摘要
将具有自学习功能的单神经元模型与常规的PID控制算法相结合 ,设计了单神经元PID控制器 ,并应用于超临界机组过热汽温控制系统。仿真结果表明 ,采用单神经元PID控制器的汽温控制系统 ,能够适应被控对象在较大范围内的变化 ,具有较强的自适应能力和鲁棒性 ,其控制品质优于常规的PID控制器。此外根据仿真结果 ,对单神经元的学习算法进行了改进 。
Combining selflearning single neuron model with conventional PID control algorithm, a single neuron PID controller has been designed and used for the superheated steam temperature control system. The simulation results show that the said control system is adaptable for a large range of controlled object variation, having high robustness and adaptability, the control quality of single neuron PID controller being better than that of conventional PID controller. In addition, on the basis of said simulation results, the learning algorithm of single neuron has been further improved, enhancing the better control quality of single neuron PID controller.
出处
《热力发电》
CAS
北大核心
2004年第2期59-62,共4页
Thermal Power Generation
基金
国家 973项目 (G1 9990 2 2 30 4 )资助