Research On Charging Strategy Of Lithium-Ion Battery Based On Improved Particle Swarm Optimization By Natural Selection

As the energy and environmental problems are becoming increasingly severely,electric vehicles have developed rapidly due to their advantages of cleanliness and environmental protection.As an important energy source for electric vehicle driving,the research on lithium power battery has also received extensive attention.However,there are still many technical bottlenecks in the application of lithium-ion batteries in electric vehicles,such as safety and charging technology.Fast and effective lithium battery charging technology is very important for efficient driving,safe operation and long service life of electric vehicles.In this thesis,the common charging methods of lithium-ion batteries are reviewed.Not only the traditional constant current charging method but also the constant current and constant voltage charging method exists some problems such as slow charging speed and obvious charging temperature rise,which can not meet the demand of safe and rapid charging of lithium batteries in electric vehicles.Considering the important influence of charging rate and temperature on the charging process of lithium-ion battery,the electrical performance tests of lithium-ion battery under different conditions were carried out,which provided data support for the subsequent research on optimal charging strategy of lithium-ion battery.Firstly,the modeling method of electro-thermal coupling model of lithium-ion battery was studied.The parameters of electro-thermal model of lithium-ion battery were identified by HPPC test,and the accuracy of the model was verified by standard constant current and constant voltage charging experiment.The fitness function considering charging time and charging temperature rise is established,and the current of multi-stage constant current charging method is iteratively optimized by particle swarm optimization.Hence,the optimal multi-stage current value and corresponding multi-stage constant current charging strategy are obtained.The experimental and simulation results indicate that the proposed multi-stage constant current charging method needs shorter charging time than the traditional charging method,and the temperature rise has not been significantly improved.Secondly,committed to overcoming the shortcomings that the traditional particle swarm optimization algorithm has poor global searching capacity and is liable to sink into local optimization,this thesis proposes a charging strategy based on particle swarm optimization algorithm improved by natural selection.The simulation results indicate that fitness function which is improved by particle swarm optimization algorithm based on natural selection converges to a better value,and the number of iterations is less,which further improves the charging speed and has a better charging optimization effect on the premise of ensuring that the temperature rise does not increase significantly,and experimental results also verify this result.Finally,a fast charging device for lithium-ion batteries is designed,and charging verification experiments under different temperature environments are carried out to verify the effectiveness of the charging optimization method proposed in this thesis.The experimental results show that the improved multi-stage constant current optimization charging method achieves the design goal of improving charging speed and controlling charging temperature rise.The research in this thesis provides a train of thought for the research of lithium-ion battery charging strategy,and has certain positive significance in the rapid charging application of lithium-ion batteries in electric vehicles.