基于高效拉氏松弛迭代算法的炼钢-连铸主辅设备调度方法的研究

The process of steelmaking-continuous casting is the bottleneck of steelmaking production, the dynamic equilibrium of logistics, resource and time among the main equipment (including multiple converters, multiple refining position and multiple casting machines), the rational allocation of production resources and the collaborative operation of main equipment for the accessorial equipment (including multiple shop travelers and multiple trolleys) for this process are of important significance for improving the efficiency of steelmaking production and reducing the material and energy consumption. First, this application studies the complex production process scheduling models of the main and accessorial equipment for the steelmaking-continuous casting process based on the different mathematical programming modeling methods, analyses the influence of the different relaxation strategies under the different mathematical models for the solution quality and solution speed of lagrange relaxation method, proposes the static schedule model, the reschedule model and the related lagrange relaxation strategies of the main and accessorial equipment for the steelmaking-continuous casting process with separable structure. Then, this application studies the optimal methods for the static schedule and the reschedule of the main and accessorial equipment for the steelmaking-continuous casting process based on effective lagrange relaxation iterative algorithm, improves the optimal efficiency of the static schedule and the response speed for the reschedule of the main and accessorial equipment for the steelmaking-continuous casting process. Finally, the industrial application is verified based on the background of the actual steelmaking and continuous casting schedule in domestic steelmaking plant. The research will improve the schedule method for the actual steelmaking-continuous casting process, promote the development of the schedule theory with the separable structure characteristics and lay a foundation in the industry application for the schedule theory.

炼钢-连铸过程是钢铁生产的瓶颈环节，该过程主体设备（多台转炉、多类多个精炼位及多台连铸机）间物流、资源与时间的动态平衡，辅助设备（多台天车及多台台车）合理配置生产资源、协同主体设备作业，对提高钢铁生产效率、降低物耗和能耗具有重要意义。本项申请首先研究基于不同数学规划建模方法的炼钢-连铸主辅设备生产过程的调度模型，分析基于不同数学模型的不同松弛策略对拉氏松弛方法求解质量和求解速度的影响，提出适用于可分离结构特征的炼钢-连铸主辅设备静态调度和重调度模型及其拉氏松弛策略；其次研究基于高效拉氏松弛迭代算法的炼钢-连铸主辅设备静态调度和重调度方法，用以提高炼钢-连铸过程主辅设备静态调度编制的效率及重调度编制的响应速度；最后以国内某大型钢铁企业的炼钢-连铸调度实例为背景进行工业应用验证。研究成果将完善实际复杂炼钢-连铸过程的调度方法，推动具有可分离结构特征调度理论的发展，为调度理论的工业应用奠定基础。

本项目依托国内某大型钢铁企业炼钢-连铸生产过程为研究背景，将高效拉氏松弛迭代算法应用于炼钢-连铸过程主辅设备优化调度问题中，在提高炼钢-连铸主辅设备静态调度的求解质量及炼钢-连铸主辅设备重调度的响应速度的基础上，探索和研究了复杂大规模混合流水车间调度问题的建模方法及拉格朗日对偶问题的优化方法。首先，针对炼钢-连铸主辅设备调度过程不同数学规划建模方法及其不同松弛策略下的多种子问题分解方法对拉氏松弛方法求解质量和求解速度的影响分析，建立了适用于复杂大规模炼钢-连铸主辅设备静态调度和重调度数学模型并选择了相应的松弛策略，为获取高效的调度方案奠定了基础；其次，针对拉氏框架下次梯度算法的迭代优化过程在可行域出现的震荡现象以及次梯度算法收敛条件过于严格而难以实现的问题，设计了方向可控的拉格朗日松弛迭代优化算法，用以提高炼钢-连铸主辅设备静态调度问题的求解质量；再次，针对传统拉氏框架下次梯度算法因精确求解而导致效率低的问题，提出了基于拉氏松弛框架的可控误差近似求解优化算法，用以提高炼钢-连铸重调度问题的响应速度；最后，将本项目所提高效拉氏松弛迭代优化算法应用到实际炼钢-连铸主辅设备静态调度和重调度过程中，验证了所提算法的有效性。研究成果完善了炼钢-连铸过程的调度方法，推动了具有可分离结构特征调度理论的发展，为调度理论的工业应用提供了理论依据。基于项目的研究内容，项目主持人及其团队申请发明专利3项；发表SCI检索国际期刊论文3篇，EI检索国际期刊/会议论文13篇，学术专著1部；培养硕士研究生18名；参加本领域相关国际会议12次；均已超额完成预期研究成果。