CIESC Journal

• RESEARCH PAPERS • 上一篇    下一篇

多目标进化算法求解过程综合中的混合整数线性规划(MILP)与混合整数非线性规划问题(MINLP)

石磊; 姚平经   

  1. Laboratory of Proceas System Engineering, Dalian University of Technology, Dalian 116012,
    China
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2001-06-28 发布日期:2001-06-28
  • 通讯作者: 石磊

Multi-objective Evolutionary Algorithms for MILP and MINLP in Process Synthesis

SHI Lei; YAO Pingjing   

  1. Laboratory of Proceas System Engineering, Dalian University of Technology, Dalian 116012,
    China
  • Received:1900-01-01 Revised:1900-01-01 Online:2001-06-28 Published:2001-06-28
  • Contact: SHI Lei

摘要: Steady-state non-dominated sorting genetic algorithm (SNSGA), a new form of multi-objective
genetic algorithm, is implemented by combining the steady-state idea in steady-state
genetic algorithms (SSGA) and the fitness assignment strategy of non-dominated sorting
genetic algorithm (NSGA). The fitness assignment strategy is improved and a new self-
adjustment scheme of σshare is proposed. This algorithm is proved to be very efficient
both computationally and in terms of the quality of the Pareto fronts produced with five
test problems including GA difficult problem and GA deceptive one. Finally, SNSGA is
introduced to solve multi-objective mixed integer linear programming (MILP) and mixed
integer non-linear programming (MINLP) problems in process synthesis.

关键词: multi-objective programming;multi-objective evolutionary algorithm;steady-state non- dominated sorting genetic algorithm;process synthesis

Abstract: Steady-state non-dominated sorting genetic algorithm (SNSGA), a new form of multi-objective
genetic algorithm, is implemented by combining the steady-state idea in steady-state
genetic algorithms (SSGA) and the fitness assignment strategy of non-dominated sorting
genetic algorithm (NSGA). The fitness assignment strategy is improved and a new self-
adjustment scheme of σshare is proposed. This algorithm is proved to be very efficient
both computationally and in terms of the quality of the Pareto fronts produced with five
test problems including GA difficult problem and GA deceptive one. Finally, SNSGA is
introduced to solve multi-objective mixed integer linear programming (MILP) and mixed
integer non-linear programming (MINLP) problems in process synthesis.

Key words: multi-objective programming, multi-objective evolutionary algorithm, steady-state non- dominated sorting genetic algorithm, process synthesis