Research On Equilibrium System Based On State Of Charge Of Lithium Ion Battery Pack

As a new type of environmentally friendly transportation,pure electric vehicles have great development potential.However,there are still some deficiencies in battery management for pure electric vehicles.If the battery SOC estimation accuracy is relatively low,the capacity,voltage,internal resistance,etc.of the battery cells in the battery pack are inconsistent.The battery cells have inconsistencies in the manufacturing process,and due to differences in the use of the car,driving environment,driving conditions and driving habits,the inconsistency of the battery cells will be exacerbated.The voltage,internal resistance,capacity,etc.of each battery cell There are significant differences in the parameters,which may cause serious damage or even failure to the battery,and the damage to the large-capacity series battery pack is more obvious.The battery balancing system is a key technology for improving the performance of the battery pack and extending the life of the battery pack,and is also one of the main research directions of the battery management system.This article takes series lithium battery packs as the research object,takes battery state of charge(SOC)as the equilibrium goal,and uses reasonable models and algorithms to achieve accurate estimation of SOC.Aiming at the problems of slow balance time and large balance energy loss of the current balance system,study the balance topology with relatively simple balance control,fast balance speed,and high balance efficiency,and design the corresponding balance control strategy for the battery pack charging and discharging process,so that It can effectively improve the inconsistency between battery cells,increase the speed of equalization,reduce the energy loss in the process of equalization,and has strong practicality.Its main research work is as follows:(1)Selection of equilibrium variables.Firstly,the reasons for the inconsistency of lithium battery packs are analyzed,mainly due to the inconsistencies generated during production,use and storage.Then introduce the inconsistent manifestation of the battery pack in voltage,internal resistance,ohmic internal resistance and polarization internal resistance,SOC.Finally,through the analysis of the battery pack charge and discharge related test data,the important significance of the battery pack balance is obtained,and the SOC is used as the balance variable of the battery pack in different states.(2)Establish an improved Thevenin equivalent circuit model.Based on the Thevenin battery model and the modified ampere-hour integration algorithm,a state space expression is established for the battery model,and the system parameter identification of the equivalent circuit model is carried out through multiple HPPC cycle experiments to identify the parameters of Thevenin model.(3)Use the fuzzy neural network algorithm to modify the Kalman filter algorithm(FNN-EKF)to estimate the SOC of the lithium battery.Aiming at the SOC estimation problem of power battery pack,several commonly used SOC estimation methods are analyzed,and the FNN-EKF algorithm is designed to estimate the SOC of the power battery.First,the temperature and charge and discharge current are used as the input of the fuzzy controller,fuzzy control The output value of the device is the modified Kalman gain matrix,and then the three parameters of the modified Kalman gain matrix,the current data of the lithium battery during normal operation and the state estimation update error are used as the input of the BP neural network,and the output value is The difference between the true SOC value and the EKF filtered optimal estimate,and finally the difference is added to the EKF filtered optimal estimate,which is the optimal estimate based on FNN-EKF.(4)Design an active balanced topology based on inductanceThe principles and characteristics of common balanced topology structures are analyzed.At present,the main balanced topology structures are designed based on inductors and voltage transformers.By comparing with the transformer balanced topology structure,the active balanced topology structure of the inductor is finally selected and The working principle is introduced in detail,a balanced circuit simulation model is built,the structural parameters are adjusted according to the simulation results,and the circuit control parameters are determined.(5)Balance control strategy of k nearest neighbor genetic algorithmThe K nearest neighbor genetic algorithm is used to realize the balanced control of the battery pack,which is compared with the average equalization strategy.In MATLAB / Simulink,simulation experiments of three equalization working conditions of self-equalization,charging and discharging are completed.The research results show that the balancing strategy in this paper has a good balancing effect,which can effectively improve the inconsistency between the battery packs and shorten the balancing time of the battery pack.The balancing speeds under the three operating conditions are increased by 26.70%,22.70%,and 24.80%,respectively.;Reduce the number of switching actions and thus reduce the energy loss of the battery pack,and the energy conversion rate is increased by 5.06%.