A wind farm operation and maintenance management method and system

Abstract

The present invention relates to the field of wind power technology, and more specifically, to a method and system for wind farm operation and maintenance management. The method provided by the present invention mainly includes establishing a wind farm abnormality diagnosis model, outputting the abnormality of the current wind farm through the wind farm abnormality diagnosis model, and judging whether scheduling is currently required based on the abnormality of the current wind farm. The method provided by the present invention greatly reduces the reliance on manual inspections and improves work efficiency to a certain extent. Secondly, the model provided by the present invention can detect and warn of wind farm faults in real time, and issue scheduling instructions when manual processing is required to improve the stability of the wind farm.

Description

Technical Field

[0001] The present invention relates to the field of wind power technology, and in particular to a wind farm operation and maintenance management method and system. Background Art

[0002] The management application of intelligent operation and maintenance of the centralized control center in the operation and maintenance mode of wind farms is one of the hot topics in the current wind power industry. With the continuous increase in my country's wind power installed capacity, the scale of wind farms has gradually expanded, and the operation and maintenance work has become more and more complicated. The traditional wind farm operation and maintenance model has been unable to meet the growing operation and maintenance needs, so it is necessary to introduce the concept and technology of intelligent operation and maintenance to improve the efficiency and quality of operation and maintenance.

[0003] At present, there are some problems and challenges in the operation and maintenance of wind farms. First, the traditional operation and maintenance mode of wind farms mainly relies on manual inspection and manual operation, which is inefficient and prone to human errors. In addition, the traditional operation and maintenance mode cannot achieve real-time monitoring of the status of wind farm equipment and fault warning, resulting in a long fault handling time, affecting the operational stability and reliability of the wind farm. Summary of the invention

[0004] The purpose of the present invention is to provide a wind farm operation and maintenance management method and system to solve the above problems in the prior art.

[0005] The embodiments of the present invention are implemented by the following technical solutions: In a first aspect, the present invention provides a wind farm operation and maintenance management method, comprising: Obtain historical data of wind farm equipment, pre-process the historical data, and obtain training sets and verification sets of the historical data; Establish a wind farm anomaly diagnosis model, train the wind farm anomaly diagnosis model through the training set, and verify the wind farm anomaly diagnosis model after multiple trainings through the verification set, and output the optimal wind farm anomaly diagnosis model; Obtain the operating data of the current wind farm equipment and output the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; Determine whether dispatching is needed based on the current abnormal situation of the wind farm.

[0006] Preferably, the historical data preprocessing includes normalizing and extracting features from the historical data to obtain a feature subset of the historical data, and dividing the data set into a training set and a validation set in proportion, including: Setting a relationship function between a feature set and a feature subset, wherein the relationship function is used to represent an abnormal situation of a current wind farm; Calculate the correlation of all feature subsets in the feature set through the relationship function, extract the features with the greatest correlation and add them to the feature subset; The remaining features in the feature set are calculated with the features with the highest correlation obtained in the previous calculation respectively, and the features with the highest correlation each time are extracted and added to the feature subset until the upper limit of the feature subset is reached, and the feature subset is output.

[0007] Preferably, the relationship function between the feature set and the feature subset is set as follows:

[0008] The remaining features in the feature set are calculated with the features with the greatest correlation obtained in the last calculation, including:

[0009] In the formula, is the feature with the greatest correlation, is the number of features, For the Features, For the The eigenvalues ​​of the features, It is the feature calculation function with the greatest correlation between this feature and the feature calculated last time. is the number of remaining features, is the first of the remaining features Features.

[0010] Preferably, the establishing of the wind farm abnormality diagnosis model comprises: Get the feature subset and combine it with the data set to get the data set vector, establish the input layer, hidden layer and output layer respectively, and define the initial objective function; The initial objective function is trained with the training set to minimize the objective, and the learning rate is set. After each iteration, the weights and biases are replaced by the learning rate until the number of iterations is reached, and the optimized wind farm abnormality diagnosis model is output.

[0011] Preferably, the objective function includes:

[0012] In the formula, is the objective function value, is the activation function, is the regularization coefficient, For the samples, For the The output layer to the The connection weights of the hidden layers are is the output value of the hidden layer, is the weight coefficient, is the output value of the output layer, is the bias from the hidden layer to the output layer.

[0013] Preferably, the outputting of the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model includes: Output the abnormal situation of the current wind farm through the operation data of the current wind farm equipment and the optimized wind farm abnormality diagnosis model; If the current wind farm is abnormal, dispatching will be carried out; if the current wind farm is normal, no dispatching will be carried out.

