二进制粒子群优化算法在化工优化问题中的应用

优化问题是化工过程的一个主要问题,而由化工问题建模所得到的优化问题大多较为复杂,此时要求的优化算法具有良好的优化性能。粒子群优化算法是新近发展起来的一种优化算法,但其对多极值函数的优化时,易陷局部极值。本文在分析粒子群优化算法的机理、考虑二进制比十进制更易于学习等的基础上,提出采用二进制表示粒子群优化算法,使每个粒子更易于从个体极值与全局极值中学习,从而使算法具有更强的搜索能力与更快的收敛速度,性能测试说明了所提出的算法是有效的．最后将算法用于求解换热网络的优化问题,取得良好效果。

Binary system particle swarm optimization algorithm for optimization of chemical processes

Optimization problem is principal problem in chemical industry process, and most of optimization problems come from chemical industry are complicated, so a good property algorithm was needed for solving it. Particle swarm optimization algorithm (PSO) was proposed recently, but it is easy to be trapped into local minima in optimizing multimodal function. Basing on analyzing the mechanism of the PSO, and considering binary system is easier to learn than decimal system, binary system particle swarm optimization was proposed (BPSO). Taking advantage of binary system, every particle can learn easily from present global optimal point and the optimal point of each particle in each of iteration, so BPSO has powerful global search property and converges fast. Experimental results show the proposed method was successful and it is efiiective introducing binary system to PSO. At last, the algorithm was applied to heat exchangers network problem and got satisfying results.