基于RBF网络滑模的电动助力制动系统液压力控制

针对汽车电动助力制动系统(Electro-booster,EBooster)的液压力控制中液压负载的非线性和不一致性问题,提出一种基于径向基函数(Radial based function,RBF)神经网络的滑模变结构控制方法。设计EBooster系统压力控制架构,建立液压制动系统等效结构简化模型,据此设计基于RBF网络滑模变结构的液压力控制方法,通过设计RBF网络的自适应律来实现系统滑模控制参数的自适应调整,并利用李雅普诺夫函数方法分析算法的稳定性。最后搭建电动助力制动系统的快速原型试验平台来验证算法的有效性。试验结果表明,采用RBF神经网络滑模变结构的控制策略对电动助力制动系统液压力的控制误差在2%以内,具有良好的控制效果。研究成果为EBooster系统的压力控制提出一种具有良好自适应性的算法设计思路。

Sliding Mode Control Based on RBF Network for Hydraulic Pressure in Electric Power-assisted Brake System

To solve the problem of nonlinearity and inconsistency of hydraulic brake system in the pressure control of automobile electric booster(EBooster) braking system, a sliding mode variable structure control method based on radial based function(RBF) neural network is proposed. The pressure control architecture of EBooster is designed and a simplified equivalent structure model of hydraulic braking system is established. Then the hydraulic pressure sliding mode control method based on RBF network is designed. The sliding mode control parameters of the system are adaptively adjusted by designing the adaptive law of RBF network. The stability of the algorithm is analyzed by Lyapunov function. Finally, the rapid control prototyping(RCP) experiment platform of electric power-assisted brake system is built and the algorithm is verified by RCP test. The experimental results show that the control strategy of the sliding mode control method based on RBF network has a good performance with a control error no larger than 2%. A design method of adaptive pressure control algorithm is provided for the EBooster system.