Multi-objective Optimization by non-dominationsearching based NSGA-II Algorithm
Chen Zhen Huan, Zhang Dong Liang, Zhao Wei, Hu Chun Guang, Wang Pan Pan
Abstract
Chen Zhen Huan, Zhang Dong Liang, Zhao Wei, Hu Chun Guang, Wang Pan Pan
Abstract
The core elements of THE NSGA-II algorithm are genetic search and non-dominant sorting based on elite retention strategy It belongs to passive search and lacks an active search mechanism. In some cases, it will affect the efficiency of the algorithm in searching for the optimal Pareto front. According to the distribution characteristics of each generation of population samples, this paper calculates the non-dominated direction of the Pareto front from the solution set with Pareto ranks 1 and 2 in the population of that generation, and actively searches for a step toward the non-dominated direction from each sample of the Pareto front to find the better Non-dominant may solve and participate in the next generation of subgroup reconstruction. In the experimental verification, the effectiveness of the proposed method is verified through classic calculation examples and multi-objective optimization problems after power grid black start.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The core elements of THE NSGA-II algorithm are genetic search and non-dominant sorting based on elite retention strategy It belongs to passive search and lacks an active search mechanism. In some cases, it will affect the efficiency of the algorithm in searching for the optimal Pareto front. According to the distribution characteristics of each generation of population samples, this paper calculates the non-dominated direction of the Pareto front from the solution set with Pareto ranks 1 and 2 in the population of that generation, and actively searches for a step toward the non-dominated direction from each sample of the Pareto front to find the better Non-dominant may solve and participate in the next generation of subgroup reconstruction. In the experimental verification, the effectiveness of the proposed method is verified through classic calculation examples and multi-objective optimization problems after power grid black start.
Key concepts: Sorting, Multi-objective optimization, Mathematical optimization, Pareto principle, Population, Computer science, Set (abstract data type), Algorithm