Energy storage power station fault prediction method based on large model

Abstract

The embodiment of the invention provides an energy storage power station fault prediction method based on a large model, and the method comprises the steps: obtaining the operation data of various batteries in a plurality of energy storage power stations, and carrying out the preprocessing operation of the data; performing power station operation fault feature extraction on the preprocessed operation parameters of the energy storage power station, and performing model training on feature extraction data to obtain a pre-trained large model; and then obtaining historical operation data and prediction data of a to-be-predicted target energy storage power station to carry out fine tuning on the pre-trained large model, wherein the fine-tuned small model can be used for fault diagnosis of the target energy storage power station.

Description

Technical Field

[0001] The present application relates to the field of information technology, and in particular to a method for predicting energy storage power station failures based on a large model. Background Art

[0002] In the wave of global energy transformation, promoting low-carbon energy structure has become the core goal of achieving sustainable development, and the rise of smart grids marks that the power system is moving towards a new era of intelligence and efficiency. In this transformation, the importance of energy storage systems as the cornerstone of smart grids has become increasingly prominent, stimulating the industry's unprecedented enthusiasm and exploration for energy storage technology. Energy storage is as important as power generation. Once an energy storage power station fails, it will also have a significant impact on the country's economy and security. Therefore, energy storage power station failure prediction is as important as power station failure prediction, but it has its own particularity; therefore, directly applying the power station failure prediction method has great inaccuracy. Summary of the invention

[0003] In order to improve the accuracy of energy storage power station fault prediction, the present application provides an energy storage power station fault prediction method based on a large model.

[0004] In order to achieve the above-mentioned purpose, an embodiment of the present application proposes a method for predicting energy storage power station faults based on a large model, comprising: obtaining operating data of various batteries in multiple energy storage power stations and performing preprocessing operations on the data; extracting power station operating fault features from the preprocessed operating parameters of the energy storage power station, and performing model training on the feature extraction data to obtain a pre-trained large model; then obtaining the historical operating data and prediction data based on the target energy storage power station to be predicted to fine-tune the pre-trained large model, and the fine-tuned small model can be used for fault diagnosis of the target energy storage power station.

[0005] The target energy storage power station data preprocessing operation includes: obtaining historical data such as current, voltage, fault information, etc. of the energy storage power station for a certain time length according to the time series, thereby obtaining voltage, current, and fault sequence data, and performing preprocessing operations based on various historical operating data and prediction data obtained from multiple energy storage power stations.

[0006] Among them, the method of adjusting network parameters based on the historical operation data and predicted data of the multiple energy storage power stations to obtain a pre-trained large model includes: extracting the features of each power station using the pre-processed operation parameters of the energy storage power station to extract the features of the power station operation faults, and using them as training samples; converting the historical data of the energy storage power station into a word embedding vector, and position encoding the word embedding vector, processing the vector representation of the word based on the attention mechanism, and obtaining the attention vector of each word; performing nonlinear transformation processing on the attention vector through a neural network to obtain a representation vector of each word of the fault data to be checked; inputting the predicted data information and the knowledge embedding feature vector into pre-training to obtain a pre-trained large model, and obtaining pre-trained data.

[0007] Among them, the target model is fine-tuned based on the historical operation data and the predicted data of the energy storage power station to be predicted, including: inputting the historical actual voltage, current, and fault information of the target energy storage power station into the large prediction model, performing forward iterative operations on the predicted fault information and the fault information, and updating the neural network parameters of the target prediction model through back propagation to obtain a fine-tuned small model until the small model converges.

[0008] The beneficial effects of the above technical solution of the present application are as follows: The technical solution proposed in the present application can accurately diagnose the faults of the energy storage power station and improve the maintenance efficiency of the energy storage power station. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic diagram of a method flow of an embodiment of the present application;

[0010] Figure 2 A schematic diagram of a method flow of an embodiment of the present application in one implementation mode;

[0011] Figure 3 The present invention is a flowchart of another embodiment of the present invention. DETAILED DESCRIPTION

[0012] In order to make the technical problems, technical solutions and advantages to be solved by the present application clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0013] The field of energy storage power station fault diagnosis is experiencing an unprecedented leap, and its development trajectory shows four distinct trends:

[0014] 1. The leap forward in sensor technology has given wings to the comprehensive monitoring of energy storage systems. High-precision and high-reliability sensors continue to emerge. They are like sharp sentinels, accurately capturing every subtle change in the energy storage process, ensuring all-round and all-round monitoring of the energy storage status, and providing solid data support for fault warning and diagnosis.

[0015] Second, the deep integration of big models and artificial intelligence technology has opened up a new path for intelligent diagnosis of energy storage faults. In particular, the application of large language models, with its powerful data processing and learning capabilities, has injected the soul of intelligence into the fault analysis of energy storage systems. These technologies can not only automatically identify fault modes, but also continuously optimize diagnostic strategies through deep learning to achieve accurate prediction and rapid response to faults.

[0016] 3. Model-based diagnostic methods are becoming increasingly mature and have become an important means to improve the reliability of energy storage systems. By building an accurate physical model or a virtual model using machine learning algorithms, combined with real-time operating data, it is possible to deeply analyze the internal mechanism of the energy storage system and accurately locate and diagnose the root cause of the fault. This method not only improves the accuracy of diagnosis, but also significantly shortens the troubleshooting time.

