Supervision of an industrial steam generator. Part II: Online implementation

A bond graph model based Fault Detection and Isolation approach applied to design of a supervision system for a complex steam generator is presented. The non-linearities and energetic couplings involved in the dynamics require a suitable and adaptive supervision environment. The analytical redundancy relations are derived from bond graph model of the process by utilising of the structural and causal properties of the model. The decision procedure to detect and isolate the faults uses the residuals, evaluated online. Due to measurement noises and uncertainties in the process, many complementary decision procedures based on different methods of signal treatment are used to form an integrated supervision environment. The integrated platform allows to monitor the plant in normal operation as well as in the presence of faults. Finally, redundancy information is used for both automatic and manual reconfigurations to safely increase process availability by using adaptive real-time modifications to process control laws as well as equations of the process and the residuals to reflect the dynamics of the new configuration.