Accurate estimation of the state-of-energy(SOE)in lithium-ion batteries is critical for optimal energy management and energy optimization in electric vehicles.However,the conventional recursive least squares(RLS)algor...Accurate estimation of the state-of-energy(SOE)in lithium-ion batteries is critical for optimal energy management and energy optimization in electric vehicles.However,the conventional recursive least squares(RLS)algorithm struggle to track changes in battery model parameters under dynamic conditions.To address this,a multi-timescale estimator is proposed.A variable forgetting factor RLS approach is used to determine the model parameters at a macro timescale,and the H infinity filter is utilized to estimate the SOE at a micro timescale.The proposed algorithm is verified and analyzed and shown to have accurate and robust identification of battery model parameters.Finally,experiments under dynamic cycles demonstrate that the proposed algorithm has a high level of accuracy for SOE estimation.展开更多
基金the financial support provided by the National Key R&D Program of China(Grant No.2020YFB1600605).
文摘Accurate estimation of the state-of-energy(SOE)in lithium-ion batteries is critical for optimal energy management and energy optimization in electric vehicles.However,the conventional recursive least squares(RLS)algorithm struggle to track changes in battery model parameters under dynamic conditions.To address this,a multi-timescale estimator is proposed.A variable forgetting factor RLS approach is used to determine the model parameters at a macro timescale,and the H infinity filter is utilized to estimate the SOE at a micro timescale.The proposed algorithm is verified and analyzed and shown to have accurate and robust identification of battery model parameters.Finally,experiments under dynamic cycles demonstrate that the proposed algorithm has a high level of accuracy for SOE estimation.
