Research On Multi-Objective Cuckoo Search Algorithm And Its Application

In reality,the majority of optimization problems involve multiple conflicting objectives.For such multi-objective optimization problems,population-based evolutionary algorithms is an efficient approach to obtaining multiple Pareto optimal solutions in a single run,as opposed to the classic methods of repeatedly applying an algorithm.Among these algorithms,the cuckoo search algorithm stands out as a novel bio-inspired algorithm within the swarm intelligence category of evolutionary algorithms.Since its proposal,the multi-objective cuckoo search algorithm has garnered significant attention from the academic community.This paper aims to enhance the performance of the multi-objective cuckoo search algorithm and apply it to practical problems.The work presented in this paper focuses on three main aspects:1.A dynamic scaling factor suitable for the multi-objective cuckoo algorithm is proposed.This factor improves the convergence speed of the algorithm and enhances solution accuracy.To verify the effectiveness of DSF,it is integrated with three classic multi-objective algorithm frameworks and tested on CEC2019 benchmark functions.The simulation test data is analyzed and compared.The experimental results demonstrate the excellent performance of the three types of multi-objective cuckoo algorithms combined with DSF.The proposed DSF exhibits better versatility,particularly when combined with an indicator-based algorithmic framework.2.A hypervolume-based cuckoo search algorithm with enhanced diversity and dynamic scaling factor(Hyp ECS)is proposed.To fully leverage the potential of the best individual in guiding the population and get a swarm intelligence algorithm with superior performance,Hyp ECS is designed.Building upon the Hyp E framework and incorporating the DSF method,Hyp ECS introduces the use of hypervolume as an intuitive indicator to measure the performance of population.Additionally,a diversity enhancement module is added to improve population diversity when an uneven distribution is observed,preventing the algorithm from converging to local optima.To validate the algorithm’s effectiveness and versatility,extensive tests are conducted on 31 functions,including ZDT,WFG,and the CEC2019 test function group.Simulation results are compared with those obtained from six classic multi-objective evolutionary algorithms and five state-of-the-art multi-objective evolutionary algorithms.The experimental results demonstrate the superior diversity and high efficiency of Hyp ECS.3.Application of Hyp ECS to practical problems.In order to demonstrate the practical applicability of Hyp ECS,eight classic multiobjective practical problems with constraints are selected,along with an urban tourism carrying capacity model.Hyp ECS is combined with the prevailing and effective constraint processing technology to solve these problems and the urban tourism carrying capacity model.The obtained results are compared with those generated by classic multi-objective evolutionary algorithms.The experimental results confirm the practicality and effectiveness of Hyp ECS,providing valuable references and examples for future real-world applications.