Tuning of Auto Disturbance Rejection Controller based on multi-objective Genetic Algorithm

Abstract

Conventional tuning approaches for Auto Disturbance Rejection Controller (ADRC) are basically trial ones. The procedure to adjust parameters is complicated and tedious and therefore is hard to achieve the expected goal. This paper proposed a three-step method based on the separation principle. First, the proper TD settings are verified through the step response. Then, the parameters of the ESO are designed. A NSGA-II was used to obtain the approximate parameters in line with observer to achieve a satisfied solution. Finally, the ADRC three modules were combined to decide NLSEF settings. This optimization problem with overshoot, settling time and control energy requirements used NSGA-II to obtain the approximate Pareto set of multi-objective optimization. The 2nd-order heading ADRC for a 150,000-ton tanker with steering gear was studied to verify the proposed method. The results show that it is simple and efficient, and therefore it should be very useful in engineering applications.