基于神经网络的模糊控制在装甲车电传动控制系统中的应用

Application of Fuzzy Control Based on Neural Networks in the Electrical Drive Control System of Armored Vehicle

针对装甲车辆电传动系统是一个复杂的多变量、非线性系统 ,其控制系统的扰动变化大 ,对驱动电机的控制要求苛刻的特点 ,根据模糊控制器对参数的变化不敏感 ,对非线性系统进行控制时往往能取得较好的控制效果 ,以及神经网络的学习功能、联想记忆功能、分布并行式处理功能可以更好地实现模糊逻辑控制中的规则表示、知识获取和并行推理 ,因此利用神经网络来实现无刷直流电机调速系统的模糊控制。在MATLAB中进行计算机仿真 ,通过仿真分析表明 ,当突然加、减负载时 ,神经网络模糊控制与PID控制相比 ,具有对参数变化不敏感以及超调和振荡小等特点。

Aimed at characteristics that the armored vehicle electrical drive system is a multivariable and nonlinear system with the control system of diverse disturbance and the motor of rigorous control demand, and based on principles that fuzzy controller is not sensitive to the change of parameters and can get better control effects for nonlinear system and neural networks of learning, associative memory and distributing parallel processing function can better realize rule representation, knowledge acquisition, and parallel inference, so a neural network is used to realize the fuzzy control in a BLDCM (brushless direct current motor) speed regulator. Simulation analyses using MATLAB show that the neural network fuzzy control system is less sensitive to the change of parameters than that in PID control system and the overshooting and oscillation are also weakened during the resuming period when the loads are suddenly increased or decreased. This method is feasible for the electrical drive control system of armored vehicles.