[0017] Based on this, Figure 1 As shown, the embodiment of the present disclosure proposes a method for predicting energy storage power station faults based on a large model, including: obtaining operating data of various batteries in multiple energy storage power stations, and preprocessing the data; extracting power station operating fault features from the preprocessed operating parameters of the energy storage power station, and training the feature extraction data to obtain a pre-trained large model; then obtaining the historical operating data and prediction data based on the target energy storage power station to be predicted to fine-tune the pre-trained large model, and the fine-tuned small model can be used for fault diagnosis of the target energy storage power station.

[0018] The target energy storage power station data preprocessing operation includes: obtaining historical data such as current, voltage, fault information, etc. of the energy storage power station for a certain time length according to the time series, thereby obtaining voltage, current, and fault sequence data, and performing preprocessing operations based on various historical operating data and prediction data obtained from multiple energy storage power stations.

[0019] like Figure 2 As shown, the method of obtaining a pre-trained large model by adjusting network parameters based on the historical operating data and predicted data of the multiple energy storage power stations includes: extracting the features of each power station using the pre-processed operating parameters of the energy storage power station to extract the features of the power station operation faults, and using them as training samples; converting the historical data of the energy storage power station into a word embedding vector, and position encoding the word embedding vector, processing the vector representation of the word based on the attention mechanism, and obtaining the attention vector of each word; performing nonlinear transformation processing on the attention vector through a neural network to obtain a representation vector of each word of the fault data to be checked; inputting the predicted data information and the knowledge embedding feature vector into the pre-training to obtain a pre-trained large model, and obtaining pre-trained data.

[0020] like Figure 3As shown, the target model is fine-tuned based on the historical operating data of the energy storage power station to be predicted and the predicted data, including: inputting the historical actual voltage, current, and fault information of the target energy storage power station into the large prediction model, performing forward iterative operations on the predicted fault information and the fault information, and updating the neural network parameters of the target prediction model through back propagation to obtain a fine-tuned small model until the small model converges.

[0021] The technical solution proposed in the embodiment of the present disclosure is not only applicable to the fault prediction of energy storage power stations, but also applicable to the prediction of major components such as transformers in energy storage power stations. Since the method is simple to implement and has high prediction accuracy, it has broad engineering application prospects.

[0022] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs. When the computer program is loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage medium such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0023] A person skilled in the art may understand that the various numerical numbers such as first and second involved in the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application, and also indicate the order of precedence.

[0024] At least one in the present application can also be described as one or more, and a plurality can be two, three, four or more, which is not limited in the present application. In the embodiments of the present application, for a technical feature, the technical features in the technical feature are distinguished by "first", "second", "third", "A", "B", "C" and "D", etc., and there is no order of precedence or size between the technical features described by the "first", "second", "third", "A", "B", "C" and "D".

[0025] The corresponding relationships shown in each table in the present application can be configured or predefined. The values ​​of the information in each table are only examples and can be configured as other values, which are not limited by the present application. When configuring the corresponding relationship between the information and each parameter, it is not necessarily required to configure all the corresponding relationships illustrated in each table. For example, in the table in the present application, the corresponding relationships shown in some rows may not be configured. For another example, appropriate deformation adjustments can be made based on the above table, such as splitting, merging, etc. The names of the parameters shown in the titles of the above tables can also use other names that can be understood by the communication device, and the values ​​or representations of the parameters can also be other values ​​or representations that can be understood by the communication device. When implementing the above tables, other data structures can also be used, such as arrays, queues, containers, stacks, linear lists, pointers, linked lists, trees, graphs, structures, classes, heaps, hash tables or hash tables.

[0026] The predefined in the present application may be understood as defined, predefined, stored, pre-stored, pre-negotiated, pre-configured, solidified, or pre-burned.

[0027] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0028] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0029] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for predicting energy storage power station failure based on a large model, characterized in that: include: Obtain the operating data of various batteries in multiple energy storage power stations and perform pre-processing operations on the data; The preprocessed operating parameters of the energy storage power station are used to extract the power station operation fault features, and the feature extraction data is used for model training to obtain a pre-trained large model; then the historical operating data and prediction data based on the target energy storage power station to be predicted are obtained to fine-tune the pre-trained large model, and the fine-tuned small model can be used for fault diagnosis of the target energy storage power station.

2. The method for predicting energy storage power station failure based on a large model according to claim 1, wherein: The target energy storage power station data preprocessing operation includes: obtaining historical data such as current, voltage, fault information, etc. of the energy storage power station for a certain time length according to the time series, thereby obtaining voltage, current, and fault sequence data, and performing preprocessing operations based on obtaining various historical operating data and prediction data from multiple energy storage power stations.

3. The method for predicting energy storage power station failure based on a large model according to claim 1, wherein: The method for obtaining a pre-trained large model by adjusting network parameters based on the historical operating data and predicted data of the multiple energy storage power stations includes: extracting the features of each power station using the pre-processed operating parameters of the energy storage power station to extract the features of the power station operation faults, and using them as training samples; converting the historical data of the energy storage power station into a word embedding vector, positionally encoding the word embedding vector, processing the vector representation of the word based on the attention mechanism, and obtaining the attention vector of each word; performing nonlinear transformation processing on the attention vector through a neural network to obtain a representation vector of each word of the fault data to be checked; inputting the predicted data information and the knowledge embedding feature vector into pre-training to obtain a pre-trained large model, and obtaining pre-trained data.

4. The method for predicting energy storage power station failure based on a large model according to claim 1, wherein: The target model is fine-tuned based on the historical operating data and the predicted data of the energy storage power station to be predicted, including: inputting the historical actual voltage, current, and fault information of the target energy storage power station into the large prediction model, performing forward iterative operations on the predicted fault information and the fault information, and updating the neural network parameters of the target prediction model through back propagation to obtain a fine-tuned small model until the small model converges.

Images