprocessing method database according to the operation feature set, and the initial data is preprocessed to obtain data to be analyzed. By extracting key features and selecting appropriate preprocessing methods, the present invention can effectively prepare data for subsequent abnormal analysis and predictive analysis, improve the quality and analyzability of the data, and thus provide more accurate data support for subsequent operation and maintenance decisions.

[0090] An intelligent operation and maintenance management method for an energy storage power station provided by an embodiment of the present invention selects a matching preprocessing method from a method database based on an operation feature set, including:

[0091] Based on the operation feature set and in combination with a preset feature-factor comparison table, a first matching factor corresponding to each feature in the operation feature set is obtained;

[0092] Based on all first matching factors corresponding to the running feature set, a corresponding preprocessing method is selected from a method database.

[0093] In this embodiment, the preset feature-factor comparison table is a preset table used to compare features with factors so that the first matching factor corresponding to each operating feature can be accurately found in the subsequent matching process. For example, features such as battery temperature and voltage may correspond to factors such as device health status and charge and discharge efficiency.

[0094] In this embodiment, the first matching factor: the first matching factor corresponding to each feature in the operating feature set, is used to select the corresponding preprocessing method in the method database. For example, for the feature "battery temperature", the possible first matching factor is "device health status".

[0095] The implementation principle and beneficial effects of this embodiment: The present invention obtains the first matching factor corresponding to each feature according to the running feature set combined with the preset feature-factor comparison table, and selects the corresponding preprocessing method in the method database based on these first matching factors. The present invention can more accurately select the preprocessing method suitable for the running feature set, thereby improving the accuracy and efficiency of data processing.

[0096] An intelligent operation and maintenance management method for an energy storage power station provided by an embodiment of the present invention includes, in step 3:

[0097] Compare and analyze the data to be analyzed with the corresponding data in the preset abnormal database to obtain the abnormal analysis results;

[0098] At the same time, based on the running feature set, a matching preset prediction model is selected in the model database, and the data to be analyzed is input into the preset prediction model for prediction analysis to obtain a prediction analysis result;

[0099] Comprehensively analyze the abnormal analysis results and predictive analysis results to obtain the operation and maintenance analysis results.

[0100] In this embodiment, the model database: a database storing various preset prediction models, used to select a matching prediction model for prediction analysis based on the running feature set, for example, including prediction models based on machine learning, such as neural network, support vector machine, random forest and other models.

[0101] The implementation principle and beneficial effects of this embodiment are as follows: the data to be analyzed is compared and analyzed with the corresponding data in the preset abnormality database to obtain the abnormality analysis results; at the same time, a matching preset prediction model is selected in the model database based on the operation feature set, and the data to be analyzed is input into the prediction model for prediction analysis to obtain the prediction analysis results; finally, the abnormality analysis results and the prediction analysis results are combined to obtain the operation and maintenance analysis results. By comparing and analyzing the abnormal data and the prediction analysis results, combined with the preset prediction model selected by the operation feature set, the present invention can more comprehensively evaluate the operation status of the energy storage power station, discover potential problems in advance and perform predictive maintenance, thereby improving the reliability and efficiency of the system, reducing potential failure risks, and optimizing the accuracy and timeliness of operation and maintenance decisions.

[0102] An intelligent operation and maintenance management method for an energy storage power station provided by an embodiment of the present invention compares and analyzes the data to be analyzed with the corresponding data in a preset abnormality database to obtain an abnormality analysis result, including:

[0103] Inputting the running feature set into a preset abnormal database for feature matching, extracting abnormal data features that meet a first threshold condition from the preset abnormal database, and constructing a first feature set based on all the extracted abnormal data features;

[0104] Select corresponding source data from a preset abnormal database based on the first feature set to construct first data;

[0105] Obtaining a mapping relationship between the running feature set and each data feature in the first feature set, and constructing a feature relationship graph;

[0106] Based on the characteristic relationship graph, the first data under each mapping relationship is compared and analyzed with the data to be analyzed to obtain a first comparison result corresponding to each mapping relationship;

[0107] Summarize and analyze the first comparison results under all mapping relationships in the feature relationship graph to obtain an abnormal analysis result;

