炼钢产品质量和合同生产周期解析

发布时间：2018-05-06 12:26

本文选题：转炉炼钢 + 数据解析　；参考：《东北大学》2014年硕士论文

【摘要】：如何提高钢铁产品质量和有效控制产品生产周期是钢铁企业在激烈的市场竞争中面临的两大难题。前者可通过操作优化、过程控制等手段完善生产工艺；后者则需要对产品的生产周期进行有效地预测判断。两者相辅相成,分别从“质”和“量”的角度提升企业效益,提高竞争力。炼钢是钢铁生产中的一道重要工序。在转炉炼钢过程中,碳与温度的走势直接反应了转炉内的冶炼状态,决定着氧气与其他辅料的加入策略,影响着出钢质量。因此,预测炼钢过程中钢水温度和碳含量对提高炼钢产品质量至关重要。另一方面,钢铁企业的生产计划都是围绕产品合同和生产能力展开,合同生产周期的管理是钢铁生产管理中非常重要的一项任务。预测合同完成时间有助于从整体上把握合同进程,指导生产计划的排制和调整。本文利用数据解析方法,通过建立企业实际生产数据的解析模型,分别针对炼钢过程中钢水温度和碳含量以及产品合同生产周期的预测问题进行了解析研究。主要内容包括以下几个部分：(1)以钢铁企业转炉炼钢为背景,研究钢水温度与碳含量的预测问题。针对该问题,设计了基于改进粒子群算法的最小二乘支持向量机方法,并分别建立钢水温度和碳含量的预测模型。建模过程中,采用多阶段建模方法,实现对整个过程的动态预测。对于实际数据采集不细致的问题,采用插值算法对数据进行预处理。最后通过实际数据进行多组实验,验证了该方法的有效性及多阶段动态建模的准确性。(2)以钢铁企业生产周期管理为背景,研究合同完成时间预测问题。钢厂合同类型多,生产工序多且杂,产生了大量的合同数据信息。如何根据合同类型和特征,对合同完成时间进行快速准确的预测是本文研究重点。针对该问题,建立了基于粒子群算法的最小二乘支持向量机预测模型,利用历史数据进行解析模型训练,并用实际生产数据进行实验分析,验证了模型的有效性。(3)以某钢厂的实际合同管理为背景,设计开发了钢铁合同管理子系统,对合同完成时间进行预测并对当前各类合同完成情况进行统计。系统不仅实现了对生产过程中合同数据的监控和分析功能,还与KPI绩效管理系统兼容,将合同管理指标作为员工个人的绩效评价指标,实现绩效考评的功能。
[Abstract]:How to improve the quality of iron and steel products and how to effectively control the production cycle are two major problems faced by iron and steel enterprises in the fierce market competition. The former can improve the production process by means of operation optimization and process control, while the latter needs to predict and judge the production cycle of the product effectively. The two supplement each other, from the angle of "quality" and "quantity" to enhance the efficiency of enterprises and enhance their competitiveness. Steelmaking is an important process in steel production. In the process of converter steelmaking, the trend of carbon and temperature directly reflects the smelting state in converter, determines the addition strategy of oxygen and other auxiliary materials, and affects the quality of steel production. Therefore, it is very important to predict the temperature and carbon content of molten steel to improve the quality of steelmaking products. On the other hand, the production planning of iron and steel enterprises revolves around product contract and production capacity, and the management of contract production cycle is a very important task in steel production management. Forecasting the completion time of the contract will help to master the process of the contract as a whole and guide the scheduling and adjustment of the production plan. Based on the method of data analysis, the prediction of temperature and carbon content of molten steel and the production cycle of product contract in the process of steelmaking are studied by establishing the analytical model of actual production data in the enterprise. The main contents are as follows: (1) based on the background of converter steelmaking in iron and steel enterprises, the prediction of molten steel temperature and carbon content is studied. To solve this problem, the least squares support vector machine (LS-SVM) method based on improved particle swarm optimization (PSO) algorithm is designed, and the prediction models of molten steel temperature and carbon content are established respectively. In the process of modeling, the multi-stage modeling method is used to realize the dynamic prediction of the whole process. The interpolation algorithm is used to preprocess the actual data. Finally, the validity of the method and the accuracy of multi-stage dynamic modeling are verified by many experiments based on the actual data. Based on the production cycle management of iron and steel enterprises, the prediction of contract completion time is studied. There are many types of contracts and many production processes in steel mills, which produce a large amount of contract data information. How to predict the contract completion time quickly and accurately according to contract types and characteristics is the focus of this paper. In order to solve this problem, a prediction model of least squares support vector machine based on particle swarm optimization algorithm is established. The analytical model is trained by using historical data, and the experimental analysis is carried out with actual production data. The validity of the model is verified. (3) based on the actual contract management in a steel plant, the steel contract management subsystem is designed and developed, the contract completion time is forecasted and the current contract completion situation is counted. The system not only realizes the function of monitoring and analyzing the contract data in the production process, but also compatible with the KPI performance management system. It regards the contract management index as the performance evaluation index of the individual employees and realizes the function of performance evaluation.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F426.31;F273.2

【相似文献】