Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1180070 | Chemometrics and Intelligent Laboratory Systems | 2007 | 8 Pages |
Abstract
Diagnostics of a state of potentially dangerous technological processes and field equipment enables to detect abnormal situations and hidden (soft) failures at early stages of their development, when they are still reversible. In this paper, the combined method of diagnostics is considered. The early detection of abnormal process situations is carried out with the help of the “moving PCAÄÃÔ method by matching the values of statistics TÃ2 and Q with the thresholds. By the same method, the hidden faults of the field equipment (sensors, actuators etc.) are determined. However this method does not allow us to simply identify the abnormal situations when many variables are changing simultaneously. For this case, the methods of identification on the basis of production or frame-production diagnostic models (DM), in particular, systems on the basis of fuzzy production rules, have shown the good results. The procedure of identification consists in estimation of degree of similarity between a current situation vector S = {s1, s2, ⦠sJ} and vectors of possible abnormal situations registered in rules of a process diagnostic model Smâ= {s1mâ, s2mâ, ⦠sJmâ}. Elements siâ= μS(uiâ) reflect in opinion of the experts the “ideal” development of symptoms ui for the given fault. The quality of the method is illustrated by diagnostics of a state of a high-pressure polyethylene polymerization process.
Related Topics
Physical Sciences and Engineering
Chemistry
Analytical Chemistry
Authors
L.A. Rusinov, I.V. Rudakova, V.V. Kurkina,