SOH Prediction of Lithium-ion Batteries Based on Automatic Feature Extraction and IWOA-SVR Model
The existing state of health(SOH)estimation model for lithium batteries has many input features,requires manual selection,and is computationally intensive,and a simplified prediction model that automatically extracts battery degradation features is proposed to solve these problems.The model uses the Whale Optimiza-tion Algorithm(WOA)to automatically find features as input,and then uses the improved whale algorithm to optimize the parameters of Support Vector Regression(SVR).The simulation results on the NASA lithium bat-tery dataset show that the model not only improves the prediction accuracy compared with BP neural network and other methods,but also simplifies the prediction process and can effectively avoid manual trial and error.Final-ly,the universality of the model is verified on the No.7 lithium battery.
Lithium batteryState of healthWhale optimization algorithmSupport vector machine regression
万方数据
维普
