The Prediction Of State-of-charge For Lead-Acid Battery Based On ANFIS Model

With fuel energy drying up and environment deterioration,EVs(electric vehicles) are comprehensively seen as the main transpoartation tools in the future.Thus, many countries as well as automotive companies support and attach great significance to the research and development of EV. China has also realized the great meaning of developing EV and has put the research and development of EV on the key science and technology program list in the eighth,the ninth and the tenth Five-Plan.As one of the EV development relevant technologies,the research for State Of Charge(SOC) of battery for EV has come to a practical period.The research and use of SOC system can prolong the life of battery effectively.SOC system can not only avoid the overcharge or overdischarge of battery,but also play as a important role as auto gas instrument.This paper has analyzed and compared some achieved results of SOC prediction algorithm and related system implemtention in the world.A SOC prediction model with high accuracy and strong generalization has been presented and designed.The system principle and implementation approach as well as relevant technical platform has been introduced in details in this paper.At the beginning of the project,we designed a data-acquisition system for key parameters of lead-acid battery. Based on this platform,some relevant battery experiments were made to obtain battery data for training and testing SOC prediction model.With PIC18F258 MCU integrating CAN-bus controller interface, this data-acquisition system with CAN bus communication technology is controlled by PC program.Finally,based on the above battery experiments,some simulation for ANFIS model of SOC prediction has been made and the corresponding results are presented in the paper...