This talk gives an introduction to our group’s work in developing intelligent learning approaches for data-driven diagnosis and control. These research topics are driven by practical demands and challenging scientific problems that require urgent solutions.
In the first part of this talk, we discuss the use of industrial big data for large-scale system monitoring and control. The target is to deal with industrial plants whose mechanism models are extremely hard to build and with little known expert knowledge. Since factories can provide huge amounts of historical data generated during operation processes, we will make use of the information therein and propose machine learning approaches to capture the characteristic dynamics of the systems and develop data-driven monitoring and control strategies.
In the next part, we present proposals for industrial cyber-physical systems (ICPS). ICPS is the central research subject and backbone of Industry 4.0, and this talk will discuss the key elements of ICPS and the application to vehicular CPS. It will show how interdisciplinary cooperation benefits design, and discuss the trend of in-depth cooperation among leading international research teams.
The third part of this talk will introduce our efforts towards applying intelligent learning approaches to human health diagnosis. Medical images, audio, video, etc. are analysed by intelligent learning approaches. The effectiveness of the proposed approaches gives us confidence in opening new possibilities for developing machine learning aided diagnosis schemes.
Shen Yin is a professor in the Department of Control Science and Engineering at the Harbin Institute of Technology. His research interests include Data driven control and monitoring, system identification, Fault diagnosis and fault tolerant control, artificial intelligence and machine learning, Big data and image processing for medical applications, and industrial applications on cyber-physical systems and industrial processes.