基于NR-Transformer的集群作业运行时间预测

高性能集群的作业调度通常使用作业调度系统来实现，准确填写作业运行时间能在很大程度上提升作业调度效率。现有的研究通常使用机器学习的预测方式，在预测精度和实用性上还存在一定的提升空间。为了进一步提高集群作业运行时间预测的准确率，考虑先对集群作业日志进行聚类，将作业类别信息添加到作业特征中，再使用基于注意力机制的NR-Transformer网络对作业日志数据建模和预测。在数据处理上，根据与预测目标的相关性、特征的完整性和数据的有效性，从历史日志数据集中筛选出7维特征，并按作业运行时间的长度将其划分为多个作业集，再对各作业集分别进行训练和预测。实验结果表明，相比于传统机器学习和BP神经网络，时序神经网络结构有更好的预测性能，其中NR-Transformer在各作业集上都有较好的性能。     

Smart utilization of renewable energy sources in a microgrid system integrated with plug‐in hybrid electric vehicles

Mass roll‐out of plug‐in hybrid electric vehicles (PHEVs) and significant penetration of renewable energy sources in distribution system play a major role in delivering low carbon environment. However, placing and utilizing these units randomly result in overloading, increased power loss, and reduced voltage profile. This paper responds to these technical challenges by using a strategic placement method for locating the distributed generation (DG) and the charging station (CS) of PHEVs in a multi‐zone distribution system. For simultaneously scheduling of these units in each zone, the smart energy management framework is proposed in this paper. Apart from usual energy management constraints, this paper also incorporates the real‐time constraints involving the capacity of PHEV batteries, the mobility pattern, and the power level of the charging infrastructure. The simulation studies are carried out for each hour of a day. To cope with this time constraint execution, particle swarm optimization algorithm‐based approach is used. The proposed framework is tested in IEEE 33 and IEEE 69 bus radial distribution system. The obtained results imply that the presented energy management framework provides maximum profits for the vehicle owner, and meanwhile it fulfills preferences of the user in each zone simultaneously.     

A multiobjective integrated multiproject scheduling and multiskilled workforce assignment model considering learning effect under uncertainty

Today, organizations try to decline academically expenses using humans and resources in addition to rising managers and operators' satisfaction. Meantime, a very important step in the process of decision is the assignment of human resources, particularly in connection with research and development (R&D) projects in which the system is highly dependent on the capabilities of human resources. In this study, we tried all the assumptions that come true in the real world, considered a model for applied R&D projects to reduce costs and increase the efficiency of projects. Therefore, an integrated multiproject scheduling and multiskill human resource assignment model under uncertainty has developed for R&D projects. Furthermore, it is assumed that the activity processing time is related to human resources assignment that means the learning effect is considered. To demonstrate the proposed model efficiency, the various dimensions instance problem was solved accurately and efficiently in GAMS software, and the results have been reported. In addition, the proposed model is validated through the input parameter sensitivity analysis. The results indicate a suitable performance of the proposed fuzzy mathematical programming model is due to the complexity of the problem.     

热处理过程智能信息化生产调度技术

热处理是制造业中至关重要的一环,针对热处理企业的多品种变批量生产模式,合理协调车间生产资源、保障产品的质量,提高生产效率、降低生产能耗、提升企业智能化管理水平,是热处理企业亟待解决的难题。基于热处理的实际生产过程,利用物联网信息采集、异构信息数据融合、生产流程信息化管控等技术,结合生产调度遗传算法,构建了一套热处理智能化管控系统平台。通过系统平台,实现了热处理过程的生产实时调度,提高生产效率,提升热处理产品质量一致性的管控能力。     

分时电价下多目标绿色可重入混合流水车间调度

针对多目标绿色可重入混合流水车间调度问题（RHFSP）的特点，在机器分配和工序排序的基础上引入分时电价机制，构建了以最小化最大完工时间、总能耗成本和碳排放为目标的绿色调度优化模型，提出了一种改进的多目标文化基因算法（MOMA）来求解该问题，通过数值实验验证了所设计的MOMA算法的可行性。实验结果表明MOMA算法在非劣解的收敛性、多样性和支配性指标方面都显著优于多目标蚁狮优化算法(MOALO)、多目标粒子群优化算法（MOPSO）和带精英策略的非支配排序遗传算法（NSGA-Ⅱ），四种算法的分布性指标无显著差异。所提出的模型能够使企业有效避开高电价时段作业，合理转移用电负荷，达到降低总用电成本和碳排放的目的。     

Mixed coordination mechanisms for scheduling games on hierarchical machines

In this paper, we study scheduling games under mixed coordination mechanisms on hierarchical machines. The two scheduling policies involved are ‐ and ‐, where ‐ (resp., ‐) policy sequences jobs in nondecreasing order of their hierarchies, and jobs of the same hierarchy in nonincreasing (resp., nondecreasing) order of their processing times. We first show the existence of a Nash equilibrium. Then we present the price of anarchy and the price of stability for the games with social costs of minimizing the makespan and maximizing the minimum machine load. All the bounds given in this paper are tight.     

低碳环境下钢铁企业燃气系统动态平衡调度仿真

在“节能减排、降本增效”的大背景下,从碳素流视角切入,在煤气介质预测基础上,建立了0-1混合线性规划模型,以能源成本和碳交易成本总和最小化为目标,研究钢铁行业副产煤气生产调度问题。采用基于遗传算法的动态迭代调度算法,通过混合交叉算子和平均变异算子提高遗传算法局部搜索能力,并以国内某大型钢铁企业为仿真对象进行系统仿真,结果表明:混合煤气柜可以根据煤气产耗特点调整存储量,并保证存储量在合理范围内;调节用户会起到一定的缓冲作用,可以根据某种副产煤气的产耗特点响应增加或减少该种煤气的消耗量;当某种副产煤气严重不足时,固定消耗用户会根据优先级响应停调,以满足生产需求。可见副产煤气的优化调度应用可以为中国钢铁行业的低碳发展和节能减排提供技术支撑。     

A robust mixed integer linear programming framework for underground cut-and-fill mining production scheduling

ABSTRACT

A review of general optimization studies that have been proposed for underground mining shows that previous works lack flexibility, operability and practicality in relation to cut-and-fill mining production scheduling. This paper presents a robust mixed integer linear programming (MILP) formulation for underground cut-and-fill mining. The objective function of the model is to maximize the net present value (NPV) of the operation while meeting all mining and processing operational and technical constraints. The MILP model features stope development and extraction sequencing constraints, mining and processing tonnage fluctuation constraints, and extraction duration and active levels control constraints. These features make the model more practical and expandable. The MILP model is verified and validated with two case studies from an existing mine and the results are compared with the actual mining strategy. The comparison shows a 9% to 17% improved NPV in both case studies resulting from mining higher grades and processing less tonnes thereby generating a better cash flow.     

变压器经济运行区间划分的节电降耗

开展变压器经济运行是节电降耗的重要途径,而变压器经济运行区间是开展变压器经济运行的重要概念。本文阐释变压器经济运行区间的概念和求解原理;并以实际某变电站为例,说明经济运行区间的求解方法及其应用;同时给出取得的节电降耗效果和经济效益。     

白云鄂博矿区五大矿体的一体化利用及整合

介绍了国家规划矿区--白云鄂博矿区的资源概况,以及当前开发利用的现状和出现的问题,简述了五大矿体利用的进展,从充分利用和保护资源入手,有针对性地提出了下一步整合中应注意的问题、解决问题的方法和思路.