State Of Charge And Capacity Estimation Of Battery Based On Intelligent Optimization Algorithm

The state of Charge(SOC)represents the relative remaining power of the battery,which characterizes the discharge capacity of power batteries.The available capacity of the battery is an important parameter to measure the aging state and energy storage capacity of the power battery.Accurate estimation of state of charge and available capacity is of great significance for battery management systems to achieve state monitoring and energy balance.In this paper,optimization algorithms,improved unscented Kalman filter(UKF)algorithm and improved robust extreme learning machine algorithm(ORELM)are used to identify model parameters and estimate battery SOC and battery capacity.The specific process is as follows:Aiming at the problem that the Antlion optimization algorithm applied to battery parameter identification has problems such as lower accuracy and slower convergence speed;a kind of parameter identification method based on improved Antlion algorithm is proposed.Aiming at the problem that the Antlion optimization algorithm has slow convergence speed and poor optimization results,IALO algorithm adopts chaotic Logistic mapping initialization,which is conducive to improving the initial population diversity of the algorithm and preventing the algorithm from falling into the local optimum at the beginning;adopts adaptive inertial weight and random Cauchy mutation strategy,which is conducive to improving the convergence rate of the algorithm;adopts the strategy of elite reverse learning to match the reverse group optimal value,which is conducive to improving the algorithm’s global search capabilities.The combination of the three improved methods is conducive to improving the convergence speed and optimization accuracy of the ALO algorithm.Applying IALO to the parameter identification of the third-order Thevenin equivalent circuit model obtains a better estimation error,which verifies the effectiveness of the IALO algorithm.Aiming at the problem that the battery use is accompanied by aging.As the battery life decreases,the effective capacity of the battery gradually decreases.The effective capacity of the battery is used as the reference value for calculating the initial value of SOC,and its decline with aging will affect the accuracy of SOC estimation.In this paper,A joint estimation method is proposed,which gives the open circuit voltage and the SOC and battery effective capacity Two-variable polynomial description of the nonlinear model;when the number of battery cycles exceeds the preset value,the whale optimization algorithm is used to estimate the current battery capacity and battery model parameters;according to the model parameters and capacity values,the battery SOC is estimated through WOA-UKF;in the SOC estimation process,Use the whale optimization algorithm to update the UKF’s observation noise variance and process noise variance to achieve adaptive adjustment of the noise variance,thereby improving the estimation accuracy.Aiming at the problem that the phenomenon of capacity recovery in the whole life cycle range makes modeling difficult;a scheme for estimating the capacity of lithium-ion batteries is proposed.The SSA-ORELM algorithm is used to establish a battery capacity prediction model based on the selected health factors.In the battery capacity estimation,the salvia optimization algorithm(SSA)is used to optimize the number of hidden layer nodes and the adjustment factor,which improves the accuracy and stability of the ORELM algorithm capacity estimation.Finally,the simulation verifies the effectiveness of the battery capacity estimation scheme.