Research On Life Prediction Method Of Lithium Battery In Unmanned Photovoltaic Charging Station

With the development of science and technology and the support of national policies,photovoltaic power generation has become a popular research direction in the new energy industry.Lithium battery,as a recyclable green energy storage unit,has significant advantages in weight,energy storage capacity,life,etc.It is an ideal energy storage module for unmanned photovoltaic charging stations.Accurate prediction of the life of lithium batteries can provide important health information for the operation of unmanned photovoltaic charging stations,reduce and avoid safety accidents caused by lithium batteries.In this paper,by studying the lithium battery life prediction method of unmanned photovoltaic charging stations,two methods of lithium battery life prediction are proposed,and the corresponding life prediction software system is designed to visually display the prediction results.The main work completed is as follows:1.Aiming at the problem that the monitoring data of the battery management system of unmanned photovoltaic charging stations are various,the selection of training data is complex,and it is difficult to accurately predict the life of the lithium battery,it is constructed using Support Vector Machine(SVM)and phase space reconstruction The prediction model reduces the complexity of data selection and improves the accuracy of life prediction.Aiming at the difficulty of SVM parameter selection,the improved particle swarm optimization(IPSO)algorithm is used to optimize the model parameters,and a phase space reconstruction IPSO-SVM lithium battery life prediction method is proposed,which can improve the overall Forecast accuracy.2.In practical engineering applications,it is difficult to obtain real-time battery capacity data and predict the life of the lithium battery online.Through analysis and extraction of the monitoring data of the battery management system of the unmanned photovoltaic charging station,a use is proposed Voltage drop discharge time,combined with Extreme Learning Machine(ELM)model for lithium battery life prediction method.For the unstable output of ELM prediction results,Immune Genetic Algorithm(IGA)is adopted to optimize the input weight and hidden layer threshold of ELM,which improves the stability of the output.3.Software system design is carried out for the two life prediction methods mentioned above.The system mainly includes three main modules:user login,data loading,and prediction method selection.The user can select the corresponding prediction method according to the actual situation,use the system to predict the life of the lithium battery,and visually display the prediction results so that the user can timely make effective use and maintenance decisions for unmanned photovoltaic charging stations.The experimental results show that the life prediction method of lithium battery in unmanned photovoltaic charging station can accurately and efficiently complete the life prediction of lithium battery.The designed life prediction system can not only accurately predict the life of lithium battery,but also visually present the prediction results,which meets the expected design requirements.