The aim of software module clustering problems (SMCPs) is to automatically find a good quality clustering of software modules based on relationships among modules. In this paper, we propose a multi-agentevolutionary ...
详细信息
The aim of software module clustering problems (SMCPs) is to automatically find a good quality clustering of software modules based on relationships among modules. In this paper, we propose a multi-agent evolutionary algorithm to solve this problem, labeled as MAEA-SMCPs. With the intrinsic properties of SMCPs in mind, three evolutionary operators are designed for agents to realize the purpose of competition, cooperation, and self-learning. In the experiments, practical problems are used to validate the performance of MAEA-SMCPs. The results show that MAEA-SMCPs can find clusters with high quality and small deviations. The comparison results also show that MAEA-SMCPs outperforms two existing multi-objective algorithms, namely MCA and ECA, and two existing single-objective algorithms, namely GGA and GNE, in terms of MQ.
In view of a large number of distributed generators penetrated into distribution network, an improved hierarchical island operation was proposed to make the DG play the role as backup power supply in the service resto...
详细信息
ISBN:
(纸本)9783037852668
In view of a large number of distributed generators penetrated into distribution network, an improved hierarchical island operation was proposed to make the DG play the role as backup power supply in the service restoration of the distribution network;As the renewable distributed generators was strongly advocated, A new mathematical model was proposed to make full use of them. The service restoration problem is a constrained multi-objective optimization problem, a new algorithm which was the combination the heuristic algorithm and multi-agent evolutionary algorithm was proposed, it improved the multi-agent evolutionary algorithm with the Niche technologies to ensure the variety of the populations in the post-optimization and used the adaptive updating strategy to improve the convergence speed. The results of a case show feasibility and correctness of the model and the algorithm.
暂无评论