摘要
随着制造业向智能化和柔性化方向发展，柔性制造系统（FMS）在应对多样化生产需求方面展现出显著优势，但其复杂性对调度与优化提出了更高要求。本研究旨在针对FMS中的动态调度问题，提出一种融合智能算法的优化方法以提升系统效率和资源利用率。具体而言，研究基于遗传算法、粒子群优化及深度强化学习等技术，设计了一种多目标混合智能调度算法，能够同时考虑任务优先级、设备负载均衡及加工时间等因素。通过构建仿真模型并结合实际工业数据进行验证，结果表明该算法在处理复杂调度场景时具有更高的适应性和鲁棒性，相较于传统方法可将平均完工时间减少约15%，并显著降低设备闲置率。此外，研究创新性地引入了动态权重调整机制，使算法能够在不同工况下自动优化参数配置，从而进一步提高调度决策的质量。总体而言，本研究不仅为FMS的智能调度提供了有效解决方案，还为相关领域的算法设计与应用拓展奠定了理论基础。
关键词：柔性制造系统；智能调度算法；多目标优化；动态权重调整；深度强化学习
Abstract
With the development of manufacturing towards intelligence and flexibility, flexible manufacturing systems (FMS) have demonstrated significant advantages in addressing diverse production demands; however, their complexity imposes higher requirements on scheduling and optimization. This study focuses on the dynamic scheduling problem in FMS and proposes an optimization approach integrating intelligent algorithms to enhance system efficiency and resource utilization. Specifically, a multi-ob jective hybrid intelligent scheduling algorithm is designed based on genetic algorithms, particle swarm optimization, and deep reinforcement learning, which can simultaneously consider task priority, equipment load balancing, and processing time. By constructing a simulation model and validating it with real industrial data, the results indicate that the proposed algorithm exhibits superior adaptability and robustness in handling complex scheduling scenarios, reducing the average completion time by approximately 15% compared to traditional methods while significantly decreasing equipment idling rates. Moreover, this research innovatively incorporates a dynamic weight adjustment mechanism, enabling the algorithm to automatically optimize parameter configurations under varying operating conditions, thereby further improving the quality of scheduling decisions. Overall, this study not only provides an effective solution for intelligent scheduling in FMS but also lays a theoretical foundation for the design and application expansion of algorithms in related fields.
Keywords：Flexible Manufacturing System; Intelligent Scheduling Algorithm; Multi-ob jective Optimization; Dynamic Weight Adjustment; Deep Reinforcement Learning
目 录
摘要 I
Abstract II
一、绪论 1
(一) 柔性制造系统与智能调度的研究背景 1
(二) 智能调度与优化算法的研究现状 1
(三) 本文研究方法与技术路线 2
二、柔性制造系统的调度基础 2
(一) 柔性制造系统的基本概念与特征 2
(二) 调度问题的数学建模分析 3
(三) 关键约束条件及其影响因素 3
三、智能调度算法的设计与实现 4
(一) 常见智能调度算法的分类与比较 4
(二) 遗传算法在调度中的应用研究 4
(三) 粒子群优化算法的改进与实现 5
四、优化算法的性能评估与案例分析 5
(一) 性能评估指标体系的构建 6
(二) 实验设计与仿真环境搭建 6
(三) 典型案例分析与结果讨论 7
结 论 8
参考文献 9