[0108] The abnormal analysis results are input into the preset result analysis model for analysis, and the abnormal state matching the abnormal analysis results is obtained by combining the preset result-state comparison table, and the result-state analysis table is output;

[0109] Input the result-state analysis table into the state database for identification and judgment, and determine the state type to which each abnormal state in the result-state analysis table belongs and the risk level under the corresponding type;

[0110] Based on the result-status analysis table and the status type and risk level corresponding to each abnormal status in the table, output the real-time status determination result;

[0111] Based on the real-time status judgment results, the corresponding alarm instructions are matched in the early warning database and an alarm signal is sent.

[0112] In this embodiment, feature matching: matching the operating feature set with feature data in the abnormality database, for example, matching the temperature feature of the battery in the energy storage power station with the battery overheating feature data in the abnormality database;

[0113] In this embodiment, the first threshold condition: the condition that the abnormal data feature needs to meet in order to be extracted;

[0114] In this embodiment, abnormal data features: features extracted from data in the abnormal database, such as abnormal battery temperature, abnormal charge and discharge rate, etc.;

[0115] In this embodiment, the first feature set: a set consisting of abnormal data features extracted according to the first threshold condition;

[0116] In this embodiment, source data: corresponding data selected from an abnormal database based on each abnormal data feature in the first feature set;

[0117] In this embodiment, the first data: that is, a data set constructed based on all source data;

[0118] In this embodiment, the mapping relationship is used to characterize the association between the operation feature set and each data feature in the first feature set, for example, the mapping relationship between the battery temperature feature and the battery health status;

[0119] In this embodiment, feature relationship graph: a graph constructed based on the mapping relationship between features;

[0120] In this embodiment, the first comparison result is a result obtained by comparing and analyzing the first data under each mapping relationship with the data to be analyzed, for example, determining the battery overheating state based on the comparison and analysis of the battery temperature abnormality characteristics;

[0121] In this embodiment, the preset result analysis model: a preset model for analyzing abnormal analysis results, for example, judging the remaining battery life through a preset model based on abnormal battery temperature data;

[0122] In this embodiment, the preset result-state comparison table: a table containing the comparison relationship between abnormal analysis results and abnormal states, is preset;

[0123] In this embodiment, abnormal state: a state matching the abnormal analysis result, for example, abnormally high temperature, high pressure, etc.;

[0124] In this embodiment, the result-state analysis table: an analysis table including abnormal analysis results and corresponding abnormal states, state types and risk levels;

[0125] In this embodiment, the state database: a database storing abnormal states and related information;

[0126] In this embodiment, state type: the type to which the abnormal state belongs;

[0127] In this embodiment, risk level: describes the risk level corresponding to the abnormal state;

[0128] In this embodiment, the real-time status determination result: a real-time determination result based on the abnormal status and risk level, for example, determining that the battery life is insufficient and needs to be processed immediately;

[0129] In this embodiment, the early warning database: a database storing early warning information;

[0130] In this embodiment, the alarm instruction: an alarm instruction obtained by matching the status real-time determination result, for example, sending an instruction to replace the battery;

[0131] In this embodiment, the alarm signal: an alarm signal sent to relevant personnel or systems, for example, notifying operation and maintenance personnel to replace batteries.

[0132] The implementation principles and beneficial effects of this embodiment: The present invention can quickly and accurately identify abnormal conditions, saving manpower and time costs; it can timely identify abnormal conditions and risks, and by classifying abnormal conditions and assessing risk levels, it can take timely measures to reduce potential risks, which helps prevent equipment failures and optimize system operations; it can also integrate data analysis, state identification and alarm systems to realize intelligent operation and maintenance management of energy storage power stations and improve the overall efficiency of the system.

