Integration of Fault Detection and Isolation with Control Using Neuro-fuzzy Scheme

• Main Authors: A. Asokan, D. Sivakumar
• Format: Article
• Language: English
• Published: IFSA Publishing, S.L. 2009-10-01
• Series: Sensors & Transducers
• Subjects: Fault detection; Neuro fuzzy scheme; Chi-square test; Integration of fault diagnosis and control
• Online Access: http://www.sensorsportal.com/HTML/DIGEST/october_09/P_509.pdf

In this paper an algorithms is developed for fault diagnosis and fault tolerant control strategy for nonlinear systems subjected to an unknown time-varying fault. At first, the design of fault diagnosis scheme is performed using model based fault detection technique. The neuro-fuzzy chi-square scheme is applied for fault detection and isolation. The fault magnitude and time of occurrence of fault is obtained through neuro-fuzzy chi-square scheme. The estimated magnitude of the fault magnitude is normalized and used by the feed-forward control algorithm to make appropriate changes in the manipulated variable to keep the controlled variable near its set value. The feed-forward controller acts along with feed-back controller to control the multivariable system. The performance of the proposed scheme is applied to a three- tank process for various types of fault inputs to show the effectiveness of the proposed approach.