A comparative analysis and validation for double-filters-based state of charge estimators using battery-in-the-loop approach

The state of charge (SOC) estimation is extremely important for the wide commercialization and safe operation of electric vehicle (EV), especially under cold conditions, which is also a critical technology for battery system in EVs used in the 2022 Beijing winter Olympics. Three efforts have been made in this paper: (1) A general joint estimation framework with dual estimators is set up. Based on this frame, a joint algorithm using the recursive least square (RLS) and the adaptive H infinity filter (AHIF) is realized. (2) Four filter-based algorithms have been systematically compared and analyzed at the wide temperature range. The results show that RLS-AHIF algorithm has better performance for SOC estimation even at low temperatures, such as -10 degrees C, and the SOC error is within 3.5%. (3) A hardware-in-loop validation platform including the battery management system (BMS) and battery test instruments has been built to verify the proposed method. The results from the platform show that the maximum error of SOC is less than 2% at 0 degrees C and 25 degrees C. Consequently, the proposed algorithm can achieve the application over a wide temperature range in an actual BMS.