基于节点流的流程工业生产物流调度方法及应用

流程工业生产物流动态平衡调度是典型的多工序、多目标、多约束的多机并行调度问题,难以解析建模.为此,提出了节点流控制及切换的概念,建立了生产调度模型,并采用多目标分层求解调度模型.在此基础上,提出了流程工业生产的物流平衡调度方法.实际应用表明,所提出方法提高了物流动态调度的准确性和生产效率.     

基于深度强化学习的微电网在线优化调度

提出一种基于深度强化学习的微电网在线优化调度策略.针对可再生能源的随机性及复杂的潮流约束对微电网经济安全运行带来的挑战,以成本最小为目标,考虑微电网运行状态及调度动作的约束,将微电网在线调度问题建模为一个约束马尔可夫决策过程.为避免求解复杂的非线性潮流优化、降低对高精度预测信息及系统模型的依赖,设计一个卷积神经网络结构学习最优的调度策略.所提出的神经网络结构可以从微电网原始观测数据中提取高质量的特征,并基于提取到的特征直接产生调度决策.为了确保该神经网络产生的调度决策能够满足复杂的网络潮流约束,结合拉格朗日乘子法与soft actor-critic,提出一种新的深度强化学习算法来训练该神经网络.最后,为验证所提出方法的有效性,利用真实的电力系统数据进行仿真.仿真结果表明,所提出的在线优化调度方法可以有效地从数据中学习到满足潮流约束且具有成本效益的调度策略,降低随机性对微电网运行的影响.     

机床产品制造系统能效的最优控制

本文结合机床产品制造系统的能量流特性,研究机床产品制造系统能效的最优控制.首先,利用无线传感器网络,构建制造系统能效感知网络,并设计了网络能量高效的通信协议,实时获取制造系统的能效数据.进而,利用能效感知数据,分别从单机设备局部优化与综合资源全局优化两方面,设计能效优化控制算法.根据单机设备任意两工步间空载能耗特性,给出单机设备空载能效最优控制模型.同时,建立以缩短生产周期、减少机器空转时间、提高产品合格率为优化目标的综合生产资源能效多目标优化方案.考虑到所论综合资源能效优化问题是离散组合优化问题,本文提出了文化基因支配排序粒子群算法进行求解,并采用层次分析(analysis hierarchy process,AHP)决策方法从Pareto解集中选取最优综合能效的优化策略.最后,利用实例与仿真相结合的方法,验证了文中所提方法的有效性.     

微电子生产过程调度问题基于指标快速预报的分解算法

微电子生产过程调度问题具有规模大和约束复杂等特点,如菜单、Setup时间和组批约束等,其优化调度具有一定难度.针对以最小化平均流经时间为调度目标的较大规模微电子生产过程调度问题,提出一种基于指标快速预报的分解方法(DM-IFP).首先,通过松弛不可中断约束,设计一种代理方法,即基于机器负载的操作完工时间快速预测方法(CTP-ML);其次,设计基于CTP-ML的问题分解方法,将原问题迭代分解为多个连续交迭的子问题;然后,提出一种基于双信息素的蚁群算法(ACO-D)用于求解分解后的子问题,其全局调度目标采用CTP-ML获取,有效保证了全局优化性能;最后,针对一些不同规模的仿真数据,将所提出方法与一些代表性的算法进行详尽的数值对比,计算结果表明所提出方法在所获解的质量和收敛性上均有改善.     

多目标环境下的生产资源平衡和优化调度研究

基于调度周期订单所映射的工件集合,讨论了以生产资源和工艺方案为约束的多工艺方案推荐方法,有目的地为生产调度进行工艺选择,提高生产调度的运行效率.基于遗传算法,考虑设备负荷均衡、关键设备利用率、所有设备利用率等多目标,进行了基于可用资源的生产调度基础数据优化处理.结合实例,根据不同的目标得到了有效的可选工艺方案,仿真结果表明,经过生产资源平衡后,较易得到满意的调度方案.     

基于负荷平衡的电动汽车模糊多目标充电调度算法

居民小区电动汽车（EV）的单相充电方式导致配电网出现三相不平衡和负荷峰谷差问题,因此提出基于负荷平衡的EV模糊多目标充电调度策略。基于三相网络,将总延迟时间和充电平衡作为目标函数,考虑三相不平衡度和负荷峰谷差等约束,建立静态和在线调度问题下EV充电调度模型。采用改进非支配排序遗传算法-Ⅱ（NSGA-Ⅱ）进行多目标求解,通过设计交叉算子、自适应调整变异概率和局部优化等来优化结果。通过设置一定容量的外部档案和拥挤距离判定来获得Pareto最优前沿,并用模糊隶属度方法得到折中最优解。最后,通过算例分析可同时活动充电点和三相不平衡度的不同取值对优化结果的影响,并与无序充电进行比较,验证了所提模型和策略的有效性。     

就绪时间受限的负荷单机环境下返工工件重调度方法

研究了返工工件的单机重调度问题.在初始调度中初始工件带有不同的就绪时间,优化目标为最小化初始工件等待时间和;重调度时在满足每个初始工件最大等待时间约束情况下安排返工工件的生产,优化目标为最小化所有工件等待时间和.文中首先建立了RRSM （Rescheduling for reworks on single machine）问题模型,并证明其为NP难问题.然后,提出并证明了三个RRSM问题性质,进而根据诸性质设计了求解RRSM问题的动态插入启发式（Dynamic insert heuristic,DIH）算法.证明了应用DIH算法能在多项式时间内求得两种特殊RRSM问题的最优解. 最后,分析了DIH算法解的特点,给出了最优解的判定方法,并通过算例说明了DIH算法的有效性.     

