Application of Rough Set Theory to Data Mining of Condenser Diagnosis in Power Plants

Rough set theory is a powerful tool in deal with vagueness and uncertainty. It is particularly suitable to discover hidden and potentially useful knowledge in data and can be used to reduce features and extract rules. This paper introduces the basic concepts and fundamental elements of the rough set theory. A reduction algorithm that integrates a priori with significance is proposed to illustrate how the rough set theory could be used to extract fault features of the condenser in a power plant. Two testing examples are then presented to demonstrate the effectiveness of the theory in fault diagnosis.