Application of NSGA-II in Parameter Optimization of Extended State Observer
Active Disturbance Rejection Control (ADRC) is an effective solver for object with large inertia, strong coupling or hardly modeling. However, how to tune and optimize the parameters of this control system, especially of ESO, is the key problem which block its aboard appliance. Now the tuning methods mainly includes manual tuning with experience, simple model and genetic algorithm. The performance of the parameters of ESO shows in the errors of some state variables. And so it is a multi-objective optimal problems. In these scheme, the aim function is constructed from these errors with some weight values through linear or no-linear function. Then the multi-objective problem becomes one objective problem. In such scheme, how to get weight values is very important. So it is hard to get the global optimal solutions. In fact, we can optimize these parameters with multi-objective solutions. NSGA-II is a very good choice in these multi-objective solutions. This article optimized the parameters of ESO with NSGA-II. There are some faults in NSGA-II, such as more calculation and more calculation time. So this article developed this algorithm. These showed good performance in emulation.
ESO NSGAII Multi-objective optimization ADRC
Qiang Ma Daping Xu Peng Iv Yuntao Shi
North China Electricity Power University, Beijing, 102206, China
国际会议
2007杭州国际动力工程会议(The International Conference on Power Engineering 2007)
杭州
英文
2007-10-23（万方平台首次上网日期，不代表论文的发表时间）