Industrial Control Practice Forum

 

1.

 

Yuanming Zhu

East China University of Science and Technology

(朱远明,华东理工大学)

 

Yuanming ZHU received his B.S. and Ph. D. degree in BJTU and work as a vice professor in ECUST. The candidate of "Sailing Plan Project" in Shanghai. He has undertaken more than 10 national scientific research projects. His research interest includes data-driven control, robust control and intelligent control for industry process. Recent years, his research direction is aimed at the national major strategic demands, committed to the research of key theories and technologies for cement green manufacturing. He has published more than 20 academic papers, including in TNNLS, TII, IJRNC, IECR, etc.; applied for more than 10 national invention patents and obtained 6 software copyrights in the field of optimization and control for cement manufacturing.

 

Title: Data-driven control technology in cement industry

 

The dry process for manufacturing of cement consists of three stages: raw material preparation, clinker calcination, and cement grinding, which involves the mixing of gas and solid phases and high-temperature chemical reactions. However, the sampling delay for composition analysis, the lack of online measurement for critical process variables (such as fineness of product, the decomposition rate of raw meal, and the concentration of the free calcium oxide in clinker), and the variety of fuel sources and quality, have posed great challenges to the quality control and energy and emission reduction. This talk will discuss some key data-driven technologies of intelligent control for cement industry. 1. An embedding learning algorithm based soft-sensor; 2. Parameterized controller based data-driven control; 3. A deep reinforcement learning approach. Meanwhile, some application scenarios will be included.

 

 

 

2.

Minglei Yang

East China University of Science and Technology

(杨明磊,华东理工大学)

 

Yang Minglei was born in 1985. He received his B.S. and Ph. D. degree in Chemical Engineering from East China University of Science and Technology. He is now the vice director of automation research institute. For the past 10 years, he has been devoted to the steady / dynamic state of complex petrochemical processes, multi-scale mechanism modeling, process optimization and decision making. As the project leader, he undertook 2 National Natural Science Foundation Projects and 1 basic scientific research project of Central University project. As the technical director, he undertook over 10 projects from industry and MIIT, such as the development of large-scale petrochemical process mechanism modeling, plan optimization, profit maximization for reforming process, etc. The developed technics have been successfully applied in Sinopec Jiujiang Branch, Zhenhai Branch, Shanghai Branch and Yangzi Branch. He has published over 30 papers in Chemical Engineering journals, and authorized over 20 Chinese patents. In 2019, he won the First prize of Shanghai Science and Technology Progress Award and Technological Invention.

 

Title: Value chain evaluation and optimization for petrochemical process

 

Abstract: Smart manufacturing is very hot and popular in petrochemical industry in China due to the great improvements in economic profit and operation efficiency. This report is to introduce three important technologies applied in Smart Factory: (1) rapid characterization of refinery oil. Using near infrared spectroscopy technology, online evaluation of crude oil, distillate oil, product oil and other oil products can provide accurate and fast information for business optimization; (2) digital twin model of the whole plant. The plant-wide model is built by combing the mechanism and operation data, so as to improve the accuracy in predicting the yield and properties of products; (3) plant-wide optimization. By integrating the mechanism model in planning, the crude oil purchase, resource distribution, and unit operation mode are optimized to maximize the production value of the whole plant. Finally, several application cases in smart manufacturing for petrochemical industry are shared.

 

 

3.

Hao Chen

Chinese Academy of Sciences

(陈豪,中国科学院海西研究院)

 

Hao Chen, a researcher of Haixi Institute, Chinese Academy of Sciences & PI. He received his Bachelor degree in the major of automation from National University of Defense Technology in 2006, Master degree in the major of control system from The University of Sheffield, UK in 2009, Doctor’s degree in the major of Cybernetic from University of Reading, UK in 2013. From January 2014 to July 2015, he engaged in industrial data analysis and soft-sensor research in the University of Alberta, Canada / Syncrude oil company. Now he is the director of Fujian Provincial key laboratory for Intelligent Identification and Control of Complex Dynamic System, the committee member of data driven control, learning and optimization Committee of Chinese Association of Automation, the vice director of Fujian Provincial Industrial Internet Intelligent Sensing and Decision Engineering Research Center. He presided over nearly 30 national / Chinese Academy of Sciences / local scientific research projects, applied for / authorized 36 national invention patents, and published nearly 50 academic papers. He has won the Youth Promotion Association of Chinese Academy of Sciences, Fujian Natural Science Foundation for Distinguished Young Scholars and Fujian May 4th Youth medal.  He committed to the research road of combining theory with application. The production optimization series technology and industrial software that developed by the team have been successfully applied in nearly ten leading enterprises in the industry, such as Sunner Group, Seven Group, Shuhua Sports, etc.

 

 

Title: Exploration on the Application of Industrial Big Data in Manufacturing Industry

 

Abstract: With the deepening of the national integration policy between informatization and industrialization, China has gradually transformed from a big manufacturing country to a big data manufacturing country. This report explores how the massive industrial big data can enable industrial development through the actual case of the intelligent control and operation and maintenance of sewage treatment system in the farming industry of Sunner Group. At present, the industrial sewage treatment equipment basically adopts simple open loop control, and even many technological operations still have a trial-and-error adjustment by the manual mode, which also makes the treatment equipment in harsh environment require a lot of manual operation and maintenance. Aiming at the decentralized chicken farming model of Sunner Group, a mobile and distributed industrial sewage treatment scheme is adopted, which greatly improves the treatment efficiency and reduces the cost. At present, most of the water quality automatic detection device is based on the reagent detection method of the national standard method, which has long measurement cycle and high cost, and cannot realize real-time water quality detection. The big data-driven soft sensor designed in the case can obtain low-cost online water quality measurement data, thus realizing closed-loop sewage treatment and real-time predictive operation and maintenance.