Research on Gaussian mixture model to predict Lithium battery life

Living in an era of social economy and energy storage technology, according to the historical monitoring data, the remaining Useful life(RUL) of the lithium battery to make predictions is important. In this paper, the data driven method is used to systematically analyze the influence of single kernel function and super parameters of kernel function on prediction distribution. On this basis, the combined kernel function is derived. The Gaussian process regression model based on the combined kernel function is used to predict the health status and remaining life of lithium batteries offline. Compared with the prediction results of a single kernel function, the Gaussian process regression based on the combined kernel function not only guarantees the local learning ability, but also improves the generalization ability and effectively improves the prediction accuracy.