Optimization Of Power Management In An Extended-range Electric Vehicle

Hybrid electric vehicle fuel consumption and exhaust emissions is largely dependent on the vehicle’s energy control systems. Scholars have been working on hybrid electric vehicle energy management strategy for years with some improvements, but there has development spaces still. Based on the traditional threshold control strategy, this paper conducted a further research for E-REV energy management strategies with different road cycles, according to the ideas of "system modeling, control algorithm design, control strategy optimization and online applications". This paper proposed an energy optimization control strategy based on road pattern recognition which improved the adaptation of control strategies. Works are as follows:First of all, to ensure traffic data is reliable and comprehensive, we use the road data during the Beijing Olympics by rechargeable fuel cell hybrid city buses as the sample database. Using K-means clustering method for cluster analysis of traffic database. Using SOM neural network for real-time road pattern recognition when driving.Secondly, Designed a suitable control strategy based on DP for E-REV, which solves optimization problems that mileage, beginning and ending SOC are fixed. Optimized the optimal energy allocation sequences for the typical road conditions, which can provide a reference for other optimization methods and the basis for online applications.To meet the requirements of controller’s computing performance and system security, this paper design a vehicle controller based on TC1782, which is the latest32-bit automotive microcontroller. The controller is according to the Modular design ideas. And constructs a complete safety certification system "hardware monitor chip+external+Safe-Tcore software" to achieve real-time monitoring to ensure the security and stability of the system.Finally, this paper have do the research about how to use the optimal operation of offline to online energy management. And proposes an online energy management strategy based on road pattern recognition.The simulation results show that the energy control strategy based on road recognition can effectively reduce fuel consumption.