Research On State Of Charge Estimation Algorithm Of Power Battery In Electric Vehicle And Hardware Implementation

Nowadays, vehicle is becoming more and more popular, but the pollution, noise and energy crisis brought by gas-powered vehicle also increasingly influence lives of human beings. Electric vehicle, with the advantage of energy saving and environmental protection, has attracted more and more attention, and is regarded as the development trend of automobile industry. The world's major companies, universities and institutes have carried on a thorough research on this problem and have made some remarkable progress. However, the development of electric vehicle power battery and its applications is slow and become the bottleneck of the popularization of electric vehicle. So electric vehicle battery management system (BMS) is absolutely necessary to monitor battery and improve lifetime and efficiency of the battery. The key functions of BMS, state of charge estimation (SOC), also influence the industrialization, commercialization of electric vehicle, thus, it is very important to study state of charge estimation.This paper research on state of charge estimation algorithm of power battery and the implementation of hardware. The concrete work as follows:(1)The research on estimation algorithm of state of charge based on extended karlman filterKalman filter technique has good effect on state estimation and is increasingly used for state of charge estimation. To estimate state of charge by kalman filter technique, it needs to analyze charge/discharge characteristic of battery, build equivalent circuit model, test out charge/discharge characteristic through charge/discharge experiment and fit the electric parameters of equivalent circuit model by using classical least squares. On the basis of ampere-hour integration approach and open-circuit voltage approach, State of charge estimation algorithm is studied in this dissertation with extended kalman filter technique. A simulation model is established on the platforms of Matlab/Simulink. Finally, the feasibility of the algorithm is verified through off-line simulation.(2)The hardware design of the estimation system for state of charge of battery After the hardware requirement analysis for state of charge estimation system, a hardware platform is established based on MC9S12DG128. To the hardware platform, signal-collecting circuit is needed to acquire voltage of battery group, charging-discharging current and working temperature, which are the inputs of the algorithm; external memory is added to save computational results and guarantee for the nonvolatile of the data when lost power. And LCD screen and CAN bus circuit are all designed in the hardware platform to display battery information and achieve communication. The total hardware platform adopts the module, integration and simple design concept. And electromagnetic interference of the circuit is fully considered to ensure the reliability of the hardware platform.(3)The software design of the estimation system for state of charge of batteryCorresponding software is designed to cooperate with hardware. Based on the data acquisitioning module, external memory module and LCD display module, capacity updating automatically, battery state judgment and electrical energy computation are realized by calling master control program repeatedly. The total software system adopts the module design concept too. In addition, software filter and watch-dog are designed in the software program to improve the anti-interference abilities of the software.(4)The hardware-in-loop test of the estimation system for state of charge of batteryIn this part, software program is written in the controller and the indoor discharge test is designed. This paper selects some discharge test to inspect and verify the accuracy of the algorithm, such as variational current test, square-wave test, and completely discharge test ect. The results show that the results of the estimation algorithms is accuracy, and the algorithm is feasible...