摘要
针对常规PID控制应用于双电机变频调速系统时参数难以整定、自适应性差且无法实现系统解耦等问题,在系统物理模型建立的基础上,确定了系统的控制方案,构建一种神经网络智能解耦控制器。首先设计2个BP神经网络参数自适应PID控制器,对系统转速和张力分别进行单回路自适应控制;然后设计1个自适应神经元解耦补偿器,串联在自适应PID控制器之后,通过训练神经网络,补偿回路之间的耦合影响。仿真结果表明,系统具有良好的解耦控制性能和跟随性能,证明了该控制策略的可行性和先进性,为双电机控制优化设计提供了依据。
When common PID controller is applied into dual-motor variable frequency speed-regulating system, its parameters are difficult to set, with poor self-adaptability and being unable to realize the decoupling of the system. On the basis of system modeling, the control strategy of the system was determined, and a neural network intelligent decoupling controller was designed. Firstly, two BP neural network PID controllers were designed to perform the a- daptive control of speed and tension. Then, a neuron decoupling compensator was designed. Finally, the neuron de- coupling compensator was set to designed series connected after adaptive PID controller, through training the weighs to compensate the compensate the system coupling effect. The simulation results show that the system has good decou- pling control and tracking performances, which shows that the method is efficient and advanced.
出处
《计算机仿真》
CSCD
北大核心
2013年第11期374-378,共5页
Computer Simulation
基金
盐城工学院校级科研项目(XKR2010071)
盐城市科技计划项目(YKN2012031)
关键词
双电机
变频调速
神经网络
参数自整定
解耦补偿控制
Dual-motor
Variable frequency speed-regulating
Neural network
Parameter self- regulation
Decou-piing compensation control
