State Of Charge(SOC)estimation Of Lithium-Ion Batteries Considering Temperature Effects

With the warming of the global climate and the further increase of the energy crisis,in order to protect and save energy,new energy vehicles have been developed rapidly in recent years.As the main power source of electric vehicles,the current battery state of charge estimation plays a vital role in the Battery Management System(BMS.The State of Charge(SOC)of a battery is the ratio of its current capacity to its rated capacity.Accurately estimating SOC can prolong battery life and ensure safe operation of the battery.Due to the influence of temperature on SOC estimation,this paper uses ternary lithium batteries as the research object for model establishment and parameter identification.Design SOC estimation algorithms under different temperatures and operating conditions,and conduct in-depth analysis in terms of estimation accuracy,convergence speed,and robustness.This dissertation is organized as follows:(1)Based on the internal structure and mechanism of lithium batteries,this paper uses a second-order RC equivalent circuit model to simulate the internal characteristics of the battery.Multiple experiments have been conducted to obtain the SOC-OCV curves at different temperature,and the discrete state space equation of the system has been established.The pulse discharge experimental method was used for offline identification,and the accuracy of the parameters obtained from offline identification was verified.Lithium battery Parameter identification lay the foundation for subsequent SOC estimation of lithium battery.(2)Considering that traditional Kalman filtering algorithms ignore the impact of historical data on the current state,resulting in increased errors during the iteration process.This paper proposes an exponentially weighted multi-innovation Cubature Kalman Filter(EWMI-CKF),which combines the multi-innovation principle with the Cubature Kalman Filter,expands the innovation scalar into an innovation vector,and exponentially weights the new innovation vector.Compared to traditional linear weighted Cubature Kalman Filter and inverse proportional weighted Cubature Kalman Filter,EWMI-CKF has significant improvement in estimation accuracy.To further validate the robustness of the algorithm,EWMI-CKF was compared with traditional EKF,UKF,and CKF at different temperatures,operating conditions,and sensor errors.(3)Due to the weak resistance of traditional Cubature Kalman Filter to noise interference,while the H_∞Filter maintains strong robustness under noise interference.By combining Cubature Kalman Filter with HIF and incorporating the Sage-Husa adaptive principle,an adaptive hybrid Cubature Kalman/H_∞algorithm(ACHF)is proposed.Analyze the impact of different γ-values on SOC estimation under noise interference,and then analyze the robustness of the algorithm under different initial SOC values.Finally,ACHF,ACKF and H_∞are tested at different temperatures and operating conditions.Simulation results show that ACHF has high estimation accuracy and strong robustness.