Sensor fault detection and identification via Bayesian belief networks

A new Bayesian belief network (BBN) model with discretized nodes is proposed for fault detection and identification in a single sensor. The single-sensor model is used as a building block to develop a BBN model for all sensors in the process under consideration. A new fault detection index, a fault identification index, and a threshold setting procedure for the multi-sensor model are introduced. Single-sensor model design parameter (prior and conditional probability data) is optimized to achieve maximum effectiveness in detection and identification of sensor faults. The single-sensor model and the optimal values of the design parameters are used to develop a multi-sensor BBN model for a polymerization reactor at steady-state conditions. The capabilities of this BBN model to detect and identify bias, drift and noise in sensor readings are illustrated by an example of simultaneous multiple faults.

[1]  K. Kroschel,et al.  Applying Bayesian networks to fault diagnosis , 1994, 1994 Proceedings of IEEE International Conference on Control and Applications.

[2]  Coleman B. Brosilow,et al.  Nonlinear model predictive control of styrene polymerization at unstable operating points , 1990 .

[3]  Ann E. Nicholson,et al.  Sensor Validation Using Dynamic Belief Networks , 1992, UAI.

[4]  Nugroho Iwan Santoso,et al.  Nuclear plant fault diagnosis using probabilistic reasoning , 1999, 1999 IEEE Power Engineering Society Summer Meeting. Conference Proceedings (Cat. No.99CH36364).

[5]  Mark A. Kramer,et al.  Comparison of belief networks and rule-based expert systems for fault diagnosis of chemical processes , 1993 .