Lithium-ion battery SOC estimation under data loss and update stagnation

With the rapid development of electric vehicle technology, battery management systems (BMS) play a crucial role in estimating the state of charge (SOC) of the battery. However, in practical applications, battery data is often affected by issues such as loose connectors, leading to data loss and update stagnation, which results in decreased estimation accuracy. This study, based on the second-order RC circuit model, thoroughly analyzes the impact of data loss and update stagnation on battery SOC estimation, and experimentally verifies that voltage data loss has the most significant effect on battery parameter identification and SOC estimation. To address this issue, a voltage data reconstruction method based on a backpropagation neural network (BPNN) is proposed, and the reconstructed data is input into the simultaneous input and state estimation (SISE) algorithm for SOC estimation. The proposed method effectively reduces the errors caused by missing voltage data, significantly improving the accuracy of SOC estimation, with errors controlled within 1 %. Comparisons were made with the current mainstream methods, and the proposed methods all outperformed the comparison methods with higher accuracy. The approach provides an effective solution for battery management systems to cope with data loss and update stagnation, offering strong practical application value.