摘要
针对多目标优化过程中如何根据个人偏好确定各目标权重的问题,提出一种约束优化方法以获得各目标的最佳权重.首先,将目标权重计算问题转化为综合适应度最大方差计算问题;然后,将个人偏好转化为最大方差问题不等式约束条件;最后,利用遗传算法和梯度投影法求解约束优化问题以获得最佳的目标权重.在电力机车故障维修策略决策过程中应用该算法计算各部件经济性、安全性等目标权重,实验结果验证了所提出方法能够获得满足个人偏好的最佳目标权重.
For the problem that how to compute the objective weight based on personal preference during multi-objective optimization process, a method which decides the objective weights by solving a constrained optimization problem is proposed. Firstly, by using this method, the objective weight computation problem is transformed into a synthetical fitness optimization problem according to the statistics theory. Then the personal preference is transformed into the constrain condition of the synthetical fitness optimization problem. Finally, the gradient projection method is used to solve the constrained synthetical fitness optimization problem to get optimum objective weights. The proposed method is used to compute the objective weight of economical efficiency and safety during the electric locomotive repair strategy decision process, and the test result shows that the proposed method can get the optimum objective weight under the constraint of personal preference.
出处
《控制与决策》
EI
CSCD
北大核心
2014年第8期1471-1476,共6页
Control and Decision
基金
国家自然科学基金项目(61273174
61034006
60874047)
关键词
个人偏好
多目标优化
约束优化
方差
personal preference
multi-objective optimization
constrained optimization
variance