[0133] An intelligent operation and maintenance management method for an energy storage power station provided by an embodiment of the present invention selects a matching preset prediction model from a model database based on an operation feature set, and inputs the data to be analyzed into the preset prediction model for prediction analysis to obtain a prediction analysis result, including:

[0134] Selecting a matching first prediction model in the model database based on the operation feature set, and selecting historical operation and maintenance data whose similarity satisfies a second threshold condition in the history database based on the operation feature set;

[0135] Combining the format adaptation information corresponding to the first prediction model, the historical operation and maintenance data is format converted to obtain the data to be divided;

[0136] Combined with the obtained prediction performance requirement information of the energy storage power station for the preset prediction model, a matching data set partitioning method is selected in the method database, and the data to be partitioned is partitioned based on the data set partitioning method to obtain an operation and maintenance training set, an operation and maintenance verification set, and an operation and maintenance test set respectively;

[0137] Inputting the operation and maintenance training set and the operation and maintenance verification set into the first prediction model respectively, training and optimizing the first prediction model respectively, and obtaining a second prediction model;

[0138] Inputting the operation and maintenance test set into the second prediction model for performance testing, and determining the second prediction model whose test result satisfies the prediction performance requirement information as the preset prediction model;

[0139] The data to be analyzed is input into the preset prediction model for prediction analysis to obtain the prediction analysis results.

[0140] In this embodiment, the first prediction model: a model selected from the model database for prediction, for example, may be a linear regression model established based on historical operation and maintenance data;

[0141] In this embodiment, the second threshold condition: the condition that the historical operation and maintenance data similarity needs to meet, for example, the historical operation and maintenance data similarity needs to be above 80%;

[0142] In this embodiment, historical operation and maintenance data: past operation and maintenance data used to train and verify the model, for example, data including battery charge and discharge cycles, temperature changes, etc. during the historical operation and maintenance process;

[0143] In this embodiment, the format adaptation information is used to convert the historical operation and maintenance data into format information suitable for the prediction model;

[0144] In this embodiment, the data to be divided: data converted according to the format adaptation information, preparing data for training, verification and testing;

[0145] In this embodiment, prediction performance requirement information: performance requirements for prediction model performance, such as prediction accuracy, real-time performance, security, etc.;

[0146] In this embodiment, the data set partitioning method is a method for partitioning the data to be partitioned into a training set, a validation set, and a test set, such as random partitioning, time series partitioning, K-fold cross partitioning, and the like;

[0147] In this embodiment, the operation and maintenance training set: a data set obtained by division for model training;

[0148] In this embodiment, the operation and maintenance verification set: a data set obtained by division for model verification and optimization;

[0149] In this embodiment, the operation and maintenance test set: a data set obtained by division for model performance testing;

[0150] In this embodiment, the second prediction model: a prediction model that has been trained and optimized.

[0151] The implementation principle and beneficial effects of this embodiment: Based on historical data and prediction model requirements, the present invention obtains a prediction model that meets performance requirements through data conversion, division and model training optimization, and uses it for predictive analysis of operating data. The present invention can improve the prediction performance of the model by training and optimizing the prediction model, making it more suitable for the specific needs of energy storage power stations, integrating the training, optimization and prediction analysis functions of the prediction model, and improving the accuracy of prediction of the operating data and operating conditions of each device in the energy storage power station.

[0152] An intelligent operation and maintenance management method for an energy storage power station provided by an embodiment of the present invention includes, in step 4:

[0153] Analyze the operation and maintenance analysis results, and determine all operation and maintenance types contained in the operation and maintenance analysis results based on the analysis content, where the operation and maintenance types include: equipment maintenance, fault diagnosis and repair, energy efficiency optimization, and data monitoring management;

[0154] Classify the operation and maintenance analysis results based on the operation and maintenance type, and obtain sub-results corresponding to each category;

[0155] Combine the preset result-factor matching table to obtain the screening factor corresponding to each sub-result;

[0156] Based on the screening factors under each category, an operation and maintenance method matching each sub-result is selected from the operation and maintenance method database;

[0157] At the same time, the device information corresponding to each sub-result and the pre-maintenance data of the corresponding device are obtained, the operation and maintenance method is matched with the device corresponding to the corresponding sub-result, and the matching result that meets the third threshold condition is output as a method-device comparison table;

[0158] Based on the mapping relationship between each operation and maintenance method and equipment in the method-equipment comparison table, the corresponding equipment in the energy storage power station is operated and maintained, and the operating parameters of the corresponding equipment after operation and maintenance are obtained and output as post-maintenance data;