企业管理中多工作任务下合理调度模型仿真

在企业管理中工作任务调度效率优化问题的研究中,由于在多任务的调度过程中,任务之间的关联性很复杂,受到相互之间调度顺序的影响较大.传统的工作任务调度模型以单一线性顺序最优为基础,一旦任务过多,会造成调度中不同任务调度相互矛盾加大,模型稳定性会下降,缺少灵活性,造成调度的效率不高.提出采用多约束调度模型的企业管理中多工作任务下合理调度方法.详细阐述了企业管理中多工作任务调度的原理,构建AON网络模型,针对网络模型进行虚调度,在求解空间中搜索最优解,得到企业管理中多任务合理调度方法.实验结果表明,利用改进算法进行企业管理中多工作任务调度,能够提高资源利用率,并且提高调度效率,避免多工作掺杂的情况下造成的调度模型稳定性降低的缺陷.     

单机调度问题对偶集结迭代算法

具有到达时间约束、目标为最小化加权完工时间之和的单机调度问题是一个典型的NP-hard问题,采用时间下标建模的线性规划松弛方法可提供一个很强的下界,但优化求解存在维数困难.为此,本文提出了一种对偶集结优化策略,通过选择一个衰减集结矩阵集结对偶乘子变量,利用对偶理论获得模型的约束集结,从而降低计算复杂度.同时分析了集结模型的结构特性,并提出一种迭代算法来改善下界.仿真结果表明对偶集结迭代算法能够减少计算时间,同时改善下界性能,适用于大规模调度问题.     

一种基于神经网络的生产调度方法

提出解决具有开、完工期限制的约束Job-shop生产调度问题的一种神经网络方法. 该方法通过约束神经网络,描述各种加工约束条件,并对不满足约束的开工时间进行相应调 节,得到可行调度方案;然后由梯度搜索算法优化可行调度方案,直至得到最终优化可行调度 解.理论分析、仿真实验表明了方法的有效性.     

Development of hybrid evolutionary algorithms for production scheduling of hot strip mill

A hot strip mill (HSM) produces hot rolled products from steel slabs, and is one of the most important production lines in a steel plant. The aim of HSM scheduling is to construct a rolling sequence that optimizes a set of given criteria under constraints. Due to the complexity in modeling the production process and optimizing the rolling sequence, the HSM scheduling is a challenging task for hot rolling production schedulers. This paper first introduces the HSM production process and requirements, and then reviews previous research on the modeling and optimization of the HSM scheduling problem. According to the practical requirements of hot rolling production, a mathematical model is formulated to describe two important scheduling sub-tasks: (1) selecting a subset of manufacturing orders and (2) generating an optimal rolling sequence from the selected manufacturing orders. Further, hybrid evolutionary algorithms with integration of genetic algorithm (GA) and extremal optimization (EO) are proposed to solve the HSM scheduling problem. Computational results on industrial data show that the proposed HSM scheduling solution can be applied in practice to provide satisfactory performance.     

A two-stage scheduling method for hot rolling and its application

Hot rolling scheduling is a difficult problem in the steel processing industry. It involves many objectives and constraints in both technical and practical respects. A two-stage scheduling method is proposed in this paper. Batch planning of staple material is formulated as a VRPTW, which is solved with a modified PGA. Then, batches of the established units are optimized by adjusting rolling sequences using intelligent search algorithms to reach higher hot charge ratios. This method has been applied to a hot strip mill belonging to Baosteel in China. Our results demonstrate that the proposed technique can improve production efficiency and offer significant economic benefits.     

基于PCTSP的热轧单元计划模型与算法

根据钢铁企业热轧产品生产工艺约束条件,将热轧生产轧制单元计划模型归结为奖金收集旅行商问题,设计了蚁群最优化算法对模型进行求解.引用某钢铁企业热轧生产轧制单元计划编制的实际问题对模型和算法进行了验证,并与遗传算法的求解结果进行了对比.实验结果表明模型和算法的优化效果和时间效率都是令人满意的.该模型和算法经过改进后可应用到包含多个轧制单元计划的轧制批量计划优化问题中.     

A case of rule-based heuristics for scheduling hot rolling seamless steel tube production

Abstract: A production scheduling problem for hot rolling seamless steel tube at Tianjin Pipe Corporation of China is studied. Considering the complexity of the problem and the acceptable time for solving it, a rule-based heuristic approach is proposed and implemented. The proposed approach is a bottleneck scheduling method and considers simultaneously all production processes in three production units and 'optimizes' them as a whole. Additionally, the running result shows, on average, that a 3% increase in throughput and a 5% reduction in late deliveries have been achieved since the system implementation.     

