The Design Of Monitoring Management System And Research Of Classifying Algorithm For Li-ion Battery Formation

In recent years,lithium battery becomes the focus of new energy research due to its outstanding performance in energy density,power density,cycle life and environmental protection.According to the actual demand for voltage and capacity in application,series-parallel cell battery is used as battery pack to provide energy.However,lithium battery series system often have a consistency problem,resulting in cycle life of battery pack is much lower than the single battery.This thesis chooses one kind of lithium-iron phosphate battery as the research object,a lithium battery monitoring and management system is proposed and designed.Through the intelligent classifying,consistency of lithium battery pack is improved,and its operating characteristics and cycle performance are enhanced.The main research contents are as follows:The charge-discharge characteristics and operating principle for lithium-ion ferrous phosphate battery(LFP battery)is described in this thesis.According to the basic principle and technical requirements of lithium battery formation,the whole frame of the lithium battery formation management system is designed based on the idea of distribution,hardware and software modules are divided.According to the related CAN communication technology are designed,the strategy and protocol are designed.The lithium battery monitoring and management system software design is completed.Through the analysis of software technology requirements,Visual Studio 2010 and SQL Server 2008 are used as the software development platform and database platform separately.Through the establishment of system E-R model,the database structure is builded,than it establishes a connection to the database by ADO interface access technology.Software platform is used to achieve the system login function,realtime monitoring of lithium battery data,storage analysis and other functions on software platform.The consistency between lithium batteries and intelligent classifying are studied,the lithium battery inconsistency is defined,including the reasons and the impact on the battery;than combined with the record in SQL Server database,research on classifying algorithm is designed in Matlab with the use of Fuzzy C clustering means algorithm,besides The FCM algorithm is improved based on genetic algorithm and Gaussian density weighted.Comparing the improved algorithm with FCM algorithm through test,the correct rate and speed of the battery classifying are enhanced,it confirms the effectiveness of the proposed algorithm.At last,the formation-detecting device is built to test the consistent classifying of lithium batteriy with the improved method on the experimental test and analysis.The functional modules of the system software platform are tested.Comparing the method which combines dynamic curve to the method only capacity and internal resistance parameters on the storage performance and cycle characteristics of lithium battery pack,the results verify the improvement on consistency and achieve the purpose of the research.