Detection and Isolation of Sensor Faults for a Heat Exchanger Based on Local Linear Fuzzy Models

Abstract

The development of a reliable fault detection and isolation (FDI) scheme for nonlinear processes is often laborious and difficult to achieve due to the complexity of the system. Neural networks and fuzzy models, able to approximate nonlinear dynamic functions offer a powerful tool to cope with this problem. In this paper, a multi-model approach for FDI of sensor faults on a highly nonlinear real world processes is introduced. The approach is based local linear fuzzy models and provides residuals which are structured in a way that detection of six and isolation of five different sensor faults over all ranges of operation becomes possible. The approach enables on-line supervision of the process. Furthermore, a sensitivity analysis of the residuals to different faults can be made, based on the parameters of the fuzzy models.