抄録
A model-based performance monitoring and diagnostic method for a heat exchange process in a distributed energy system is developed. This method compares the values of measured process variables to those estimated by an input-output model to detect performance deterioration, and then identifies its cause by pattern matching with trend database of process variables under typical faults. First, the trend of the process variables of a shell-and-tube type heat exchanger with scale adhesion inside and outside heat-transfer tubes is analyzed through a numerical simulation. The result shows that the representative transfer-tube temperature and differential pressure between inlet and outlet of the heat-transfer tube have distinctive trends under scale adhesion. Moreover, it is revealed that by using the trends of the abovementioned process variables, the scale adhesion inside and outside heat-transfer tubes can be diagnosed, and the scale adhesion and the impulse line blockage in the differential pressure transmitter can be differentiated.
