Research And Development For The Data Real-time Processing System In Battery Test Instrument

Facing the severe situation and arduous challenges of global energy shortage and environmental pollution,electric vehicles(EVs)have attracted wide attention because of their excellent effect of energy saving and emission reduction.Many countries in the world have raised the development of electric vehicles to the height of national strategy,making the electric vehicles achieve blowout development.Power batteries are the energy source of electric vehicles,whose parameters and performance indexes come from battery testing instruments.The authenticity of test data greatly affects the performance evaluation of power battery,such as state estimation and residual life prediction.However,the large amount of data,transmission congestion and strong interference and poor in battery testing process present a severe challenge to the data real-time processing system in battery testing instruments.The purpose of this paper is to develop a high-performance data real-time processing system for battery testing instruments,which can realize high-speed and reliable real-time processing of battery test data.Besides,it can realize the coordinated control of the PC,the power conversion system,the data acquisition system and the touch screen in battery test instruments to ensure the authenticity of test data.The main contents of this paper are as follows:First of all,the function and performance requirements of real-time data processing system for battery testing instruments are analyzed in detail.The overall functional frameworks of data real-time processing system are proposed and the engineering implementation plans are formulated.Secondly,high-performance multi-core controllers are selected and functional circuits are designed.The structure and working principle of these circuits are described.These circuits are used to realize Ethernet communication,serial communication,CAN communication,level conversion and drive,power off storage,simulation debugging and system power supply.And then,an IP(Intellectual Property)core integration platform is constructed by using Vivado integrated development environment based on the designed hardware circuit.The software of each processor core in heterogeneous multi-core architecture is programmed and implemented,and the software workflow is analyzed concretely.Next,the Thevenin equivalent circuit model of lithium-ion batteries and the online parameter identification algorithm of the Recursive Least Squares with Forget Factor(FFRLS)with forgetting factor are studied by using the battery data obtained by the battery test instrument.The state of charge algorithm based on extended Kalman filter and PI observer are studied respecti vely.Furthermore,several condition tests of ternarymaterial lithium-ion batteries was completed,and the research results were compared and analyzed.The experimental results show that the state of charge estimation algorithm based on PI observer algorithm has more advantages over the state of charge estimation algorithm based on extended Kalman filter in convergence speed,accuracy,complexity and real-time.Finally,a battery test platform is built and tested.The results show that the designed data real-time processing system achieves the expected functions and performance indicators,and has high practical value in engineering field.