多生产线作业协调调度的PSOGA混合优化算法

为了求解多目标多生产线调度问题,采用协同进化思想,提出了多种群PSOGA混合优化算法(MC-HPSOGA)。以最小化最大完工时间、最大化生产线利用率和最大化客户满意度为目标函数,建立了多生产线作业协调调度问题的多目标批量调度数学模型,并且设计最小批量动态分批策略,将MC-HPSOGA算法应用于BSPT公司角磨机装配线的多目标多生产线调度问题实例中,通过与粒子群(PSO)和遗传算法(GA)的比较,验证了MC-HPSOGA算法和模型的有效性。     

FMS在线刀具管理系统OLTMS

本文介绍了一种刀具管理软件框架结构，对于不同的ＦＭＳ配置及工件调度策略研制相应刀具管理软件时避免了不必要的重复劳动。另外还简要介绍了一种工业实用性启发式在线刀具动态调度算法。     

批次受限的双目标并行机等量分批调度

刀具是机械加工的重要资源,刀具数量决定了能同时加工的工件批次上限。针对此类批次受限的并行机等量分批调度问题,建立了刀具数量和刀具更换等特有约束,构建了双目标数学规划模型,以完工时间和交付时间偏差最大值的最小化为目标,力求提高生产效率、保证同一订单内工件准时且同步交付。提出融入快速非支配排序的双目标鲸鱼群算法,通过引入虚拟占位符,设计出允许批次数变化的定长编码;将多点保留交叉策略融入到个体移动规则设计中,以便跨越子批赋值向量、保持子批序列不变、增强解的多样性;将非劣个体保留策略嵌入到邻域搜索中,指导算法跳出局部最优。实验结果表明,所提算法的收敛性与多样性显著优于对比算法,有利于同时达到准时、高效的生产目标。     

基于深度学习算法的调度自动化云平台任务优化策略研究

分析了云平台任务调度的特点和目标,从任务调度算法入手,提出了基于改进粒子群算法的电力调度自动化系统的人工智能方法,开发了云计算操作的模型。基于该算法和物理模型的运行控制考虑了 QoS 要求和平台云居民的环境负载平衡,可以有效提高所提电力调度自动化系统的云平台任务调度的效率。以电力自动化云平台为分析对象,研究其架构,将修正的 PSO 算法与云资源调度模型的结构拓扑相结合,建立三级数据节点,给出了基于改进 PSO 的云平台调度模型,旨在提高云计算资源配置效率,改善云服务质量,解决电力调度自动化系统的任务调度问题。     

基于遗传-粒子群混合算法的柔性作业车间多资源调度问题

在传统柔性作业车间调度问题(FJSP)中加入运输和装配环节,提出一种柔性作业车间多资源调度问题(MRFJSP),以完工时间最短为目标建立了包含加工、运输和装配的柔性作业车间调度模型。为了提高传统遗传算法(GA)在车间调度问题中的寻优能力,将粒子群算法(PSO)的寻优过程进行改进并与遗传算法进行结合,提出一种带保优策略的遗传-粒子群混合算法,利用单层编码对模型进行求解。通过算例验证了模型的可行性,并将提出的混合算法与遗传算法和粒子群算法进行比较,证明了混合算法的优越性。     

基于粒子群优化的开放式车间调度

开放式车间调度(OSP)是重要的调度问题,它在制造领域中的应用非常广泛。优化调度算法是调度理论的重要研究内容。基于人工智能的元启发式算法是解决该问题的常用方法。分析了一种新的元启发式算法——粒子群优化(PSO)在信息共享机制上的缺陷,提出新的基于群体智能的信息共享机制。在该信息共享机制的基础上, 设计新的基于PSO的元启发式调度算法——PSO-OSP。该算法利用问题的邻域知识指导局部搜索,可克服元启发式算法随机性引起的盲目搜索。该算法应用于开放式车间调度问题的标准测试实例。仿真结果显示,PSO-OSP算法在加快收敛速度的同时提高了开放式车间调度解的质量。     

多Agent网络协议在刀具调度策略中的应用

主要从机械制造企业自动化制造系统中刀具调度存在的问题出发,利用多Agent合同网协议优化刀具调度策略,从而实现刀具调度过程中多目标问题的求解。     

基于多目标的动态车间调度问题的策略研究

动态事件在车间调度过程中不可避免,严重影响加工过程,并导致订单交货延期、生产效率不高等现象发生。而目前针对动态事件的研究多针对单一事件,对多种动态事件的研究更偏向于算法研究。由此,通过研究完全重调度策略,结合粒子群(Particle Swarm Optimization,PSO)算法,对新加类、机器占用类、交换类以及延迟类这4类动态事件进行处理。通过多规模的测试,以及对动态事件发生在不同时间段的模拟仿真,以重调度结果相对于原始调度结果的扰动影响率为指标,验证重调度策略的实用性与鲁棒性。     

