摘要
混合励磁无刷爪极发电机具有非线性、多耦合的特征,基于单神经元自适应PID控制方法,在MATLAB/Simulink中建立了它的动态响应仿真模型。该模型基于有监督的Hebb学习规则,通过在线动态调整加权系数,达到了在线控制的目的,不仅结构相对简单,而且具有较高的控制精度和鲁棒性,能应对复杂环境参数的变化。动态响应仿真结果表明,随着转速的变化,系统能自动调整励磁电流,使得发电机输出电压保持在稳定的水平,动态特性好。
The hybrid excitation brushless claw-pole alternator is highly nonlinear and variable, thus the dynamic performance simulation model was built in the MATLAB/Simulink based on the single neuron PID control method. By using supervised Hebb learning algorithm, the model can adjust the weighted factor to achieve online control. The structure of the excitation controller was simple and it had great ability in adaptability, robustness, high control accuracy. According to the simulation results, with the changes of the rotational speed, the system enables the output voltage to keep steady by adjusting the excitation current and it has good dynamic performance.
出处
《微特电机》
北大核心
2017年第2期9-12,共4页
Small & Special Electrical Machines
基金
国家自然科学基金项目(51177090)
关键词
混合励磁
爪极发电机
单神经元
PID
动态响应
hybrid excitation
claw-pole alternator
single neuron
PID
dynamic performance