[0014] Preferably, the scheduling includes: Obtain maintenance station information and the current location of the wind farm, and set up a maintenance station evaluation index model; The maintenance sites are evaluated through the evaluation index model and the maintenance station information to obtain several evaluation values, and the maintenance site with the largest evaluation value is selected for scheduling maintenance as a scheduling instruction and sent.

[0015] Preferably, the evaluation index model includes:

[0016] In the formula, is the number of maintenance station staff, is the number of equipment to be maintained at the maintenance station, is the distance between the maintenance station and the wind farm, is the area of ​​the maintenance station, It is the number of vehicles at the maintenance station.

[0017] In a second aspect, the present invention further provides a wind farm operation and maintenance management system, comprising: An acquisition module is configured to acquire historical data of wind farm equipment, preprocess the historical data, and obtain a training set and a verification set of the historical data; The model building module is configured to build a wind farm abnormality diagnosis model, train the wind farm abnormality diagnosis model through a training set, verify the wind farm abnormality diagnosis model after multiple trainings through a verification set, and output an optimal wind farm abnormality diagnosis model; The processing module is configured to obtain the operating data of the current wind farm equipment, output the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; and determine whether scheduling is currently required according to the abnormal situation of the current wind farm; The main control device is connected with the acquisition module, the model building module and the processing module, and is used to execute the above-mentioned wind farm operation and maintenance management method.

[0018] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects: The method provided by the present invention mainly includes establishing a wind farm abnormality diagnosis model, outputting the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; judging whether scheduling is needed at present according to the abnormal situation of the current wind farm. The method provided by the present invention greatly reduces the dependence on manual inspection and improves work efficiency to a certain extent. Secondly, the model provided by the present invention can detect and warn the faults of the wind farm in real time, and issue scheduling instructions when manual processing is required to improve the stability of the wind farm. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0022] The division of modules in this application is a logical division. There may be other division methods when implemented in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0023] The modules or submodules described independently may be physically separated or not: they may be implemented by software or hardware, and some modules or submodules may be implemented by software, and the processor may call the software to implement the functions of these modules or submodules, and other modules or submodules may be implemented by hardware, such as by hardware circuits. In addition, some or all of the modules may be selected according to actual needs to achieve the purpose of the present application.

[0024] Please refer to Figure 1 The present invention provides a wind farm operation and maintenance management method, comprising: S101: Acquire historical data of wind farm equipment, pre-process the historical data, and obtain a training set and a verification set of the historical data; Among them, the historical data of wind farm equipment includes monitoring of wind turbine operating speed, temperature, vibration and other parameters, as well as monitoring of the operating status of transformers and transmission lines. The operating data of wind farms is managed and analyzed, including real-time data collection, data storage and data mining, to obtain information on operating status analysis, fault diagnosis and efficiency improvement.

[0025] Through intelligent sensors and data acquisition systems, the operating status of wind turbines can be monitored in real time, including parameters such as temperature, vibration, current, and voltage, achieving comprehensive monitoring of the equipment.

[0026] These data can be stored, managed and analyzed to evaluate and predict the performance and health status of wind turbines, detect and repair potential faults in a timely manner, and improve the reliability and operating efficiency of wind farms.

[0027] S102: Establishing a wind farm abnormality diagnosis model, training the wind farm abnormality diagnosis model through a training set, and verifying the wind farm abnormality diagnosis model after multiple trainings through a verification set, and outputting an optimal wind farm abnormality diagnosis model; S103: Acquire the operating data of the current wind farm equipment, and output the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; By acquiring the wind farm's operating data and conducting data analysis, fault diagnosis and prediction, and through centralized monitoring and management, equipment failures and abnormal behaviors can be discovered and handled in a timely manner, thereby improving the operating efficiency and reliability of the wind farm.

[0028] S104: Determine whether dispatching is currently required based on the current abnormal situation of the wind farm.

[0029] The method provided by the present invention mainly includes establishing a wind farm abnormality diagnosis model, outputting the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; judging whether scheduling is needed at present according to the abnormal situation of the current wind farm. The method provided by the present invention greatly reduces the dependence on manual inspection and improves work efficiency to a certain extent. Secondly, the model provided by the present invention can detect and warn the faults of the wind farm in real time, and issue scheduling instructions when manual processing is required to improve the stability of the wind farm.

