材料科学技术转型发展与钢铁创新基础设施的建设

摘要：在第4次工业革命浪潮的推动下,钢铁科学与技术正在经历数字化、智能化转型。钢铁行业全流程各工序均为“黑箱”，为多场、多相、多变的巨系统,具有复杂相关关系和遗传效应等。这些不确定性带来了巨大的挑战。挑战和机遇并存。这些不确定性提供了智能化和数字化技术的应用场景资源;钢铁行业极为丰富的大数据提供了挖掘其中蕴含客观规律的数据资源;现代的数据科学、智能技术为解决不确定性问题提供了强大的手段。以数据为中心，以工业互联网为载体，以实验工具、数字数据、计算工具为支撑,建设钢铁企业材料创新基础设施,将可以大幅度提高研发效率,降低研发成本，有力地支撑钢铁材料科学与技术的转型发展。实验工具平台除了传统的实验室仪器装备和中试装备之外，实际生产线被作为主要的实验工具。这些实验工具提供丰富、精准、写实的历史数据和现实生产数据,特别是生产线装备提供实际生产大数据,蕴含着生产过程中的全部规律,是极宝贵的数据资源。利用机器学习、深度学习等现代数据挖掘技术为计算工具,对这些数据资源进行处理、分析、计算,将数据转换为高保真度模型,可以得到具有“原位分析能力”的数字孪生。在工业互联网的总体架构下，以数字孪生为核心，组成信息物理系统,构建起基于数据自动流动的状态感知、实时分析、科学决策、精准执行的闭环赋能体系,解决生产制造、应用服务过程中的复杂性和不确定性问题,提高资源配置效率,实现资源优化,对材料行业转型发展提供关键技术支撑。虚实映射、实时交互、精准控制的信息物理系统与材料创新基础设施合二为一,以材料创新基础设施为基盘,形成具有“原位分析能力”的数字孪生,建设钢铁生产全流程、一体化的信息物理系统,必将推进钢铁行业智能制造蓬勃开展和数字化、智能化转型。

Transformation and development of materials science and technology and construction of iron and steel innovation infrastructure

Driven by the fourth wave of industrial revolution, iron and steel science and technology are undergoing digital and intelligent transformation. The whole process of iron and steel industry is "black box", multi field, multi phase, variable giant system, complex correlation and genetic effect, etc. These huge uncertainties bring great challenges. Challenges and opportunities coexist. These uncertainties provide the application scenario resources of intelligent and digital technology; the extremely rich big data of iron and steel industry provide the data resources of mining the objective laws contained in the data; modern data science and intelligent technology provide a strong means to solve the uncertainty problems. Taking data as the center, industrial Internet as the carrier, as well as experimental tools, data and calculation tools as the support platform, building the material innovation infrastructure of iron and steel enterprises will significantly improve R & D efficiency, reduce R & D cost, and strongly support the transformation and development of iron and steel science and technology. In addition to the traditional laboratory equipment and pilot equipment, the actual production line is regarded as a main experimental tool of the experimental tool platform. These experimental tools provide rich, accurate and realistic historical data and real production data, especially the production line equipment provides the actual production big data, which contains all the rules in the production process, and is a very valuable data resource. Modern data mining technologies, such as machine learning and deep learning, are used as computing tools to process, analyze and calculate these data, and transform the data into high fidelity models, so that digital twins with "in situ analysis ability" can be obtained. Under the overall framework of industrial Internet, digital twin as the core combined with physical system constitutes an Cyber Physical System, which constructs a closed loop enabling system based on automatic data flow, such as data perception, real time analysis, scientific decision making and accurate execution, solves the complexity and uncertainty problems in the process of manufacturing and application services, improves the efficiency of resource allocation, realizes resource optimization and improves the quality of materials to provide key technical support for the transformation and development of the industry. The Cyber Physical System with virtual real mapping, real time interaction and precise control is integrated with the material innovation infrastructure. Based on the material innovation infrastructure, a digital twin with "in situ analysis ability" is formed. The construction of the integrated Cyber Physical System along the whole steel production chains will promote the vigorous development of smart manufacturing and digital and intelligent transformation of the steel industry.