Remaining Useful Life Prediction Of Lithium-ion Battery Based On Deep Gaussian Process

As people's demand for energy grows day by day..Lithium ion battery is widely used in various industries because of its advantages such as stable energy output,long cycle life,clean and environmental protection,small size and light weight.However,lithium battery has the problem of performance degradation.As the energy supply unit of a whole system,once it causes failure due to degradation,it may lead to very serious consequences,endangering people's life and property safety.Therefore,it is very important for the safety assessment of lithium battery to predict the remaining life of lithium battery and make an early warning before its degradation to the failure threshold.In this thesis,the deep Gaussian process is applied to the prediction of the remaining life of lithium batteries(1)The main theoretical methods for the prediction of the remaining life of lithium battery are analyzed.Combined with the characteristics of the degradation curve of lithium battery,the deep Gaussian process is selected as the main method to predict the remaining life of lithium battery.(2)Based on the theory of deep Gaussian process,this thesis proposes a new method for NASA Two sets of Engineering lithium battery data of PCo E and University of Maryland CALCE are designed and established respectively by reasonably selecting network layers and kernel functions,and the corresponding deep Gaussian process model is used to predict the data,and the high prediction accuracy is obtained,which proves that the deep Gaussian process method has high feasibility in the field of lithium battery residual life prediction.(3)Compared with wavelet neural network,autoregressive moving average,long and short-term memory network,the deep Gaussian process method has some advantages in prediction accuracy and robustness.Through the above work,the feasibility and superiority of deep Gaussian process to predict the degradation data of different types of lithium batteries in the field of remaining life of lithium batteries are verified in this thesis,and provides guidance for the use of lithium batteries.