[0030] Preferably, the historical data preprocessing includes normalizing and extracting features from the historical data to obtain a feature subset of the historical data, and dividing the data set into a training set and a validation set in proportion, including: Setting a relationship function between a feature set and a feature subset, wherein the relationship function is used to represent an abnormal situation of a current wind farm; Calculate the correlation of all feature subsets in the feature set through the relationship function, extract the features with the greatest correlation and add them to the feature subset; The remaining features in the feature set are calculated with the features with the highest correlation obtained in the previous calculation respectively, and the features with the highest correlation each time are extracted and added to the feature subset until the upper limit of the feature subset is reached, and the feature subset is output.

[0031] Preferably, the relationship function between the feature set and the feature subset is set as follows:

[0032] The remaining features in the feature set are calculated with the features with the greatest correlation obtained in the last calculation, including:

[0033] In the formula, is the feature with the greatest correlation, is the number of features, For the Features, For the The eigenvalues ​​of the features, It is the feature calculation function with the greatest correlation between this feature and the feature calculated last time. is the number of remaining features, is the first of the remaining features Features.

[0034] Preferably, the establishing of the wind farm abnormality diagnosis model comprises: Get the feature subset and combine it with the data set to get the data set vector, establish the input layer, hidden layer and output layer respectively, and define the initial objective function; The initial objective function is trained with the training set to minimize the objective, and the learning rate is set. After each iteration, the weights and biases are replaced by the learning rate until the number of iterations is reached, and the optimized wind farm abnormality diagnosis model is output.

[0035] Preferably, the objective function includes:

[0036] In the formula, is the objective function value, is the activation function, is the regularization coefficient, For the samples, For the The output layer to the The connection weights of the hidden layers are is the output value of the hidden layer, is the weight coefficient, is the output value of the output layer, is the bias from the hidden layer to the output layer.

[0037] Preferably, the outputting of the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model includes: Output the abnormal situation of the current wind farm through the operation data of the current wind farm equipment and the optimized wind farm abnormality diagnosis model; Among them, the normalized samples are output for this model. The value range of is (-1, 1), so in this embodiment, the abnormal situation is judged as follows: , it is considered abnormal. , it is judged as not abnormal.

[0038] If the current wind farm is abnormal, dispatching will be carried out; if the current wind farm is normal, no dispatching will be carried out.

[0039] Preferably, the scheduling includes: Obtain maintenance station information and the current location of the wind farm, and set up a maintenance station evaluation index model; The maintenance sites are evaluated through the evaluation index model and the maintenance station information to obtain several evaluation values, and the maintenance site with the largest evaluation value is selected for scheduling maintenance as a scheduling instruction and sent.

[0040] In this embodiment, the location of the wind farm is generally large in area and there may be multiple maintenance stations. By setting an evaluation index model to determine the most suitable maintenance station, work efficiency is improved and abnormal problems of the current wind farm can be solved more accurately.

[0041] Preferably, the evaluation index model includes:

[0042] In the formula, is the number of maintenance station staff, is the number of equipment to be maintained at the maintenance station, is the distance between the maintenance station and the wind farm, is the area of ​​the maintenance station, It is the number of vehicles at the maintenance station.

[0043] Secondly, the evaluation indicators can be divided into multiple levels, from the overall to the details. The evaluation indicators can be divided into multiple levels such as basic indicators, intermediate indicators and comprehensive indicators, and the relationship and weight between each level can be determined.

[0044] In order to accurately evaluate the effect of the intelligent operation and maintenance management application of the centralized control center, it is necessary to determine the weight of each evaluation indicator. The weight can be determined based on expert evaluation, hierarchical analysis method, fuzzy comprehensive evaluation method and other methods. By reasonably allocating the weight of each indicator, the effect of the intelligent operation and maintenance management application of the centralized control center can be more accurately reflected.

[0045] According to the determined evaluation indicators and weights, a mathematical model of the evaluation indicator system is established. The evaluation model can be established using methods such as hierarchical analysis, fuzzy mathematics, and weighted linear comprehensive scoring. Through the establishment of a mathematical model, each evaluation indicator can be quantified and comprehensively evaluated.

[0046] In a second aspect, the present invention further provides a wind farm operation and maintenance management system, comprising: An acquisition module is configured to acquire historical data of wind farm equipment, preprocess the historical data, and obtain a training set and a verification set of the historical data; The model building module is configured to build a wind farm abnormality diagnosis model, train the wind farm abnormality diagnosis model through a training set, verify the wind farm abnormality diagnosis model after multiple trainings through a verification set, and output an optimal wind farm abnormality diagnosis model; The processing module is configured to obtain the operating data of the current wind farm equipment, output the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; and determine whether scheduling is currently required according to the abnormal situation of the current wind farm; The main control device is connected with the acquisition module, the model building module and the processing module, and is used to execute the above-mentioned wind farm operation and maintenance management method.