Dual-Objective Mixed Integer Linear Program and Memetic Algorithm for an Industrial Group Scheduling Problem

Group scheduling problems have attracted much attention owing to their many practical applications. This work proposes a new bi-objective serial-batch group scheduling problem considering the constraints of sequence-dependent setup time, release time, and due time. It is originated from an important industrial process, i.e., wire rod and bar rolling process in steel production systems. Two objective functions, i.e., the number of late jobs and total setup time, are minimized. A mixed integer linear program is established to describe the problem. To obtain its Pareto solutions, we present a memetic algorithm that integrates a population-based nondominated sorting genetic algorithm II and two single-solution-based improvement methods, i.e., an insertion-based local search and an iterated greedy algorithm. The computational results on extensive industrial data with the scale of a one-week schedule show that the proposed algorithm has great performance in solving the concerned problem and outperforms its peers. Its high accuracy and efficiency imply its great potential to be applied to solve industrial-size group scheduling problems.      

Two-station single-track railway scheduling problem with trains of equal speed

In this paper, the single-track railway scheduling problem with two stations and several segments of the track is considered. Two subsets of trains are given, where trains from the first subset go from the first station to the second station, and trains from the second subset go in the opposite direction. The speed of trains over each segment is the same. A polynomial time reduction from the problem under consideration to a special case of the single-machine equal-processing-time scheduling problem with setup times is presented. Different polynomial time algorithms are developed for special cases with divers objective functions under various constraints. Moreover, several theoretical results which can be ranked in a series of similar investigations of NP-hardness of equal-processing-time single-machine scheduling problems without precedence relations are obtained.     

A hybrid Lagrangian-simulated annealing-based heuristic for the parallel-machine capacitated lot-sizing and scheduling problem with sequence-dependent setup times

This paper examines the parallel-machine capacitated lot-sizing and scheduling problem with sequence-dependent setup times, time windows, machine eligibility and preference constraints. Such problems are quite common in the semiconductor manufacturing industry. In particular, this paper pays special attention to the chipset production in the semiconductor Assembly and Test Manufacturing (ATM) factory and constructs a Mixed Integer Programming (MIP) model for the problem. The primal problem is decomposed into a lot-sizing subproblem and a set of single-machine scheduling subproblems by Lagrangian decomposition. A Lagrangian-based heuristic algorithm, which incorporates the simulated annealing algorithm aimed at searching for a better solution during the feasibility construction stage, is proposed. Computational experiments show that the proposed hybrid algorithm outperforms other heuristic algorithms and meets the practical requirement for the tested ATM factory.     

冷轧机组批量作业计划模型与算法

针对编制冷轧机组作业计划受到钢卷宽度跳跃、入口厚度跳跃和出口厚度跳跃等多个工艺约束的问题, 把排产过程归纳为非对称双旅行商问题, 建立了冷轧机组生产作业计划的Pareto多目标模型. 提出了基于Pareto非支配集的自适应多目标蚁群算法, 利用自适应蚁群算法和Pareto非支配集思想, 综合考虑多个目标, 自适应地提供蚂蚁路径搜索参数, 并对得到的非支配解集对应路径更新信息素, 引导蚂蚁向最优解集方向搜索, 最终提供多个可行的批量作业计划, 根据生产要求从中选择合适的最优排产结果. 利用某冷轧薄板厂实际的生产数据进行仿真实验, 表明模型与算法在冷轧机组批量作业计划编制过程中具有可行性.     

考虑订单扰动因素的热轧重调度分布估计算法

热轧工序作为钢铁生产的核心环节,具有严格的生产连续性和复杂的产品工艺要求,而紧急订单的随机到达和紧急交货期要求会对生产连续性和质量稳定性产生不利影响。针对这类紧急订单插入的动态事件,提出一种热轧重调度优化方法。首先,分析了订单扰动因素对调度方案的影响,并以最小化订单拖期惩罚和板坯跳跃惩罚加权和为优化目标,建立了热轧重调度问题的数学模型。然后,设计了热轧重调度分布估计算法（EDA）。该算法针对紧急订单的插入式处理方式,提出一种基于插入位置的整数编码方案;结合模型特征设计了概率模型;并综合考虑目标与约束,定义了基于惩罚值的适应度函数。通过实际生产数据进行仿真实验,验证了模型和算法的可行性和有效性。     

基于预防维护的单机调度问题

在两种维护约束下,研究完工时间之和最小化的单机调度问题.第1种维护约束是,固定周期预防维护;第2种维护约束是,机器工作期间可连续加工的最大工件个数受限.对于这种带有约束的调度问题,根据问题的规模,采用4种方法进行求解.针对小规模问题,建立一个二值整数规划模型,并根据最优解的特性制定剪枝规则,进而给出分支定界算法.针对中、大规模问题,采用遗传算法进行求解,为缓解遗传算法中常见的早熟问题,对变异算子进行改进,采用动态变异方法,提出动态遗传算法.最后通过仿真实验对各种算法进行性能评估.