Fast Non-dominated Sorting Genetic Algorithm with Three Crossover Individuals for Network Topology Optimization in Industrial Internet of Things
Fanrong Kong, Yan Huang, Maoqing Zhang · 2019
With the deep integration of information and communication, people's demand for more reliable network topology structure is becoming higher and higher. Multi-objective model of switched industrial Ethernet topology structure is a recently proposed mode, which has been proved to have a flexible network structure and high communication efficiency. However, the fast non-dominated sorting algorithm II (NSGA-II), which is used to tackle the mode above, has the drawback that multiple repeatedly selected parent individuals may result in low diversity of the offspring population. To tackle this issue, this paper proposes to employ three parent individuals to generate offspring individuals. Experimental results on ZDT test suit demonstrate that the proposed method can evidently improve NSGA-II. Based on the further experiment results on multi-objective model of switched industrial Ethernet topology structure, it can be concluded that the proposed method also has better performance in tackling practical problem.