Parameter Optimization And Applycation Research Of Active Desturbance Rejection Controller

With the development of modern industry, people has higher requirement for the control quality, and the traditional controller is difficult to obtain satisfactory control effect. ADRC as a new control strategy has better adaptability and control effect than conventional controllers. However, when using ADRC to control the system, the controller parameters is determined to a large extent depend on the experience, and it is difficult to regulate.In order to more accurately determine the parameters of the turbine governor systems linear ADRC, this article combines Nelder-Mead simplex method and gravitational search algorithm together, proposed the Nelder-Mead simplex search Gravitational Search Algorithm. The search and replace mechanism in NM simplex search and the particles update method in gravitational search algorithm are combined together in the algorithm. Simplex method has strong local search ability, and it can make the particles falling into local oscillation out in time, thus avoiding the gravitational search algorithm plunging local optimum.The controlled object in this Paper have different characteristics, Including superheated steam temperature cascade control system and hydraulic turbine system.When control superheated steam temperature cascade control system, compared with PID-P controller based on internal model method, indicating the effectiveness of the proposed method. Aim at turbine governor systems, the control method in the paper can well suppress overshoot, reduce the negative tone and reduce adjustment time. When system conditions change, there are still good control results. It has strong robustness,and can achieve better performance than the PID control in anti interference.