With the rapid development of new energy vehicles, the echelon utilization of retired power battery has become an important factor to promote the healthy development of this industry, while the Remaining Useful Life (RUL), as the key reference factor for the echelon utilization of retired power battery, has attracted the attention and research of many scholars in recent years. At present, most prediction methods are based on off-line data, which cannot process real-time data in time, so it is difficult to realize online prediction of RUL. In order to realize the real-time online monitoring and high-precision calculation of lithium-ion battery RUL, this paper proposes a lithium-ion battery RUL prediction method based on data-driven and multi-model fusion.
The one-dimensional Convolutional Neural Network (1D_CNN) is used for fast online feature extraction of one-dimensional battery capacity time series data to mine potential hidden information. Then the size of the amount of information retained in before and after data is controlled through the Long and Short-term Memory neural network (LSTM), so as to solve the problem of long-dependent characteristics in time series data and efficiently identify the data mode. Using the keras platform to build a 1D_CNN-LSTM deep learning model, and use multiple data sets for training. Finally, the Bayesian Model Average (BMA) algorithm with the ability to express uncertainty is used to fuse the trained models according to the calculated weight coefficients to obtain a predictive aging model. Then calculate the remaining service life according to the capacity failure threshold. This method not only makes up for the lack of uncertainty expression in the deep learning model, but at the same time the fusion model combines the merits of the two neural networks, and further improves the prediction accuracy of the lithium-ion battery RUL. Experiments are conducted using the lithium-ion battery aging data sets published by NASA, and the prediction accuracy obtained is greatly improved compared with other single model predictions.