The differential evolution(DE)algorithm relies mainly on mutation strategy and control parameters'selection.To take full advantage of top elite individuals in terms of fitness and success rates,a new mutation oper...The differential evolution(DE)algorithm relies mainly on mutation strategy and control parameters'selection.To take full advantage of top elite individuals in terms of fitness and success rates,a new mutation operator is proposed.The control parameters such as scale factor and crossover rate are tuned based on their success rates recorded over past evolutionary stages.The proposed DE variant,MIDE,performs the evolution in a piecewise manner,i.e.,after every predefined evolutionary stages,MIDE adjusts its settings to enrich its diversity skills.The performance of the MIDE is validated on two different sets of benchmarks:CEC 2014 and CEC 2017(special sessions&competitions on real-parameter single objective optimization)using different performance measures.In the end,MIDE is also applied to solve constrained engineering problems.The efficiency and effectiveness of the MIDE are further confirmed by a set of experiments.展开更多
Differential evolution (DE) is a kind of evolutionary algorithms, which is suitable for solving complex optimization problems. Mutation is a crucial step in DE that generates new solutions from old ones. It was argu...Differential evolution (DE) is a kind of evolutionary algorithms, which is suitable for solving complex optimization problems. Mutation is a crucial step in DE that generates new solutions from old ones. It was argued and has been commonly adopted in DE that the solutions selected for mutation should have mutually different indices. This restrained condition, however, has not been verified either theoretically or empirically yet. In this paper, we empirically investigate the selection of solutions for mutation in DE. From the observation of the extensive experiments, we suggest that the restrained condition could be relaxed for some classical DE versions as well as some advanced DE variants. Moreover, relaxing the restrained condition may also be useful in designing better future DE algorithms.展开更多
为平衡算法收敛速度和全局搜索能力,克服差分进化算法易"早熟"的缺陷,在分析引起种群多样性下降及个体进化停滞原因的基础上,通过引入高斯变异操作,提出了基于高斯变异改进的差分进化算法(modified differ-ential evolution b...为平衡算法收敛速度和全局搜索能力,克服差分进化算法易"早熟"的缺陷,在分析引起种群多样性下降及个体进化停滞原因的基础上,通过引入高斯变异操作,提出了基于高斯变异改进的差分进化算法(modified differ-ential evolution base on Gauss mutation,GMDE).数值仿真及2个工程优化问题的求解结果表明本文算法能有效避免"早熟"收敛,且在算法收敛速度和全局搜索能力上取得了较好的平衡.展开更多
基金supported by the A*STAR under its RIE2020 Advanced Manufacturing and Engineering(AME)Industry Alignment Fund-Pre-Positioning(IAF-PP)(Award A19D6a0053)the Japan Society for the Promotion of Science(JSPS)KAKENHI(JP22H03643)。
摘要The differential evolution(DE)algorithm relies mainly on mutation strategy and control parameters'selection.To take full advantage of top elite individuals in terms of fitness and success rates,a new mutation operator is proposed.The control parameters such as scale factor and crossover rate are tuned based on their success rates recorded over past evolutionary stages.The proposed DE variant,MIDE,performs the evolution in a piecewise manner,i.e.,after every predefined evolutionary stages,MIDE adjusts its settings to enrich its diversity skills.The performance of the MIDE is validated on two different sets of benchmarks:CEC 2014 and CEC 2017(special sessions&competitions on real-parameter single objective optimization)using different performance measures.In the end,MIDE is also applied to solve constrained engineering problems.The efficiency and effectiveness of the MIDE are further confirmed by a set of experiments.
基金Acknowledgements The authors would like to thank the anonymous reviewers for their very constructive and helpful suggestions. This work was supported in part by the National Basic Research Program (973 Program) of China (2011CB013104), in part by the Innovation-driven Plan in Central South University (2015CXS012 and 2015CX007), in part by the National Natural Science Foundation of China (Grant Nos. 61273314 and 61673397), in part by the EU Horizon 2020 Marie Sldodowska-Curie Individual Fellowships (Project ID: 661327), in part by the Hunan Provincial Natural Science Fund for Distinguished Young Scholars (2016JJ1018), in part by the Program for New Century Excellent Talents in University (NCET-13-0596), and in part by State Key Laboratory of Intelligent Control and Decision of Complex Systems, Beijing Institute of Technology.
摘要Differential evolution (DE) is a kind of evolutionary algorithms, which is suitable for solving complex optimization problems. Mutation is a crucial step in DE that generates new solutions from old ones. It was argued and has been commonly adopted in DE that the solutions selected for mutation should have mutually different indices. This restrained condition, however, has not been verified either theoretically or empirically yet. In this paper, we empirically investigate the selection of solutions for mutation in DE. From the observation of the extensive experiments, we suggest that the restrained condition could be relaxed for some classical DE versions as well as some advanced DE variants. Moreover, relaxing the restrained condition may also be useful in designing better future DE algorithms.
摘要为平衡算法收敛速度和全局搜索能力,克服差分进化算法易"早熟"的缺陷,在分析引起种群多样性下降及个体进化停滞原因的基础上,通过引入高斯变异操作,提出了基于高斯变异改进的差分进化算法(modified differ-ential evolution base on Gauss mutation,GMDE).数值仿真及2个工程优化问题的求解结果表明本文算法能有效避免"早熟"收敛,且在算法收敛速度和全局搜索能力上取得了较好的平衡.