| 摘要: |
| 通过分析多目标数字印刷智能排活系统中各要素的相互关系，提出了PSO与局部搜索策略的混合算法，并引入了新的学习策略进行分层局部优化，用多目标分散搜索，逐步缩小复杂的搜索空间，改善了PSO算法的早熟收敛缺限，并取得了较高的求解质量。采用了一种随机键的编码方式，利用析取图编码将有序表作为优先决策，来决定发生冲突时各印刷活件的排列顺序。仿真实验验证了混合算法的有效性。 |
| 关键词: 混合粒子群（PSO） 多目标数字印刷智能排活 随机键编码 |
| DOI： |
| 分类号:TS808; TP18 |
| 基金项目:上海市科委基金项目(09220502700) |
|
| Optimization of Intelligent Multi-objective Dispatching System Based on Hybrid Particle Swarm |
|
MA Li-li, WANG Xiao-hong
|
|
Shanghai University of Science &Technology, Shanghai 200093, China
|
| Abstract: |
| By analyzing the mutual inner relationships among components in multi-intelligent printing dispatching system, an algorithm mixed with PSO and searching strategy for portion area was put forward. A new studying method to optimize the system partially in every layer was imported and the complicated searching space was reduced gradually by distribute searching, which solved the problem of premature convergence of PSO algorithm and achieved high solution quality. An encoding method with random keys was applied, which can make the ordered table to solve conflict. Simulation results showed that the hybrid algorithm is valid. |
| Key words: PSO intelligent multi-objective dispatching random key encoding |
渝公网安备 50010702501716号