Fault Detection, Isolation, and Prognosis for Complex System
1Zhejiang University, Zhejiang, China
2Hong Kong University of Science and Technology, Kowloon, Hong Kong
3National Tsing Hua University, Hsinchu, Taiwan
4Harvard University, Cambridge, USA
5Hill-Rom, Batesville, USA
Fault Detection, Isolation, and Prognosis for Complex System
Description
A complex system can be thought of as multiple interdependent working subsystems. With the increasing level of complexity, it may present more frequent unstable operation statuses. Once system faults have occurred, they can cause unrecoverable losses and unacceptable environmental pollution and so forth. It thus demands more effective and efficient techniques to monitor operation status, detect the occurrence and propagation of faults, and enable suitable decision making, before such anomalies result in great damage. As one of the most active research areas over the last few decades, fault diagnosis including detection, isolation, and prognosis has been of importance and necessary for improving the economy and safety of a complex system, ranging from industrial processes, such as steel production, papermaking, car manufacturing, and mineral processing, to biological processes.
In the era of big data of process industries, new challenge emerges for fault diagnosis with amount of data grown exponentially. In particular, there are many uncertainties in the system which show the complexity of characteristics and include multimode and dynamics, multilevel and multiscale, nonlinearities, and strong coupling effects amongst the variables. This special issue focuses on the state of the art of fault diagnosis methods and their different applications, as well as future trends.
Potential topics include but are not limited to the following:
- Fault diagnosis problems for batch processes
- Fault prognosis for industrial processes
- Fault self-recovery/self-healing control
- Product quality monitoring and prediction
- Process performance assessment
- Incipient fault detection and diagnosis
- Fault modeling, detection, and estimation
- Fault classification and discrimination
- Data-driven approaches and knowledge-based approaches
- Data-driven modeling and operational automation
- Intelligence-based supervisory control