"In recent years, China's steel industry has formed a three-dimensional technology integration of information technology integration, process technology integration, and full life cycle technology integration. These technology integration are the basis of information technology. Therefore, the steel industry is the closest industry to intellectualization, with the best conditions to implement intellectualization, the demand for intellectualization is also the most urgent. On September 27, Wangguodong, a member of the Chinese Academy of Engineering and a professor at Northeastern University, co-sponsored by the China Metallurgical Newspaper and the Tangshan City Bureau of Industry and Information Administration, and 2019(5th) co-organized by Jiujiang Line Materials Co., Ltd. of Qian 'an City.) China Steel Industry Internet + Promotion Conference.
Intelligent Manufacturing Platform Architecture for Iron and Steel Industry
Wangguodong introduced the platform architecture of intelligent manufacturing in the steel industry.
In his view, intelligence has the following characteristics: First, it has the ability to perceive the outside world and obtain external information. It is a prerequisite and necessary condition for the generation of intelligent activities. The second is the ability to remember and think, that is, to store perceived external information and knowledge generated by thinking, and to use existing knowledge to analyze, calculate, compare, judge, associate, and make decisions on information. The third is the ability to learn and adapt, that is, through the interaction with the environment, continue to learn to accumulate knowledge, so that they can adapt to environmental changes. The fourth is the ability to make behavioral decisions, that is, to respond to external stimuli, form decisions and convey corresponding information.
The core of intelligent control is the Information Physics System(CPS). The overall architecture of the system is: perceptual(physical) layer, network layer, computing layer. Through the integration of advanced information technology and automatic control technologies such as perception, calculation, communication, and control, it has built a complex system in which elements such as people, machines, objects, environment, and information in physical space and information space map each other, interact in a timely manner, and collaborate efficiently. Realize the system resource configuration and running on demand response, rapid iteration, dynamic optimization. At present, the embedded industrial control system of metallurgical enterprises is composed of physical layer, network layer, computational control layer and application layer.
Wangguodong pointed out that the principle of building smart manufacturing platforms is to make full use of existing systems and existing data of enterprises, adopt advanced smart manufacturing technologies, enhance network connectivity and data collection and processing, and develop digital twins of CPS systems. Digital perception system, To upgrade the enterprise's existing system into an intelligent system with self-learning, self-adaptation and self-organization, to achieve full-process intelligent manufacturing, and to promote the high-quality development of the enterprise.
The second is to implement true intelligence. In the process of implementing the whole process and integrated intelligent manufacturing, the enterprise shall meet the requirements of the system functions. We will gradually and steadily upgrade existing enterprise resource planning(ERP), manufacturing execution systems(MES), process control, and basic automation systems. It has the ability of perception, memory and thinking, learning ability and adaptive ability, and behavioral decision-making ability, and becomes a real intelligent system.
"The steel industry is a process industry. Therefore, the iron and steel industry information system must be full process and integrated, and it can achieve the comprehensive analysis, control, and management of material flow, information flow, and energy flow. "Wangguodong stressed.
He further pointed out that with the enrichment of technical means and the improvement of computational capabilities, the solution to problems will change from the traditional way of dividing complex problems into relatively simple different levels and solving problems layer by layer to "cross-layer and cross-domain". The flat development method.
The flat double-layer architecture is "cloud intelligence layer" and "local decision-making level." The cloud intelligence layer has low real-time MES, ERP, and BI(business intelligence) functions, and achieves comprehensive coordination of multiple goals such as quality, equipment capabilities, costs, resources, and human resources. The "local decision-making level" has the functions of high real-time PCS(process control system), BA(building equipment automation system) and sensors, supplemented by new technologies such as artificial intelligence and large data, strengthening computing capabilities and real-time. The online real-time dynamic adjustment between the two levels has flexibly and harmoniously achieved the improvement of quality and efficiency.
