1. Fault diagnosis in linear systems with non-parametric method
- Author
-
Alexey Zhirabok, Alexey E. Shumsky, S. Solyanik, Alexander Zuev, and Sergey Pavlov
- Subjects
0209 industrial biotechnology ,Engineering ,021103 operations research ,business.industry ,Technical systems ,Linear system ,0211 other engineering and technologies ,Nonparametric statistics ,02 engineering and technology ,Fault (power engineering) ,Residual ,Fault detection and isolation ,Matrix decomposition ,020901 industrial engineering & automation ,Control theory ,Feature (computer vision) ,business - Abstract
The paper is devoted to the problem of fault isolation in technical systems described with linear dynamic models. So-called non-parametric, or model-free, or data-driven method is used to solve the problem. The feature of this method is that parameters of the system under consideration may be unknown. The methods of fault isolation and residual generation are suggested. The theoretical results are illustrated by an example.
- Published
- 2016
- Full Text
- View/download PDF