FMS刀具管理优化技术研究与实验系统开发

设计开发了基于嵌入式控制技术的FMS刀具控制管理实验系统,其结构紧凑、功能完善、配置灵活,能够模拟FMS实际生产加工中刀具控制过程,既可应用于相关理论与技术的实验验证,也可应用于实验教学和培训。并运用遗传算法在有限刀具资源约束下对FMS刀具调度优化问题进行了研究,采用静态和动态两级调度优化策略,对减少换刀次数,提高FMS生产效率具有积极的作用。     

加工工序非紧密衔接的车间调度策略

针对以往车间调度算法忽略非紧密衔接工序调度,从而导致调度结果不精确这一情况,提出了一种考虑工序间空闲时间延迟约束的调度策略,即解决加工工序非紧密衔接调度问题的方法。该方法在对由延迟约束造成的工序非紧密衔接型调度问题进行定义和研究的基础上,提出了标准工序、延迟工序和扩展加工工艺树概念。设计了一种将延迟约束转换为顺序约束的转换策略,使得转换后的非紧密衔接工序变为虚拟紧密衔接工序,并可采用经典车间调度算法调度。实例验证表明,该调度策略可有效地解决非紧密衔接工序调度问题,可在不增加算法复杂度的情况下,得到更为实用的调度方案。     

FMS刀具流死锁控制策略与分派算法的研究

刀具流死锁控制及刀具分派问题是柔性制造系统(FMS)调度中的重要核心内容,为实现合理的死锁回避和刀具分派,在前期研究的基础上,通过建立刀具申请分配图,分析了死锁的相关性质,指出了FMS中刀具流死锁的两大根源是工件选择的不合理和刀具分派的不合理。提出了一种解决刀具流调度问题的两级死锁控制策略,建立了动态调度原理模型,给出了死锁检测算法和刀具分派算法。分析表明该策略与算法能够使得刀具流死锁的两大根源均得以回避,实现了刀具流死锁控制与刀具分派。     

DEADLOCK-FREE SCHEDULING METHOD BASED ON PETRI NET AND GA

An efficient algorithm for finding an optimal deadlock-free schedule in a manufacturingsystem with very limited buffer is presented. This algorithm is based on the effective genetic algorithm(GA) search method, and a formal Petri net structure is introduced to detect the token player assuringdeadlock-free. In order to make the scheduling strategy generated by GA meet the required constraintof deadlock-free, Petri net is involved to make the implementation of the job scheduling in an FMSdeadlock-free. The effectiveness and efficiency of the proposed approach is illustrated by using anexample.     

Intelligent scheduling of FMSs with inductive learning capability using neural networks

With the growing uncertainty and complexity in the manufacturing environment, most scheduling problems have been proven to be NP-complete and this can degrade the performance of conventional operations research (OR) techniques. This article presents a system-attribute-oriented knowledge-based scheduling system (SAOSS) with inductive learning capability. With the rich heritage from artificial intelligence (AI), SAOSS takes a multialgorithm paradigm which makes it more intelligent, flexible, and suitable than others for tackling complicated, dynamic scheduling problems.SAOSS employs an efficient and effective inductive learning method, a continuous iterative dichotomister 3 (CID3) algorithm, to induce decision rules for scheduling by converting corresponding decision trees into hidden layers of a self-generated neural network. Connection weights between hidden units imply the scheduling heuristics, which are then formulated into scheduling rules. An FMS scheduling problem is also given for illustration. The scheduling results show that the system-attribute-oriented knowledge-based approach is capable of addressing dynamic scheduling problems.     

Integrated scheduling of flexible manufacturing system using evolutionary algorithms

Effective sequencing and scheduling of the material handling system (MHS) have an impact on the productivity of the flexible manufacturing system (FMS). The MHS cannot be neglected while scheduling the production tasks. It is necessary to take into account the interaction between machines and MHS. This paper highlights the importance of integration between production schedule and MHS schedule in FMS. The Giffler and Thompson algorithm with different priority dispatching rules is developed to minimize the makespan in the FMS production schedule. Its output is used for MHS scheduling where the distance traveled and the number of backtrackings of the automated-guided vehicles are minimized using an evolutionary algorithms such as an ant colony optimization algorithm and particle swarm optimization (PSO) algorithm. The proposed evolutionary algorithms are validated with benchmark problems. The results available for the existing algorithms are compared with results obtained by the proposed evolutionary algorithms. The analysis reveals that PSO algorithm provides better solution with reasonable computational time.     