In order to achieve flat development, enterprises need edge calculation. The application of edge computing can achieve intelligent control of near-field devices, devices and systems, solve the uncertainty caused by network latency, reduce the cloud burden, and take into account the coordination of cloud computing and terminals.
Digital perception system is the core of steel intelligent manufacturing platform
Iron and steel enterprises are facing some problems in the process of intelligent development. Wangguodong pointed out that the steel industry is a process industry with multiple processes and long processes. The steel production process is a "black box" of the entire process, chemical reactions and physical changes within the smelting reactor, continuous casting and rolling workpieces, and related important parameters. Can not see, can not feel, can not measure, real time, internal information perception is extremely lacking; At present, most of the models used in the system are theoretical models or empirical models. There are large uncertainties, poor forecasting accuracy, and it is difficult to achieve high-precision dynamic coordination control. There are rough or even blind problems in control decision-making. Due to the complex changes in production conditions caused by the changes in process conditions and equipment operating conditions, the relationship between process input conditions, state variables, and control targets is complex.
"Digital perception provides us with new ideas and feasible methods for using modern information technologies such as big data, the Internet, cloud computing, and AI(artificial intelligence) to overcome obstacles that could not be overcome in the past. It provides an ideal solution for the precise perception and control of metallurgy and other industries. Wangguodong further stated that "under the control of information flow, material flow reacts chemically or physically with energy flow. Therefore, mathematical models based on physical-chemical reactions can be solved with sufficient precision to obtain material flow, energy flow, and information flow. Unified solution. "
Due to the characteristics of uncertainty, strong non-linear, and extremely complex, the mathematical model of the steel industry must adopt the combination of the mathematical model and the AI / large data, use the mathematical model as the basis of the main line, and use artificial intelligence and measured large data to optimize the mathematical model. High-precision digital perception, Cracking the steel industry "black box" system.
The basic components of the CPS system include sensors, actuators, and decision control units. The basic component, combined with the feedback loop control mechanism, constitutes the basic functional logic unit of the CPS and performs the most basic detection and control function digital perception system: a giant digital sensing system based on large data, artificial intelligence, the Internet of Things, and the physical-chemical model. The digital twins of all physical quantities in the real physical world are realized.
Among them, the decision control unit can make intelligent decisions according to the digital perception system, perceived information and user input rules, and issue the process control parameter adjustment instruction to ensure that the control target value is obtained under the established conditions.
Wangguodong pointed out that the number of physical systems is the core. Industrial 4.0's digital perception system can fully perceive and control the overall flow laws of information flow, material flow, and energy flow. The whole process integrated CPS system can solve the problems such as breakpoints and off-line automatic data acquisition, and realize the whole process integration and coordination control.
The main functions of intelligent manufacturing system are: one is information depth perception, which is to realize process quality accurate prediction and online state perception; The second is data processing storage, that is, the process quality data acquisition and processing; The third is network connectivity, including the 5G Internet of Things, the Internet connecting everything and people; Fourth, the equipment diagnosis and maintenance, namely the realization of equipment health state diagnosis and intelligent maintenance; The fifth is the optimization of quality control, that is, to achieve the cause of quality anomaly tracing and production process optimization; The sixth is precision coordination control, that is, realizing quality precision control and intelligent optimization of multi-process coordination; The seventh is intelligent optimization decision, that is, intelligent optimization decision that realizes the whole process of horizontal production, supply and marketing, and the cooperation of longitudinal levels.
"To build an information system for the iron and steel industry, we should, in accordance with the process of steel production, establish an intelligent sensor, decision-making and control agent in the main production link to achieve intelligent control of each link. These agents include: raw material processing agents, blast furnace smelting agents, steelmaking-continuous casting agents, hot rolling agents, cold rolling agents, and heat treatment agents. The steel enterprise should make up the multi-agent which covers the whole process and realize the cohesion, coordination, balance and optimization between the various agents. "Wangguodong suggested.
