摘要
针对化工过程系统优化中广泛存在着边值固定的动态优化问题,该问题的求解数学上还没有有效的方法,现今的方法之一是将问题转化为多目标优化问题。本文在粒子群优化(PSO)算法的基础上,提出在PSO算法中加入惩罚项,同时对局部极值与全局极值作进一步的调整,使PSO算法适用于求多目标优化问题理想有效解,该算法对多目标问题起到边优化边求理想有效解的功效;即只用一步即可求理想有效解,这使得在求解速度上大为加快。最后将其用于间歇反应器的最佳反应温度边值固定动态优化控制的实际运用中,取得良好效果。
In order to make particle swarm optimization （PSO） apply to multi-objective optimization problem ideal pareto solution （IPS）, a penalty term is injected into PSO and makes further adjustment for partial extremum and global extremum based on PSO. This algorithm can solve IPS while optimizing for multi-objective problem, viz. it can solve IPS in just one step and accelerates the speed for solving consumedly. An intermittence reactor dynamic system temperature optimal control with fixed boundary problem is employed for examining the validity of the proposed method. Experimental results show that the method is effective.
出处
《化工自动化及仪表》
EI
CAS
2006年第4期18-21,共4页
Control and Instruments in Chemical Industry
基金
国家自然科学基金资助项目(20276063)
