点击次数：

论文类型：会议论文

发表时间：2018-01-01

收录刊物：CPCI-SSH

页面范围：75-79

关键字：wolf pack algorithm; optimization; function optimization; One Belt One Road

摘要：There are many optimization problems in port logistics automation in promoting One Belt One Road policy. Swarm intelligent algorithms are regarded as promising approach to complex port logistics optimization problems. Wolf pack algorithm (WPA) is a newly proposed algorithm, which has been widely applied in many fields such as automated container terminals, port logistics optimization, and regional economic planning. However, defects still exist in WPA. Aiming at the defects of WPA, such as fixed scouting direction and R artificial wolves with worst function values will directly be removed, an improved wolf pack algorithm named RHWPA is proposed based on random scouting direction and hunger value strategy. During each scouting, fixed directions h is selected randomly within a certain range. It can avoid the scout wolves falling into local optima, and improve the convergence speed and optimization precision. To further improve the convergence speed, the hunger value strategy is applied instead of the stronger-survive renewing rule for the wolf pack. Simulation results on the benchmark functions show that RHWPA is feasible and effective. Our work is expected to provide inspiration and help to solving corresponding port logistics optimization problems perfectly on promotion One Belt One Road policy.