"AI+steel" forges New Quality Productive Forces

"The steel industry is an important basic industry. It has a huge industrial scale, deep application accumulation and urgent need for high-quality development. It is an ideal practice field for the deep empowerment of the new generation of artificial intelligence." Jiang Wei, vice president and secretary-general of the China Iron and Steel Industry Association, said recently.

At present, the China Iron and Steel Association has listed digital transformation as one of the three major projects in the industry and launched a three-year action plan for the digital transformation of the steel industry. Under the guidance of a series of national "artificial intelligence +" policies, a large number of steel enterprises have made substantial progress in promoting digital transformation, and relying on digitalization to support high-end and green development.

In full swing

In the 13th list of global "lighthouse factories" announced at the beginning of this year, Shougang Co., Ltd. Cold Rolling Company was successfully selected. With the construction of cloud-edge collaborative technology architecture, the company has become the third "beacon factory" in the country's steel industry and the seventh "beacon factory" in the global steel industry.

In order to cope with the challenges brought by stricter quality requirements and increasingly diversified product specifications in the field of high-end automobile manufacturing, Shougang Cold Rolling Company has deployed 67 Industry 4.0 digital application cases, of which 61% use artificial intelligence, improving end-to-end process control accuracy, eliminating bottlenecks that restrict key quality and capacity improvement, and solving the quality problems faced by customers.

"The production line efficiency increased by 21.2 percentage points, the product defect rate decreased by 35 percentage points, the sales of high-end products increased by 36 percentage points, and the customer complaint rate decreased by 55 percentage points." A relevant person in charge of Shougang Co., Ltd. Cold Rolling Company spoke with data.

This is just a microcosm of the upgrade of intelligent manufacturing in the steel industry. According to the coordinated research of China Steel, 95.1% of steel enterprises have included digital transformation strategies into the overall development process of enterprises, and have continued to invest funds to implement digital transformation and upgrading projects. The application density of robots (including unmanned equipment) reaches 65 sets per 10,000 people; 82.9% of enterprises have built intelligent centralized control centers, mainly concentrated in front of iron and steel rolling, and 63.4% of enterprises use three-dimensional visual simulation systems to build digital factories.

Redefine steel with artificial intelligence. Baosteel Co., Ltd., a subsidiary of China Baowu, has formulated a new round of digital transformation strategy marked by "AI+", launched the "full field, full system, and full scenario" action plan, and create more than 1,000 AI-enabled application scenarios in the next three years.

"We must adhere to innovation-driven development, strengthen technological empowerment, and accelerate the intelligent and green transformation of the steel industry." Liu Jian, chairman of Hesteel Group, said that Hesteel has incorporated intelligent manufacturing into the enterprise development strategy, actively promote the deep integration of artificial intelligence technology and steel production, vigorously implement the "intelligent improvement action", and accelerate the transformation and upgrading of enterprises with artificial intelligence.

According to reports, Hegang independently developed the "Visabo" industrial Internet platform, integrating innovative capabilities such as intelligent equipment, industrial simulation, model control, etc., forming a basic industrial Internet base that runs through the entire production process and coordinates software and hardware. Hegang has also promoted the integration of technologies such as artificial intelligence and digital twins with the entire chain of manufacturing processes, and built a large number of "smart factories".

Focus on cracking the "black box"

At present, the steel industry is in a critical stage of moving from digital to intelligence, and the new generation of artificial intelligence represented by big models is expected to become a new engine driving high-quality development. At the "AI+Steel" technology press conference held recently, experts and scholars discussed how to promote the innovation and application of "AI+Steel" technology.

Tian Guobing, chairman of Shanghai Baoxin Software Co., Ltd., said that in order to promote high-quality development, steel companies are shifting from scale to efficiency and benefit orientation, but traditional automation, information technology and systems have obvious shortcomings in data, models, computing power, etc., which restricts the ultimate improvement of enterprise efficiency and benefits, and urgently needs to break through digital and intelligent technologies.

"To achieve a new breakthrough in 'artificial intelligence + steel' and to enhance the competitiveness of the industry in all aspects." Wang Guodong, an academician of the Chinese Academy of Engineering, emphasized that as a large and complex process industry, the steel industry has long had the problem of "black box" in all processes. It is necessary to integrate the advantages of experts in my country's steel materials field and AI field to promote the deep integration of the new generation of AI technology with the real economy of the steel industry.

