Development of a Data-Based Method for Performance Monitoring of Heat Exchangers

A multivariate analysis method is developed for processing measurements, and for detecting and isolating faults and monitoring performance degradation in heat exchanger control loops. A heat exchanger inside a typical temperature to flow cascade loop is considered. This system includes a constant speed pump with flow control valves, pressure and temperature measurement. A proportional-integral-differential (PID) controller is used to maintain a temperature set point for the exit flow on one side of the exchanger. A thermal-fluid model for the components in the system is developed. A Fault Detection and Isolation (FDI) rule-base is formulated from results of simulations performed using these models. Measurements from an installed laboratory heat exchanger control loop are also used. Faults simulated and induced on the physical heat exchanger loop include tube fouling, sensor drift, fluid leakage, unresponsive valves, plugged process lines, and controller errors. The rule base allows the identification of faults in a heat exchanger control loop given suitable process measurements