FMS调度与控制的统一建模与集成设计

分析了FMS调度与控制统一建模与集成设计的必要性，从FMS运行过程的IDEF0功能建模入手，建立了基于模块化彩色赋时Petri网（CTPN）的FMS调度与控制的统一模型。该模型不仅可描述FMS的调度问题，而且能表达FMS的运行控制逻辑与仿真，从而为FMS控制系统的集成设计奠定了基础。为简化FMS控制系统的设计，提出了虚拟工作站的概念与方法，将FMS控制器的部分控制功能下放到低层的工作站控制层。基于上述CTPN统一模型，开发了FMS调度与控制的集成设计原理和方法。     

Scheduling decisions in FMS using a heuristic approach

The paper deals with the multilevel scheduling decisions of a Flexible Manufacturing System (FMS) to generate realistic schedules for the efficient operation of the FMS. The primary concern of an Operations Management System (OMS) for a FMS is production scheduling, Material Handling System (MHS) scheduling, Automated Storage/Retrieval System (AS/RS) operation and control and tool management. Scheduling is a critical issue and determines how efficiently the production resources are utilised and how the selected parts are affected in the system. In this paper, the integrated scheduling of FMS, namely, the production scheduling conforming with the MHS scheduling, is addressed. An enumerative heuristic is used, namely Giffler and Thompson, which is an evolutionary combining a Genetic Algorithm (GA) and a stochastic neighborhood search technique using a Simulated Annealing (SA) algorithm is employed.A. Noorul Haq received his PhD in Manufacturing Management from the Indian Institute of Technology (IIT), New Delhi, India, a Master of Engineering degree from Madras University, and a Bachelor of Engineering degree from Annamalai University, India. He is currently an assistant professor in the Department of Production Engineering, Regional Engineering College, Tiruchirappalli 620015, India. His research interests include aggregate production planning, facility layout and scheduling and optimisation techniques.T. Karthikeyan received his Master's degree from Bharathidhasan University. Currently, he is a research scholar in the Department of Production Engineering, Regional Engineering College, Tiruchirappalli 620015, India. His research interests includes facility layout, FMS and simulation.M. Dinesh received his bachelor's degree in engineering from Bharathiar University, India and is currently working on his master's degee in engineering in the Department Of Production Engineering, Regional Engineering College, Tiruchirappalli 620015, India. His current research interests are in optimisation techniques, facility layout and scheduling.     

基于粒子群优化和变邻域搜索的混合调度算法

提出了用于解决作业车间调度问题的离散版粒子群算法.该算法采用基于工序的编码和新的位置更新策略,使具有连续本质的粒子群算法直接适用于调度问题.同时,针对粒子群算法容易陷入局部最优的缺陷,利用粒子群算法和变邻域搜索算法的互补性能,设计了粒子群-变邻域搜索算法、改进的粒子群算法、粒子群-变邻域搜索交替算法和粒子群-变邻域搜索协同算法4种混合调度算法.仿真结果表明,混合算法能够有效地、高质量地解决作业车间调度问题.     

一种有约束FMS资源调度的新方法

提出利用神经网络和模拟退火技术来求解有约束有ＦＭＳ资源调度问题的一种新方法。有约束的ＦＭＳ资源调度被分解为一系列时间间隔的调度,这些时间间隔的调度由事件驱动,随着这些时间间隔的调度的完成,整个调度过程结束。     

柔性加工环境中机器和AGV的集成调度

研究由几台加工中心（WS）和一台（多台）自动导向小车（AGV）组成的柔性制造系统（FMS）的调度问题,描述了采用AGV的FMS高度的特点,建立了可变工艺路径、包含AGV的FMS调度问题的模型,形成一种新的基于有向图的可变路径表示方法,采用结合启发式规则优点的遗传算法（GA）同时调度机器和AGV,使得AGV调度成为FMS集成环境下调度的一部分,提出了面向可变路径特征的交叉算子,采用自适应的交叉、变异策略和灵活的群体控制策略,对几个调度实例进行了计算。