现实工业生产应用中存在大量的混合变量优化问题,这类问题的决策变量既包含连续变量,又包含离散变量.由于决策变量为混合类型,导致问题的决策空间变得不规则,采用已有的方法很难进行有效求解.引入协同进化策略,提出一种基于竞争式协同进化的混合变量粒子群优化算法(competitive coevolution based PSO,CCPSO).设计基于容忍度的搜索方向调整机制来判断粒子的进化状态,从而自适应地调整粒子的搜索方向,避免陷入局部最优,平衡了种群的收敛性和多样性;引入基于竞争式协同进化的学习对象生成机制,在检测到粒子进化停滞时为每个粒子生成新的学习对象,从而推动粒子的进一步搜索,提高了种群的多样性;采用基于竞争学习的预测策略为粒子选择合适的学习对象,充分利用了新旧学习对象的学习潜力,保证了算法的收敛速度.实验结果表明:相比其他主流的混合变量优化算法,CCPSO可以获得更优的结果.
Mixed-variable Particle Swarm Optimization Algorithm Based on Competitive Coevolution
For the current algorithm,it is difficult to obtain the available solution due to the irregularity of problem decision space caused by the numerous mixed variable optimization problems during real industrial applications.The coevolution strategy is introduced and a mixed variable particle swarm optimization algorithm(CCPSO)based on competitive coevolution is proposed.The search direction adjustment mechanism based on tolerance is designed to judge the evolution state of particles,adaptively adjust the search direction of particles,avoid falling into local optimum,and balance the convergence and diversity of the population.The learning object generation mechanism is adopted for each particle to generate new learning objects when particle evolution stagnation is detected to promote the evolution of particles and improve the diversity of the population.The prediction strategy based on competitive learning is applied to select the appropriate learning objects for particles,which makes full use of the learning potential of new and old learning objects and ensures the convergence speed of the algorithm.Experimental results show that,CCPSO can obtain the better results compared with the other main mixed variable optimization algorithms.
NETL
NSTL
万方数据
