基于神经网络的汽车离合器综合性能试验台控制系统设计

Designed of Control System of Automobile Clutch Test Platform Based on Neural Network

基于神经网络的PID控制算法设计了汽车离合器综合性能试验台的转速控制系统。设计了试验台的总体结构并建立了转速控制系统的数学模型。转速控制系统是试验台控制的关键,针对转速控制系统干扰力矩较大和控制精度要求高的情况,提出了一种基于神经网络的自适应PID控制算法对转速进行控制,将神经网络鲁棒性强与PID控制稳态误差小的优点结合起来,降低了干扰和参数变化对系统的影响。仿真及试验结果表明,其比经典PID控制更好,降低了超调量,有较强的鲁棒性和自适应性,提高了检测精度与检测效率。

The rotation speed control system of the comprehensive performance test platform of the automobile clutch is designed based on neural network with self-adaptation.The structure of test platform is designed,and the mathematical model of rotation speed control system is established.The speed control system is the key to the Test Platform,and the neural network control algorithm is proposed for the rotation speed control against the large interference torque and the high control accuracy requirement.The algorithm takes advantage of the strong robust of neural network and the small steady state error of PID control,decreased the bad influence of interference and varied parameter.The results of simulation and experiment show that it is better than the traditional PID controller,with strong robustness and excellent flexibility and adaptability as well as the test accuracy and efficiency have been greatly improved.