An Improved Comentropy-based Multi-objective Optimization Algorithm

摘要：

To solve the increasing run-time complexity with the growth of population solutions in PESA, an evolutionary algorithm of multi-objective optimization, we present a comentropy- based PESA algorithm (C-PESA). With this algorithm, the gradual development and maturity of the solution sets can be observed with the continuous calculation of entropy values to determine whether to stop the optimization process or not, which reduces the run-time complexity to some extent. Simulation results show that the computational effort of C-PESA increases at a linear order with the increasing number of solutions, and the efficiency of evolutionary algorithm shows improvement.

关键词： PESA algorithm comentropy multi-objective optimization evolutionary algorithm Pareto-optimal run-time complexity

作者: Kun Wang Linlin Wang Yan Liu Yuhua Zhang Huang Guo Yue Yu

作者单位: Key Lab of Broadband Wireless Communication and Sensor Network Technology (Nanjing University of Pos College of automation and electrical engineering, Nanjing University of Technology, Nanjing 210009, Key Lab of Broadband Wireless Communication and Sensor Network Technology (Nanjing University of Pos

会议类型: 国内会议

会议名称: 第23届过程控制会议

会议地点: 厦门

会议语种:英文

页码: 1-9

在线出版日期: 2012-08-01（万方平台首次上网日期，不代表论文的发表时间）