基于GGAP-RBF神经网络逆的复杂多电机系统同步控制

GGAP-RBF Neural Network Inverse Synchronous Control of Complex Multi-motor System

针对以矢量控制工作方式的复杂多电机同步系统,以3台电机同步系统为研究对象,证明了该系统可逆,提出了基于增长和修剪的RBF(GGAP-RBF)神经网络逆的多电机同步控制方法。将RBF神经网络逆串接在三电机系统之前,组成由速度和张力子系统组成的伪线性复合系统,分别对速度和张力子系统设计闭环控制器,实现了对速度和张力的解耦控制。根据给定的性能指标,对RBF神经元数目进行适当增删和修剪,解决了BP神经网络逆控制系统难以进行优化的问题,为复杂多电机同步系统的控制提供了简化依据。仿真结果表明,该方法具有较好的动静态性能。

Three-motor synchronous system is proved to be invertible.A method of generalized growing and pruning RBF（GGAP-RBF）neural network inverse for synchronous control of multi-motor system is proposed.The inverse can be constructed by combining the RBF neural network with an integrator.And the speed and tension control of three-motor system is decoupled by combing the inverse RBF neural network with the three-motor system.Through growing and pruning RBF neural network,according to the given performance index,the three-motor variable frequency speed-regulating control system is optimized,solving the problem which can＇t be solved by BP neural network.The method provides a theoretical basis for the research of complex multi-motor synchronous systems.The simulation results show the better dynamic and static operation performance.