轴承动态测量仪神经网络控制系统设计

Design of Neural Network Control System for Dynamic Measuring Instrument of Bearing

轴承摩擦转矩的测量是一种动态测量,需要轴承在转动条件下获得数据.轴承型号及承载条件的改变,影响电机转速的控制规律.针对经典PID控制器的结构参数不能在线修改的问题,提出利用BP神经网络的自学习能力,对PID控制器进行结构参数调整,实现对电机速度规律的优化控制方法.仿真结果表明,经神经网络修正过的PID控制器,动态性能优于未修正的PID控制器.

Measurement of bearing friction torque is a kind of dynamic measurement whose data obtained in the rotation. Once the bearing type and load conditions change, will affect the control of motor speed. According to the structure parameters of the classical PID controller can not be modified,it proposed to use the self-learning ability of BP neural network to adjust the PID controller＇s structure parameters, to realize the optimization of the motor velocity＇s control method. The simulation results show that the dynamic performance of PID controller which modified by neural network is better than unmodified.