[0159] Based on the pre-maintenance data and post-maintenance data, and combined with the preset effect evaluation indicators, the operation and maintenance effect under each sub-result is evaluated, and the operation and maintenance effect evaluation result is output;

[0160] Perform state determination on the pre-dimension data and post-dimension data corresponding to each device in the method-device table, and obtain a first state and a second state respectively;

[0161] Compare and analyze the first state and the second state corresponding to each device to obtain the state analysis result of each device, and summarize them to obtain a device state change table;

[0162]

[0163] Among them, Sj represents the state analysis result of the jth device; State 1j (live) indicates the first state level corresponding to the real-time data in the pre-dimensional data under the j-th device; State 2j (live) indicates the second state level corresponding to the real-time data in the post-dimension data under the jth device; State std State represents the standard coefficient of the state level change corresponding to the jth device; Lj1 represents the first adjustment coefficient corresponding to the real-time state change of the jth device determined based on real-time data before and after operation and maintenance; State 1j (pre) represents the first state level corresponding to the predicted data in the pre-dimensional data under the j-th device; State 2j (pre) represents the second state level corresponding to the predicted data in the post-maintenance data of the j-th device; Lj2 represents the second adjustment coefficient corresponding to the predicted state change of the j-th device determined according to the predicted data before and after operation and maintenance;

[0164] Based on the operation and maintenance effect evaluation results and the equipment status change table, an operation and maintenance log is generated in combination with the corresponding timestamp.

[0165] In this embodiment, sub-results are results obtained by classifying the operation and maintenance analysis results according to different operation and maintenance types. For example, for the operation and maintenance type of equipment maintenance, the sub-results may include equipment maintenance plans, maintenance records, etc.;

[0166] In this embodiment, the preset result-factor matching table is a table used to characterize the matching relationship between each sub-result and the corresponding screening factor, which is preset. For example, the screening factor corresponding to the equipment maintenance sub-result may be the equipment operation time;

[0167] In this embodiment, screening factors are key factors or features used to screen and match sub-results. For example, for fault diagnosis and repair, the screening factors may include fault code, fault occurrence time, etc.

[0168] In this embodiment, pre-maintenance data refers to the equipment-related data collected before the operation and maintenance operation is performed. For example, the initial state of the equipment, parameter settings, etc.;

[0169] In this embodiment, the third threshold condition is used to filter out results or data that meet specific requirements, for example, triggering a maintenance operation when the equipment operation time exceeds a certain threshold;

[0170] In this embodiment, the method-equipment comparison table: a table containing the matching relationship between each operation and maintenance method and the corresponding equipment, for example, equipment A needs to perform equipment maintenance method 1;

[0171] In this embodiment, post-maintenance data refers to equipment-related data collected after the operation and maintenance operation is performed, such as the status and performance parameters of the equipment after maintenance;

[0172] In this embodiment, preset effect evaluation index: a preset index or standard for evaluating the operation and maintenance effect, for example, the energy efficiency improvement rate and failure rate reduction after equipment maintenance;

[0173] In this embodiment, the operation and maintenance effect evaluation result is: the result obtained after evaluating the operation and maintenance effect according to the preset effect evaluation index, for example, the energy efficiency improvement rate after equipment maintenance is 10%;

[0174] In this embodiment, the first state: the state of the device before maintenance;

[0175] In this embodiment, the second state is the state of the device after maintenance. For example, the state of device A before maintenance is normal operation, and the state after maintenance is performance improvement.

[0176] In this embodiment, the state analysis result is: the device state change result obtained by comparing and analyzing the first state and the second state, for example, device A is in good condition after maintenance; the performance of device B is improved by 1% after maintenance;

[0177] In this embodiment, the device status change table is a table summarizing the status change of each device. For example, after maintenance, the status of device A changes from "needs maintenance" to "normal".

