Research On The Screening And Restructuring Strategy Of Retired Power Batteries Used In Energy Storage Power Stations

Before the cascade utilization of retired power batteries,screening and classification,series and parallel reorganization are needed to improve the parameter consistency of retired power batteries and make them meet the requirements of cascade utilization.The problem of low classification accuracy and low screening efficiency is caused by the large number of retired power batteries and large differences in parameter consistency.In this thesis,a screening and reorganization method of retired power batteries applied to energy storage stations is proposed to reduce the parameter inconsistency of retired power batteries and improve the screening and classification efficiency.Based on the in-depth study of the basic characteristic parameters of retired power batteries and the causes of parameter inconsistency,the parameter inconsistency of retired power batteries is analyzed.The initial state of retired power batteries is evaluated to determine whether they have the value of cascade utilization.In addition,the relationship between the main characteristic parameters of retired power batteries is analyzed,and the parameter basis for screening and classification is determined.Based on the determination of screening classification parameters,a screening classification method for retired power batteries based on the improved Classification and Regression Tree(CART)algorithm is proposed.Firstly,the basic principle of CART algorithm is explained.Secondly,the selection of attribute threshold points of the algorithm is improved by combining the Fayyad boundary point determination theorem.Finally,the K-means clustering algorithm,CART algorithm,and improved CART algorithm are used for the screening and classification of retired power batteries,respectively.The simulation results show that compared with the other two classification algorithms,the improved CART algorithm significantly improves the screening efficiency while improving the consistency of retired power battery parameters.The voltage and capacity of retired power batteries are generally low individually,and they need to be reconfigured in series and parallel for use.In this thesis,the secondorder RC equivalent circuit model is used as the battery model,and the Forgetting Factor Recursive Least Squares(FFRLS)algorithm is used to identify the model parameters and verify the validity of the model.Based on this model,parallel-thenseries and series-then-parallel retired power battery packs were built,and the size and distribution characteristics of the reorganized capacity of the two packs were analyzed under different single cell arrangements to improve the overall utilization rate of retired power batteries.A hierarchical bi-directional active equalization control method for retired power batteries applied to energy storage power stations is proposed for the problem that the difference in the consistency of battery pack parameters after series-parallel reconfiguration of retired power batteries affects the energy utilization of the battery pack.The upper intra-group equalization is based on an isolated Cuk converter,and the lower inter-group equalization is based on a flyback transformer.The lower layer control strategy of neighboring non-neighboring parallel equalization and the upper layer control strategy of combining different intra-group equalization and inter-group equalization are used with the battery voltage as the equalization variable.Simulation results show that the proposed equalization method can improve the consistency of retired power battery packs and effectively shorten the equalization time.