Dual-stage self-adaptive differential evolution with complementary and ensemble mutation strategies

• •
  Integrate optimization experience and Gaussian probability distributions for creating novel double mutation strategy.
• •
  A two-stage parameter self-adaptation scheme is designed to dynamically adjust key parameters.
• •
  An enhanced population size reduction mode is proposed for allowing elite individuals to undergo more evolutionary generations.
• •
  LSHADE-Csode is proposed for large scale global optimization problems.

从种群进化角度进行了研究，经过两年的时间，深入验证和分析了算法性能，与CEC竞赛排名最前的算法进行了对比，整体上有新的突破，也存在一些问题和产生新的研究角度，需要进一步深化和分析。