Power Tracking of the Hybrid Vehicle Based on Hidden Markov Model

Abstract:

Power tracking is the foundation of power allocation of hybrid vehicles, the error of prediction in power tracking is also influent a lot on the steady of battery’s state of charge (SOC ) level. In this paper, driving modes of all-terrain robot have been discussed, a new power tracking model based on Hidden Markov model (HMM) was proposed, in this model the working mode is divided into acceleration mode, deceleration mode, even pace mode, mutation acceleration mode and mutation deceleration mode due to the velocity variation, in each mode we build different control strategy for the battery management, such as regenerative braking or engine-generator power unit (IGPU) export power, and those strategy would keep the SOC in a constant interval, in which the battery has a excellent charge-discharge properties. A power prediction model based on HMM has also been set up due to the historical data. A simulation which the velocity curve was simplified as a sine-wave has been made, and the true power curve and predicted curve were compared, the error of the prediction is allowable.

• Periodical: Applied Mechanics and Materials (Volumes 278-280)
• Pages: 40-43
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.278-280.40
• Online since: January 2013
• Authors: Shu Fang, Lin Xue, Li Jing Wang
• Keywords: Battery Management System, HMM, Operation Mode Choosing, Power Tracking, SoC

References

[1] Wang Junping, Cao Binggang, Chen Quanshi. Combined state of charge estimator for electric vehicle battery pack, Control Engineering Practice, 2007, 12: 1569-1576.

DOI: 10.1016/j.conengprac.2007.03.004

[2] Mark Verbrugge，Edward Tate．Adaptive state of charge algorithm for nickel metal hydride batteries including hysteresis phenomena[J]. Journal of Power Sources, 2004, 126(1—2)：236—249.

DOI: 10.1016/j.jpowsour.2003.08.042

[3] Rudi Kaiser, Optimized battery management system to improve storage lifetime in renewable energy systems[J]. Journal ofPower Sources, 2007, 168: 58-65.

DOI: 10.1016/j.jpowsour.2006.12.024

[4] Noboru Sato, Kazuhiko Yagi. The behavior analysis of nickel metal hydride batteries for electric vehicles[J]. JSAE Review 2000, 21: 205-211.

DOI: 10.1016/s0389-4304(99)00096-x

[5] Information on http: /forum. simwe. com/archiver/tid-451394. html.

[6] COHEN L．A primer on time-frequency analysis time-frequency signal analysis methods and applications．New York, Wiley Halsted Press, 1992, in press.

[7] QIAN JIANG, SU JIANBO, Online estimation of image Jaeobian matrix by Kalman-Bucy filter for uncalibrated stereo vision feedback[c]Proceedings of IEEE International Conference on Robotics and Automation. New York: IEEE, 2002: 562—567.

DOI: 10.1109/robot.2002.1013418

[8] ZHAO QINGJIE, WANG FANG, SUN ZENGQI. Using neural network technique in vision—based robot curve tracking[J], Pro-ceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems. New York: IEEE, 2006: 3817-3822.

DOI: 10.1109/iros.2006.281787

[9] BAO Ya-ping, ZHENG Jun, WU Xiao-guan. Speech Recognition Based on a Hybrid Model of H idden Markov Models and the Genetic Algorithm Neural Network. COMPUTER ENGINEERING＆SCIENCE. vol. 33. No4, (2011).

[10] Kutkut N H. Wiegman H L N, Divan D M, et aI. Design consideration for charge equalization of electric vehicle battery system[J]. IEEE Transactions on Industry Application. (1999).

DOI: 10.1109/28.740842