Method for Detecting Disorder of a Nonlinear Dynamic Plant
Xuechun Wang, Vladimir Eliseev
TL;DR
This paper introduces a new method called CCF-AE for detecting disorders in nonlinear dynamic systems using input-output data and neural networks.
Contribution
The novel CCF-AE method detects disorders without needing a reference model, using cross-correlation and autoencoders.
Findings
CCF-AE outperforms CUSUM and EWMV in true detection rates and false alarm rates.
CCF-AE is more effective for detecting disorders in complex nonlinear processes.
The method was successfully tested on a nonlinear pH neutralization reaction process.
Abstract
This paper proposes a new disorder detection method CCF-AE for a scalar dynamic plant based only on its input–output relation using a cross-correlation function and neural network autoencoder. The CCF-AE method does not use the reference model of the dynamic object, but only considers real-time behavior changes, given by input and output time series. The proposed method was used to detect disorder in the process of a nonlinear pH neutralization reaction, and was compared with the cumulative sum control chart (CUSUM) and the exponentially weighted moving variance control chart (EWMV). The CCF-AE method demonstrates a better true detection rate and lower false alarm rate than CUSUM and EWMV. Also, CCF-AE has more advantages in detecting disorder of complex nonlinear processes.
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14
Figure 15Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsFault Detection and Control Systems · Mineral Processing and Grinding · Advanced Data Processing Techniques