Development of fault detection and diagnosis schemes for industrial refrigeration systems - lessons learned

The success of a fault detection and diagnosis (FDD) scheme depends not alone on developing an advanced detection scheme. To enable successful deployment in industrial applications, an economically optimal development of FDD schemes are required. This paper reviews and discusses the gained experiences achieved by employing a combination of various techniques, methods, and algorithms, which are proposed by academia, on an industrial application. The main focus is on sharing the "lessons learned" from developing and employing fault-tolerant functionalities to a controlled process in order to meet the industrial needs while satisfying economically motivated constraints.