一维下料问题的自适应广义粒子群优化求解

Adaptive General Particle Swarm Optimization for One-Dimension Cutting Stock Problem

针对现有粒子群优化算法在求解组合优化问题时粒子速度迭代难以定义的问题,首先将粒子群优化算法与遗传算法相结合,利用交叉算子、变异算子,提出一种广义粒子群优化算法来求解一维下料问题;然后引入模拟退火算法作为自适应策略,避免算法陷入局部最优.仿真实验结果表明,采用自适应广义粒子群优化算法求解一维下料问题具有高效性和鲁棒性.

In the existing particle swarm optimization algorithms,the iteration of particle velocities is difficult to define for combinatorial optimization problems.In order to solve this problem,this paper proposes a general particle swarm optimization algorithm to solve the one-dimension cutting stock problem.In the proposed algorithm,the existing particle swarm optimization algorithm is combined with the genetic algorithm,the crossover operator and the mutation operator in genetic algorithm are employed,and an adaptive strategy based on the simulated annealing algorithm is introduced to avoid the premature convergence of particle swarm.Simulated results demonstrate that the proposed algorithm is effective and robust in solving the one-dimension cutting stock problem.