An IMFO-LSTM_BIGRU combined network for long-term multiple battery states prediction for electric vehicles.

Predicting the state of power batteries is crucial for ensuring the safe and reliable operation of electric vehicles. However, accurately forecasting multiple battery parameters under full operating conditions remains challenging. Traditional prediction models struggle to accurately represent the actual battery behavior due to irregular vehicle driving patterns. Furthermore, long-term prediction of battery states, essential for potential fault diagnosis, poses significant difficulties for conventional neural networks. To address the issue of error accumulation in recurrent neural networks (RNNs) during multi-parameter battery prediction across various operating conditions, we propose a combined model featuring multi-level feature extraction. This model integrates Long Short-Term Memory (LSTM) and Bidirectional Gated Recurrent Unit (BiGRU) networks, where the LSTM layer employs forget gates to filter out unnecessary data while preserving valuable information. The bidirectional update gate in the BiGRU layer effectively combines historical and future data to enhance time series modeling. This comprehensive approach improves sequence modeling efficiency, thereby increasing reliability. However, the multi-level feature extraction structure introduces higher-dimensional hyperparameters, complicating the optimization process. To maintain prediction accuracy across multiple battery parameters, we propose an improved moth-flame optimization (IMFO) algorithm to optimize the complex hyperparameters of the LSTM_BiGRU model. Extensive experiments conducted on a real-world vehicle dataset demonstrate that the IMFO-LSTM_BiGRU combined network surpasses existing state-of-the-art methods in terms of accuracy and stability. • We introduce a LSTM_BIGRU combined network that integrates the LSTM network with BIGRU network. We use LSTM network as a front-end unit, and transform the complex spatiotemporal evolution of batteries with bus working conditions into analysis of time series, and then BIGRU is used to study the relevant characteristics of time series, improving the feature extraction of various values of power batteries, including the total voltage, the total current and the Maximum Single-Cell Voltage enhancing the network's capability to process long-term time sequences during all the working conditions of the vehicle. • We propose an improved MFO (IMFO) algorithm for self-tuning hyperparameters of the LSTM_BIGRU combined network to further improve the prediction accuracy. The IMFO algorithm includes a chaotic initialization mechanism, non-uniform mutation operators, and an adaptive inertial weight mechanism, which helps effectively and accurately determine the optimal hyperparameters. • The training and testing datasets employ long-duration, continuous and real-world data from multiple vehicles, which prove the long-term prediction reliability and robustness of our method, and provide a practical foundation for the application of our method. [ABSTRACT FROM AUTHOR]