It is understood that the hardware, software and production environment of my country's steel industry have reached the world's top level in the industrial era, but it has been trapped in the "black box" of the entire production process for a long time. Since steel production involves complex physical and chemical processes, there are a large number of incomplete information, uncertainty and dynamic environmental problems, which have a significant impact on product quality, production efficiency, production cost and environmental effects.

In response to this industry pain point, Northeastern University took the lead in proposing the "human-machine hybrid intelligence" model, which is specially used to crack the "black box" of steel production, and can realize the full-process online high-fidelity prediction. Related technologies have been implemented in many steel companies. For example, in the intelligent smelting of blast furnaces, the "general model + personalized data" architecture is innovatively adopted to effectively improve energy efficiency and green production levels; in the hot rolling stage, a digital twin model based on human-machine hybrid intelligence is developed, integrating physical mechanisms and multimodal data, and successfully achieving online high-precision prediction and closed-loop optimization of organizational performance.

Wang Guodong further explained that in-depth integration of human intelligence with artificial intelligence, mining and analysis of big data in the production process, and promoting "digital brain replacement, model replacement, and industrial software development" through human-machine collaboration, an independent unmanned intelligent control system for the main unit of the steel production process can be built. This is equivalent to converting the overall "black box" of production into a comprehensive sensor, and using digital twins to achieve accurate predictions, so that the production process will be autonomous and unmanned and intelligently controlled.

Broad development prospects

"Under the background of rapid development of technology, artificial intelligence has brought unprecedented opportunities to the steel industry." Zhang Longqiang, director of the Institute of Metallurgical Industry Information Standards, said that the application scenarios of AI in the steel industry have gradually been implemented and the prospects are very broad. In terms of production process optimization, intelligent algorithms can accurately predict equipment failures, significantly reduce downtime, and significantly improve production efficiency; in the field of quality control, AI image recognition technology can conduct high-precision detection of steel surface defects to ensure product quality meets standards; in the supply chain management link, AI can conduct forecasts based on multiple factors such as market demand and raw material price fluctuations, optimize procurement and production plans, reduce enterprise operating costs, and enhance core competitiveness.

Looking ahead, the development space of "AI+steel" will be further expanded. On the one hand, with the continuous development of AI technology, especially the in-depth application of large-scale model technology, it will further penetrate the "black box" of the steel production process, penetrate data silos, crack the resource scheduling dilemma, and realize intelligent and refined management of the entire steel production process. By accumulating data assets and enhancing market competitiveness, we can develop "real skills" faster, more accurate and more efficient for steel enterprises. On the other hand, "AI+steel" will help the industry to develop green and sustainable, promote energy conservation, emission reduction and environmental protection. At the same time, the construction of a cross-enterprise and cross-industry AI ecosystem will become a trend, promoting in-depth cooperation and collaborative innovation between steel enterprises and upstream and downstream enterprises, scientific research institutions, and technical service providers, and jointly promoting the industry to transform to high-end, intelligent and green.

Jiang Wei said that the China Iron and Steel Association will continue to take "AI+intelligent technology application" as the core, and comprehensively turn to the reshaping of productivity driven by a new generation of artificial intelligence technology, and promote industry transformation through policy guidance, standard construction, and benchmark promotion. Focus on intelligent collaboration of the industrial chain, low-carbon intelligent integration and low-cost technology promotion.

The core value of artificial intelligence is not to replace human resources, but to build a new productivity paradigm of "human-machine collaboration". "Human-machine collaboration can give full play to the efficient computing power of AI and human creativity and judgment, ensure the stability of steelmaking operations to the greatest extent, and promote continuous optimization and innovation of the system." said Liu Shuchao, professor at Northeast University.

Liu Shuchao grew up from the front line of steelmaking and is now deeply engaged in research on steel intelligent technology in colleges and universities. He led the team to develop a full-process modular intelligent steelmaking system for converter based on the "AI+Manual Experience" model. All or part of its modules have been applied in steel companies such as Ansteel, Bengang, Nangang, Chenggang, Meigang, and Zhougang.

This experience from front-line practice to scientific research and innovation also gave Liu Shuchao a deeper understanding of smart steelmaking: "The development of smart steelmaking should not aim at 'reducing people'. The core is to make steelmaking operations simpler, more stable process, more professional technology, and more controllable costs. Only when leaders pay attention to it, grassroots active, employees participate, top and bottom are united, and beliefs are consistent, and then scientific policies and continuous improvement can smart steelmaking truly go deeper and more practical."

This picture and text on the website is for the purpose of conveying more information, and is designed to collect and provide it to meet the information needs of the majority of users. It is not advertising service information. If there is any infringement, please contact us to delete it.