Short investigation on system diagnosis regarding parameter estimation

Abstract

Fault Detection and Diagnosis (FDD) is a research field that has known a rapid development due to the increasing safety and dependability requirements in high-risk applications. Consequently, a significant number of FDD techniques and associated algorithms were developed, and their effectiveness was proven in practice. However, given the current trend that consist of designing and developing integrated real-time riskmanagement systems for safety-critical installations, it became obvious that most of these techniques, in their original form, fail to provide an efficient solution to the problems related to the implementation of the corresponding FDD schemes in realtime; this paper focuses on the use of parameter estimation techniques for achieving these objectives. First, a global overview of the current state of the FDD research is given, then the most commonly used parameter estimation techniques are described, with an emphasis on the possibility of using them within FDD schemes that respond to the requirements of modern industrial applications.