[0047] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0048] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0049] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A wind farm operation and maintenance management method, characterized in that: include: Obtain historical data of wind farm equipment, pre-process the historical data, and obtain training sets and verification sets of the historical data; Establish a wind farm anomaly diagnosis model, train the wind farm anomaly diagnosis model through the training set, and verify the wind farm anomaly diagnosis model after multiple trainings through the verification set, and output the optimal wind farm anomaly diagnosis model; Obtain the operating data of the current wind farm equipment and output the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; Determine whether dispatching is needed based on the current abnormal situation of the wind farm.

2. A wind farm operation and maintenance management method according to claim 1, characterized in that: The historical data preprocessing includes normalizing and extracting features from the historical data to obtain feature subsets of the historical data, and dividing the data set into a training set and a validation set in proportion, including: Setting a relationship function between a feature set and a feature subset, wherein the relationship function is used to represent an abnormal situation of a current wind farm; Calculate the correlation of all feature subsets in the feature set through the relationship function, extract the features with the greatest correlation and add them to the feature subset; The remaining features in the feature set are calculated with the features with the highest correlation obtained in the previous calculation respectively, and the features with the highest correlation each time are extracted and added to the feature subset until the upper limit of the feature subset is reached, and the feature subset is output.

3. A wind farm operation and maintenance management method according to claim 2, characterized in that: The relationship function of setting the feature set and the feature subset includes: The remaining features in the feature set are calculated with the features with the greatest correlation obtained in the last calculation, including: In the formula, is the feature with the greatest correlation, is the number of features, For the Features, For the The eigenvalues ​​of the features, It is the feature calculation function with the greatest correlation between this feature and the feature calculated last time. is the number of remaining features, is the first of the remaining features Features.

4. A wind farm operation and maintenance management method according to claim 3, characterized in that: The establishment of the wind farm abnormality diagnosis model comprises: Get the feature subset and combine it with the data set to get the data set vector, establish the input layer, hidden layer and output layer respectively, and define the initial objective function; The initial objective function is trained with the training set to minimize the objective, and the learning rate is set. After each iteration, the weights and biases are replaced by the learning rate until the number of iterations is reached, and the optimized wind farm abnormality diagnosis model is output.

5. A wind farm operation and maintenance management method according to claim 4, characterized in that: The objective function includes: In the formula, is the objective function value, is the activation function, is the regularization coefficient, For the samples, For the The output layer to the The connection weights of the hidden layers are is the output value of the hidden layer, is the weight coefficient, is the output value of the output layer, is the bias from the hidden layer to the output layer.

6. A wind farm operation and maintenance management method according to claim 5, characterized in that: The outputting of the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model includes: Output the abnormal situation of the current wind farm through the operation data of the current wind farm equipment and the optimized wind farm abnormality diagnosis model; If the current wind farm is abnormal, dispatching will be carried out; if the current wind farm is normal, no dispatching will be carried out.

7. A wind farm operation and maintenance management method according to claim 6, characterized in that: The scheduling comprises: Obtain maintenance station information and the current location of the wind farm, and set up a maintenance station evaluation index model; The maintenance sites are evaluated through the evaluation index model and the maintenance station information to obtain several evaluation values, and the maintenance site with the largest evaluation value is selected for scheduling maintenance as a scheduling instruction and sent.

8. A wind farm operation and maintenance management method according to claim 7, characterized in that: The evaluation index model includes: In the formula, is the number of maintenance station staff, is the number of maintenance equipment in the maintenance station, is the distance between the maintenance station and the wind farm, is the value of the maintenance station area, It is the number of vehicles at the maintenance station.

9. A wind farm operation and maintenance management system, characterized in that: include: An acquisition module is configured to acquire historical data of wind farm equipment, preprocess the historical data, and obtain a training set and a verification set of the historical data; The model building module is configured to build a wind farm abnormality diagnosis model, train the wind farm abnormality diagnosis model through a training set, verify the wind farm abnormality diagnosis model after multiple trainings through a verification set, and output an optimal wind farm abnormality diagnosis model; The processing module is configured to obtain the operating data of the current wind farm equipment, output the abnormal situation of the current wind farm through the wind farm abnormality diagnosis model; and determine whether scheduling is currently required according to the abnormal situation of the current wind farm; A main control device is connected to the acquisition module, the model building module and the processing module, and is used to execute a wind farm operation and maintenance management method as described in any one of claims 1-8.

Images