Optimization of a pump health monitoring system using fuzzy logic

Condition-monitoring plays an increasingly important role for technical processes in order to improve reliability, availability, maintenance and lifetime. As a consequence of increasing demands for product quality and efficiency as well as increased integration of control systems in costly and safety-critical mechatronic processes, the field of health monitoring is gaining interest. This paper is based on a current research project and describes the application of fuzzy-logic for the generation of health status information for pump systems. The unique quality of the presented approach is the collection of degradation information in form of a failure mode and effects analysis elaborated together with pump experts. This analysis resulted in multiple degradation indicators which are then used in a fuzzy system in order to generate health status information. This information is then concentrated by means of an algorithm which also uses an analogy to the failure modes and effects analysis. The result is a single number indicate the current health of a pump system.