Loading...

Multiple simultaneous fault diagnosis via hierarchical and single artificial neural networks

Eslamloueyan, R ; Sharif University of Technology | 2003

126 Viewed
  1. Type of Document: Article
  2. Publisher: Sharif University of Technology , 2003
  3. Abstract:
  4. Process Fault Diagnosis (PFD) involves interpreting the current status of the plant given sensor readings and process knowledge. There has been considerable work done in this area with a variety of approaches being proposed for PFD. Neural networks have been used to solve PFD problems in chemical processes, as they are well suited for recognizing multi-dimensional nonlinear patterns. In this work, the use of Hierarchical Artificial Neural Networks (HANN) in diagnosing the multi-faults of a chemical process are discussed and compared with that of Single Artificial Neural Networks (SANN). The lower efficiency of HANN, in comparison to SANN, in PFD is elaborated and analyzed. Also, the concept of a multi-level selection switch is presented and developed to improve the performance of hierarchical artificial neural networks. Simulation results indicate that application of multi-level selection switches increases the performance of the hierarchical artificial neural networks considerably
  5. Keywords:
  6. Fault diagnosis ; Process control ; Neural network
  7. Source: Scientia Iranica ; Volume 10, Issue 3 , 2003 , Pages 300-310 ; 10263098 (ISSN)
  8. URL: http://scientiairanica.sharif.edu/article_2635.html