Multi-objective grasshopper optimization algorithm based on multi-group and co-evolution

Abstract

<abstract> <p>The balance between exploration and exploitation is critical to the performance of a Meta-heuristic optimization method. At different stages, a proper tradeoff between exploration and exploitation can drive the search process towards better performance. This paper develops a multi-objective grasshopper optimization algorithm (MOGOA) with a new proposed framework called the Multi-group and Co-evolution Framework which can archive a fine balance between exploration and exploitation. For the purpose, a grouping mechanism and a co-evolution mechanism are designed and integrated into the framework for ameliorating the convergence and the diversity of multi-objective optimization solutions and keeping the exploration and exploitation of swarm intelligence algorithm in balance. The grouping mechanism is employed to improve the diversity of search agents for increasing coverage of search space. The co-evolution mechanism is used to improve the convergence to the true Pareto optimal front by the interaction of search agents. Quantitative and qualitative outcomes prove that the framework prominently ameliorate the convergence accuracy and convergence speed of MOGOA. The performance of the presented algorithm has been benchmarked by several standard test functions, such as CEC2009, ZDT and DTLZ. The diversity and convergence of the obtained multi-objective optimization solutions are quantitatively and qualitatively compared with the original MOGOA by using two performance indicators (GD and IGD). The results on test suits show that the diversity and convergence of the obtained solutions are significantly improved. On several test functions, some statistical indicators are more than doubled. The validity of the results has been verified by the Wilcoxon rank-sum test.</p> </abstract>

Enterprise financial early warning based on improved particle swarm optimization algorithm and data mining

“…Then, in order to solve the problem that the diversity measurement method selected in this paper cannot correspond to different dimensions of each dimension, a standard diversity measurement method is proposed in this paper. In addition, this paper also briefly introduces the commonly used diversity guidance strategy and multi-group strategy, and puts forward the diversity guidance strategy and multi-group parallel strategy for the following content [8]. The literature shows that the definition of data mining standards exists in incomplete and noisy data sets, and the model is expressed in language when extracting patterns that are effective, innovative, useful, and ultimately essentially understandable, and therefore can be used to describe data set subsets that require the model to be simpler than computing data subsets.…”

A multi-fault recovery oriented deployment strategy of backup controllers in SDN

“…In view of these shortcomings, the mechanism of Opposite Learning in SSA(OLSSA) was introduced 17 to extend the optimization range and prevent the method from falling into the trap of local optimum. In this work, the optimization capability of OLSSA is further improved by using the idea of population guidance 18 after considering the idea of opposite learning in the initialization of the algorithm.…”

A Competitive Particle Swarm Algorithm Based on Vector Angles for Multi-Objective Optimization

“…Grouping the population is an effective solution, and its intention is to increase the scope of the search space through grouping. Similarly, it is also mentioned in [59] to improve the performance of the algorithm by setting different parameters and search strategies for each population. It is worth noting that our grouping here is to make population 1 pay more attention to convergence and ensure the convergence of the algorithm.…”