Low-SOC Estimation of Aluminum-Air Battery Based on HSMM

The estimation accuracy of the battery state of charge (SOC) is one of the key factors that affect the performance of new energy vehicles. Owing to the large accumulated error, traditional ampere-hour method cannot meet the precise estimate of the aluminum-air battery SOC. In this paper, the approach based on hidden semi-Markov models (HSMM) of SOC prediction is applied as a complement for the ampere-hour method, making the latter estimation precision of aluminum air battery be guaranteed. Each of the different states of the model produces multiple sets of observations. According to the transition probability between the various states and the residence time, the model can more accurately predict the remaining time of each state. Through the experimental simulation and comparison with the single ampere-hour method, SOC estimation error combined with HSMM promotes the prediction accuracy when the battery is running out.