[0178] The implementation principle and beneficial effects of this embodiment: The present invention can select suitable operation and maintenance methods and match equipment according to the analysis and classification of operation and maintenance analysis results, combined with the preset result-factor matching table and screening factors, and generate operation and maintenance effect evaluation results and equipment status change tables by evaluating pre-maintenance data and post-maintenance data, and finally generate operation and maintenance logs. Through detailed operation and maintenance analysis and evaluation, as well as monitoring and recording of equipment status, the present invention can improve operation and maintenance efficiency, optimize equipment status management, and generate detailed operation and maintenance logs for subsequent analysis and decision-making, thereby improving the operating efficiency and reliability of the entire energy storage power station.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent operation and maintenance management method for an energy storage power station, characterized in that: include: Step 1: Obtain the operating data of each device in the energy storage power station in real time through the preset monitoring equipment and output the initial data; Step 2: extracting features from the initial data to construct an operating feature set, and preprocessing the initial data based on the operating feature set by matching a corresponding preprocessing method to obtain data to be analyzed; Step 3: Perform an abnormality analysis on the data to be analyzed in combination with a preset abnormality database, and at the same time, perform a predictive analysis on the data to be analyzed based on a preset predictive model, and output an operation and maintenance analysis result based on the abnormality analysis result and the predictive analysis result; Step 4: Based on the operation and maintenance analysis results, a corresponding operation and maintenance method is matched in the operation and maintenance method database, and the corresponding equipment in the energy storage power station is operated and maintained based on the operation and maintenance method, and an operation and maintenance log is generated simultaneously.

2. The intelligent operation and maintenance management method of an energy storage power station according to claim 1, characterized in that: Step 1 includes: The real-time operation data of each device is obtained by setting up preset monitoring devices at multiple preset points in the energy storage power station; The real-time operation data corresponding to each device in the energy storage power station is summarized and the initial data is output.

3. The intelligent operation and maintenance management method of an energy storage power station according to claim 1, characterized in that: Step 2 includes: Extracting features from the initial data using a preset feature extraction method, and constructing an operating feature set based on the extracted features; A matching preprocessing method is selected from a preprocessing method database based on the operation feature set, and the initial data is preprocessed based on the preprocessing method to obtain data to be analyzed.

4. The intelligent operation and maintenance management method of an energy storage power station according to claim 3 is characterized in that: Selecting a matching preprocessing method from a method database based on the operation feature set includes: Based on the operation feature set and in combination with a preset feature-factor comparison table, a first matching factor corresponding to each feature in the operation feature set is obtained; Based on all the first matching factors corresponding to the operation feature set, a corresponding preprocessing method is selected from a method database.

5. The intelligent operation and maintenance management method of an energy storage power station according to claim 1, characterized in that: Step 3 includes: Compare and analyze the data to be analyzed with the corresponding data in the preset abnormality database to obtain an abnormality analysis result; At the same time, based on the operation feature set, a matching preset prediction model is selected in the model database, and the data to be analyzed is input into the preset prediction model for prediction analysis to obtain a prediction analysis result; The abnormal analysis results and the prediction analysis results are comprehensively analyzed to obtain the operation and maintenance analysis results.

6. The intelligent operation and maintenance management method of an energy storage power station according to claim 5, characterized in that: Comparing and analyzing the data to be analyzed with the corresponding data in the preset abnormality database to obtain abnormality analysis results, including: Inputting the operation feature set into a preset abnormality database for feature matching, extracting abnormal data features that meet a first threshold condition from the preset abnormality database, and constructing a first feature set based on all the extracted abnormal data features; Based on the first feature set, corresponding source data is selected from a preset abnormal database to construct first data; Acquire a mapping relationship between the operation feature set and each data feature in the first feature set, and construct a feature relationship graph; Based on the characteristic relationship diagram, the first data under each mapping relationship is compared and analyzed with the data to be analyzed to obtain a first comparison result corresponding to each mapping relationship; Summarizing and analyzing the first comparison results under all mapping relationships in the feature relationship graph to obtain an abnormality analysis result; The abnormal analysis result is input into a preset result analysis model for analysis, and the abnormal state matching the abnormal analysis result is obtained by combining with a preset result-state comparison table, and a result-state analysis table is output; Input the result-state analysis table into a state database for identification and determination, and determine the state type to which each abnormal state in the result-state analysis table belongs and the risk level under the corresponding type; Based on the result-status analysis table and the status type and risk level corresponding to each abnormal status in the table, output the status real-time determination result; Based on the real-time status determination result, the corresponding alarm instruction is matched in the early warning database and an alarm signal is sent.

