论文部分内容阅读
自适应粒子群算法(AdaptiveParticle SwarmOptimization,APSO)是一种参数自适应的种群智能算法。该算法以种群的分布状态为依据区分优化过程中的不同状态自适应地调整算法参数。基于APSO算法具有参数自适应、快速收敛、全局搜索能力强等优点。将APSO算法应用于动态优化,通过采用按变量比例分配时间的方法构造时间变量,从而将其转化为无约束变量,通过时间变量与控制变量构造控制输入函数控制动态系统,使其达到最优。该方法提供一种转换时间变量约束的方法,使其能够作为一般优化问题,适用于其他类似演化类算法的动态性能的测试。最后,通过4个经典动态优化测试函数,比较APSO算法与蚁群算法,体现APSO算法处理动态优化的性能。
Adaptive Particle Swarm Optimization (APSO) is a parameter adaptive population intelligence algorithm. The algorithm adaptively adjusts the parameters of the algorithm by distinguishing the different states in the optimization process based on the distribution status of the population. Based on the APSO algorithm with the parameters of self-adaptive, fast convergence, global search ability and so on. The APSO algorithm is applied to dynamic optimization, and the time variable is constructed by using the method of proportionally allocating time according to the variable, so that the APSO algorithm can be transformed into an unconstrained variable and the control input function can be controlled by the time variable and the control variable to achieve the optimal control. This method provides a method of transforming time-variable constraints so that it can be used as a general optimization problem to test the dynamic performance of other similar evolutionary algorithms. Finally, four classic dynamic optimization test functions are compared to compare the APSO algorithm with the ant colony algorithm, which shows that the APSO algorithm can handle the performance of dynamic optimization.