遗传和BP算法优化与混合的驱动控制  

Research on driving control based on genetic and BP optimization and mixing

作　　者：胡丽丽 陶俊才[1] HU Lili;TAO Juncai(School of information engineering,Nanchang University,Nanchang 330031,China)

机构地区：[1]南昌大学信息工程学院,江西南昌330031

出　　处：《南昌大学学报（理科版）》2017年第6期585-590,共6页Journal of Nanchang University(Natural Science)

基　　金：国家自然科学基金资助项目(61262049)

摘　　要：作为新能源应用的场地电动车面对的工况复杂,路线长,负载多变。为了提高它们的响应速度与抗干扰能力,降低转矩脉动,延长电动车续驶里程,提出一种优化遗传神经元网络混合算法(IGA-IBP),基于该算法设计参数自学习PID控制器应用于该电动车驱动系统,相较于基于传统GA-BP算法的PID控制器,不仅电动车的速度动作响应更快,抗扰能力更强,电机转矩脉动更小,驾驶噪音更低,而且起动过程节能,延长了电动车续使里程。IGAIBP算法参数设计容易,适应性强,具有一定的理论意义和工程应用价值。The load of electric vehicles used for sites is changeable,and the traffic environment is complex.In order to improve the response speed of the electric vehicle and anti-jamming capability,an improved genetic BP neural Network(IGA-IBP)algorithm is proposed to reduce the torque ripple and extend the travel distance of the electric vehicles.Based on the proposed algorithm,a parameter self-learning PID controller is designed and applied to the driving system of the electric vehicle.The analysis results show that the IGA-IBP algorithm,compared with the traditional GA-BP,has the advantages that the electric vehicle has faster speed response,more anti-disturbance ability,and smaller motor torque ripple.Therefore,our algorithm can reduce the vehicle driving noise,increase the driving comfort,and extend the travel distance of the electric vehicles,thus playing an important role in the promotion and application of the electric field.

关 键 词：场地电动车 驱动控制 遗传算法 神经元网络 优化与混合 

分 类 号：TP276[自动化与计算机技术—检测技术与自动化装置] TM361A[自动化与计算机技术—控制科学与工程]