同步电动机励磁系统的模糊神经网络控制

由于在拖动周期性波动负载的工况下，电励磁同步电动机可以通过改变励磁电流来提高运行效率，提出了采用模糊神经网络控制方法对励磁系统加以调节，以使同步电动机的功率因数维持在近似为1的过励磁情况。模糊神经网络控制通过训练神经网络来记忆模糊控制规则，它不需要存储模糊控制表，节省内存空间，且具有较强的自学习能力与联想能力。采用Simulink子模块构建了整个系统。仿真结果表明：与PID控制方法相比，模糊神经网络控制有良好的快速跟踪性能和抗干扰性能。

Fuzzy Neural-network Control of Synchronous Motor Excitation System

Considering that the efficiency of synchronous motor may be improved by means of adjusting exciting current under driving variability load conditions,an approach to adjust exciting current is presented based on fuzzy neural network.The power factor of synchronous motor is approximately kept at 1.Fuzzy neural network stores fuzzy control rules by training neural network instead of storing fuzzy control table so as to economize memory cells,and is provided with good learning ability and association ability.The whole system is structured by Simulink,and the simulation results show that fuzzy neural network control is preferable to PID control in respect of fast tracking performance and anti-jamming performance.