A new integrated on-line fuzzy clustering and segmentation methodology with adaptive PCA approach for process monitoring
- PDF / 1,268,581 Bytes
- 18 Pages / 595.276 x 790.866 pts Page_size
- 36 Downloads / 168 Views
ORIGINAL PAPER
A new integrated on-line fuzzy clustering and segmentation methodology with adaptive PCA approach for process monitoring and fault detection and diagnosis Hesam Komari Alaei • Karim Salahshoor Hamed Komari Alaei
•
Published online: 12 August 2012 Springer-Verlag 2012
Abstract A new on-line fuzzy clustering-based algorithm is developed using integration of an adaptive principal component analysis approach with a weighted fuzzy C-means (WFCM) methodology for process fault detection and diagnosis (FDD) applications. The proposed algorithm is based on the segmentation of measured multivariate time series process data through a sliding window scheme being realized in a bottom-up cluster merging approach to enable detection of probable changes embedded in their hidden structure. The method recursively maintain updated PCA models and their corresponding fuzzy membership functions based on the most recent arrival of each independent chunk of process data. The extracted chunk features are then retained in the memory to be merged using a new on-line fuzzy C-means methodology before incoming of the following chunks of data. A new formula is then presented for cluster merging improvement by incorporating an on-line weight to address the issue of cluster’s weight updating in the on-line WFCM methodology. The cluster merging mechanism is coordinated by a compatibility criterion, utilizing both similarities of the adapted clustersbased PCA models and their center closeness. The proposed algorithm has been evaluated on an artificial case study and Tennessee Eastman benchmark process plant. H. K. Alaei (&) K. Salahshoor Department of Automation and Instrumentation, Petroleum University of Technology, Ahvaz, Iran e-mail: [email protected] K. Salahshoor e-mail: [email protected] H. K. Alaei Department of Engineering, Ferdowsi University, Mashhad, Iran e-mail: [email protected]
The observed performances demonstrate promising capabilities of the proposed algorithm to successfully detect and diagnose the introduced fault scenarios. Keywords Time series data Fault detection and isolation Adaptive PCA On-line clustering On-line weighted fuzzy C-means On-line weight (OW) Gath–Geva algorithm
1 Introduction Industrial processes generate a large amount of operational data through measurements. These huge historical data embed diverse diagnostic patterns or signatures in multivariate time-series format which can be extracted to infer about the plant status, leading to an important data mining problem. There is a constantly increasing demand in competitive process industries to maintain higher performance, safety and reliability against any probable abnormal condition. The conventional approaches usually rely on awareness of operators to carefully monitor and analyze operational data for early fault detection and diagnosis (FDD). These approaches, however, do not give a deep insight to consolidate the process diagnostic data. On the other hand, it is beyond the capabilities of an operator or enginee
Data Loading...
