杂交粒子群算法在工程几何约束求解中的应用

The Application of Crossbreeding Particle Swarm Optimizer in the Engineering Geometric Constraint Solving

工程几何约束问题等价于求解一系列非线性方程组。粒子群算法是一种进化计算方法。它通过创建更好的种群搜索解空间。新的种群基于新的个体而产生。借鉴遗传算法的思想,提出了杂交PSO算法的概念。粒子群中的粒子被赋予一个杂交概率。在每次迭代中,依据杂交概率选取指定数量的粒子放入一个池中。池中的粒子随机的两两杂交,产生同样数目的孩子粒子,并用孩子粒子代替父母粒子,以保持种群的粒子数目不变。实验表明该方法是有效的。

Engineering geometric constraint problem is equivalent to the problem of solving a set of nonlinear equations substantially. PSO is an evolution computing method. It searches the solution space by creating a better next swarm. The new swarm is produced based on the new individuals. The particle in the particle swarm is evaluated a crossbreeding probability. In every iteration it chooses a definite number particle to put into the pool according to the crossbreeding probability. The particles in the pool crossbreed each other and produce the same child particles. It replaces the parent particles for the child particles in order to maintain the particle number invariable. The experiment indicates that the algorithm is effective.