Application of Single Neuron in Cascade Control System
Because of some complicated factors in a cascade control system, such as that the state of the control object is nonlinear, of uncertainty, and of time-varying parameters, it is quite difficult to establish a mathematical model, so that the control effect is not always good enough. When a common PID controller is used, there exist some drawbacks, such as the poor performance to rejection of disturbance and the bigger static error in the system caused by the scale coefficient KP. Although the static error can be overcome by increasing KP or by increasing extent of the I adjuster, it will take a longer time. In this paper, an on-line method of tuning of parameters-self-adaptive PID control algorithm with a single neuron-is presented, in which the tuning of an on-line adaptive and time-varying parameters is used. The single neuron PID controller is used to the cascade control system. This algorithm is not only simply in mathematics, but also can overcome the nonlinearity and enhance the ability to rejection of disturbances. The experiment shows that the algorithm is feasible and effective.
cascade control system Integral separation PID Single neuron networks
Ding Fang Fang Xudong
Department of Automation, Civil Aviation University of China, Tianjin 300300, China China Eastern, Nanjing 210000, China
国际会议
北京
英文
2007-08-05（万方平台首次上网日期，不代表论文的发表时间）