论文部分内容阅读
如何在众多非劣解中为决策者推荐一个合理的方案是使用多目标粒子群算法(MOPSO)所面临的问题.为此,将逼近理想解的排序方法(TOPSIS策略)引入到算法中.为了提高求解精度和均匀性,还提出了基于Pbest的变异策略和改进的k邻近距离策略.测试结论显示,仅使用TOPSIS策略确定Gbest的算法,求解精度虽好,但均匀性较差,而包含所有改进策略的算法在精度和均匀性方面都更优,并且能够按照TOPSIS方法在非劣解集中找到一个适合向决策者推荐的“理想”方案.
How to recommend a reasonable scheme for decision-makers in many non-inferior solutions is the problem of using the multi-objective Particle Swarm Optimization (MOPSO) .To this end, the method of ranking approaches to TOPSIS (TOPSIS strategy) is introduced into the algorithm Improve the accuracy and uniformity of the solution, and also put forward Pbest-based mutation strategy and improved k-distance strategy.The test results show that using the TOPSIS strategy to determine Gbest algorithm, the solution accuracy is good, but the uniformity is poor, Algorithms to improve the strategy are better in terms of accuracy and uniformity, and can find a “ideal” solution suitable for decision-makers in non-degradable solutions according to the TOPSIS method.