论文部分内容阅读
粒子群算法是一种仿生进化算法,源于对鸟群觅食行为的模拟,由于其计算简单、快速,被广泛应用。但是基本粒子群算法在求解的过程中存在着全局搜索能力和局部求精能力两个性能之间的矛盾,算法容易陷入局部极值,进化后期的收敛速度慢。针对上述问题,提出了基于遗传交叉变异操作数的粒子群算法,用于优化网络权值。优化过程中基于遗传交叉算法采用了精英保存策略,在粒子群更新速度和位移的同时采用了线性递减权值调整的策略。此方法可以使粒子摆脱局部极值,继续优化,加快收敛速度。将粒子群算法与遗传算法相结合,用于控制倒立摆系统的平衡。仿真实验表明,改进粒子群算法优化了倒立摆系统控制器的设计,改善了控制效果。同时将此算法应用于倒立摆设备,实际控制效果理想。
Particle swarm optimization is a bionic evolutionary algorithm, which is derived from the simulation of the foraging behavior of flocks. It is widely used because of its simple and fast calculation. However, the basic particle swarm optimization has the contradiction between global search ability and local refinement ability in the process of solving. The algorithm is easy to fall into local extremum, and the evolutionary convergence rate is slow. In view of the above problems, a particle swarm optimization algorithm based on genetic crossover mutation operator is proposed to optimize network weights. In the optimization process, the elitist preservation strategy is adopted based on the genetic crossover algorithm, and the strategy of adjusting the weights of linear descending weights is adopted simultaneously with the update speed and displacement of particle swarm. This method can make the particle get rid of the local extreme value, continue to optimize, accelerate the convergence speed. Particle swarm optimization and genetic algorithm are combined to control the balance of inverted pendulum system. Simulation results show that the improved particle swarm optimization optimizes the design of inverted pendulum system controller and improves the control effect. At the same time this algorithm is applied to inverted pendulum equipment, the actual control effect is ideal.