Safety warning analysis for power battery packs in electric vehicles with running data

The Safety warning of battery packs can effectively prevent thermal runaway accidents in electric vehicles. The inconsistency evaluating of the battery pack accurately is a prerequisite for safety warning. In this work, the safety warning model for electric vehicles (EVs) power battery packs based on operational data is proposed, where the voltage, temperature, internal resistance, and electric quantity are extracted from accident vehicles over two years as the four factors for consistency evaluation of battery packs, and their changes during vehicle operation and before thermal runaway are analyzed. The Fuzzy analytic hierarchy process (FAHP) and the entropy weighting method are used to assign weights to the four factors, and the weighting coefficients of the subjective and objective assignment methods are determined to ensure accurate assignment of weights through the game theory approach. Each factor is scored comprehensively using the weighted scoring criteria, and the consistency status of the battery pack is determined based on the scoring results. The lowest consistency score of the single battery can be discerned by analyzing the real accident vehicle data through the distribution cloud map. The results reveal that the evaluation system can accurately quantify the degree of inconsistency of battery packs and identify problematic single cells timely, which is able to provide a reference for safety warning in the electric vehicles field. • The data obtained and analyzed are real driving data. • The changes during vehicle operation and before thermal runaway are analyzed by four evaluation factors. • The weights of the evaluation factors are calculated in two separate subjective and objective ways. • The game theory algorithm is used to determine the weighting coefficients to obtain the combination weights. • Safety warning for accident vehicles based on distribution cloud map.