Modeling Of Large Capacity Battery Energy Storage System Based On Data Mining

Global warming and environmental pollution is becoming more and more serious,and clean energy sources has been the focus of attention of the community.In order to alleviate the instability of clean energy output,large capacity energy storage systems become an important part of the clean energy and network.Automatic monitoring of state of charge and the fault of large capacity battery energy storage system is an important way to promote the construction of smart grid.Therefore,it is necessary to study the state of charge and fault diagnosis of large capacity battery energy storage system.In response to this situation,data mining based on large number of historical data is analyzed.Aiming at state of charge evaluation and diagnosis of large capacity battery energy storage system.Main contents are as follows:Firstly,aiming at the evaluation of state of charge,the paper dicussed the equivalent circuit and data miming.In this paper,the SOC evaluation model is established based on the equivalent circuit and the historical data.In the aspect of equivalent circuit modeling,this paper using the cloud model to mine the typical curve of OCV-SOC.In the aspect of historical data model,this paper makes use of the MEA to improve the traditional BP neural network,and determines the optimal state evaluation strategy of SOC.Secondly,based on data mining,the fault data of large-capacity battery energy storage system and determines the typical characteristic vectors of penalty angle and deviation degree is analyzed.For the soft fault diagnosis of large-capacity battery energy storage system,the concepts of fuzzy mathematics and crack degree are introduced.The soft fault state is divided into fuzzy segments to improve the accuracy of fault diagnosis.Finally,combining the foregoing findings,the paper develops a SOC evaluation and fault diagnosis platform of large capacity battery energy storage system using Matlab GUI graphical interface,and compiles it as an independent executable file.The software has a good portability and interactivity.The software can be used for SOC status evaluation,fault feature extraction and soft fault diagnosis of large-capacity battery energy storage system.