7. The intelligent operation and maintenance management method of an energy storage power station according to claim 5, characterized in that: Selecting a matching preset prediction model from a model database based on the operation feature set, and inputting the data to be analyzed into the preset prediction model for prediction analysis to obtain a prediction analysis result, including: Selecting a matching first prediction model in a model database based on the operation feature set, and selecting historical operation and maintenance data whose similarity satisfies a second threshold condition in a history database based on the operation feature set; Combining the format adaptation information corresponding to the first prediction model, converting the format of the historical operation and maintenance data to obtain data to be divided; In combination with the obtained prediction performance requirement information of the energy storage power station for the preset prediction model, a matching data set partitioning method is selected in the method database, and the data to be partitioned is partitioned based on the data set partitioning method to obtain an operation and maintenance training set, an operation and maintenance verification set, and an operation and maintenance test set respectively; Inputting the operation and maintenance training set and the operation and maintenance verification set into the first prediction model respectively, training and optimizing the first prediction model respectively, to obtain a second prediction model; Inputting the operation and maintenance test set into the second prediction model for performance testing, and determining the second prediction model whose test result meets the prediction performance requirement information as the preset prediction model; The data to be analyzed is input into the preset prediction model for prediction analysis to obtain prediction analysis results.

8. The intelligent operation and maintenance management method of an energy storage power station according to claim 1, characterized in that: Step 4 includes: Analyze the operation and maintenance analysis results, and determine all operation and maintenance types contained in the operation and maintenance analysis results based on the analysis content, wherein the operation and maintenance types include: equipment maintenance, fault diagnosis and repair, energy efficiency optimization, and data monitoring management; Classifying the operation and maintenance analysis results based on the operation and maintenance type to obtain sub-results corresponding to each category; Combined with the preset result-factor matching table, obtaining the screening factor corresponding to each of the sub-results; Based on the screening factors under each category, selecting an operation and maintenance method matching each of the sub-results from the operation and maintenance method database; At the same time, obtaining the equipment information corresponding to each of the sub-results and the pre-maintenance data of the corresponding equipment, matching the operation and maintenance method with the equipment corresponding to the corresponding sub-result, and outputting the matching results that meet the third threshold condition as a method-equipment comparison table; Based on the mapping relationship between each operation and maintenance method and equipment in the method-equipment comparison table, the corresponding equipment in the energy storage power station is operated and maintained, and the operating parameters of the corresponding equipment after operation and maintenance are obtained and output as post-maintenance data; Based on the pre-maintenance data and the post-maintenance data, and in combination with preset effect evaluation indicators, the operation and maintenance effect under each of the sub-results is evaluated, and an operation and maintenance effect evaluation result is output; Performing status determination on the pre-dimension data and post-dimension data corresponding to each device in the method-device table to obtain a first status and a second status respectively; Compare and analyze the first state and the second state corresponding to each device to obtain a state analysis result of each device, and summarize them to obtain a device state change table; Among them, Sj represents the state analysis result of the jth device; State 1j (live) indicates the first state level corresponding to the real-time data in the pre-dimensional data under the j-th device; State 2j (live) indicates the second state level corresponding to the real-time data in the post-dimension data under the jth device; State std State represents the standard coefficient of the state level change corresponding to the jth device; Lj1 represents the first adjustment coefficient corresponding to the real-time state change of the jth device determined based on real-time data before and after operation and maintenance; State 1j (pre) represents the first state level corresponding to the predicted data in the pre-dimensional data under the j-th device; State 2j (pre) represents the second state level corresponding to the predicted data in the post-maintenance data of the j-th device; L□2 represents the second adjustment coefficient corresponding to the predicted state change of the j-th device determined according to the predicted data before and after operation and maintenance; Based on the operation and maintenance effect evaluation results and the equipment status change table, and in combination with the corresponding timestamp, an operation and maintenance log is generated.

Images