关键词: FOS-ELM神经网络, 输入扰动, 自适应控制, PID增量, 系统误差

Abstract: To deal with the difficulty of input disturbance system in achieving adaptive control, we analyzed the defects of FOS-ELM controller and proposed a method of the PID compensation FOS-ELM controller design. The method increases the PID increments of system output error in FOS-ELM controller, and obtains the parameters of controller online by using FOS-ELM algorithm, then completes the controller design. The disturbance input is considered in the controller decision, so the controller can reduce the effect of disturbance on controller output effectively. Finally, the control effect of the controller was simulated and analyzed, demonstrating the feasibility and validity of the PID compensation FOS-ELM controller.

Key words: Full Online Sequential limit Learning Machine (FOS-ELM)neural networks, input disturbance, adaptive control, Proportion Integration Differential (PID)incremental, system error