摘要
建立以最小化提前和拖期时间、最小化炉重偏差为目标的混合整数线性规划模型,解决磁性材料成型-烧结两阶段生产调度问题.提出一种混合粒子群优化算法(HPSO)进行模型的求解,该算法采用基于订单的编码方式.针对粒子群算法易陷入局部最优,在迭代过程中引入模拟退火思想.改进粒子群算法的全局极值和个体极值选取方式,使算法尽快收敛到非劣最优解.生产现场实际数据仿真结果表明:该混合粒子群算法无论在求解精度,还是求解速度上均优于普通粒子群算法和遗传算法.
A mixed integer linear programming model was built to solve molding and sintering two stage production scheduling problem of magnetic material with the optimization objectives of minimizing earliness and tardiness time and minimizing furnace heavy deviation. A hybrid particle swarm optimization （HPSO） algorithm was proposed to solve the model. Encoding based on the order was adopted in the algorithm. For the particle swarm optimization （PSO） algorithm is easy to fall into local minima, simulated annealing was introduced in the iteration process. In order to make the algorithm to converge to the non-inferior opti- mal solution as soon as possible, the selection mode of PSO＇s global extreme and individual extreme was improved. Simulation results with actual data of production field showed that the proposed hybrid particle swarm algorithm is better than the general particle swarm algorithm and genetic algorithm （GA） either in solving precision or speed.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2013年第9期1517-1523,共7页
Journal of Zhejiang University：Engineering Science
基金
国家自然科学基金资助项目(61174187)
教育部基本科研业务费资助项目(N110208001)
东北大学启动基金资助项目(29321006)
关键词
多目标优化
粒子群优化算法
模拟退火
生产调度
磁性材料
multi-objective optimization
particle swarm optimization algorithm
simulated annealing
pro- duction seheduling~ magnetic materials
