Sensor Failure Detection and Diagnosis

The fundamental idea is to utilize additional relevant process information for assessing the correctness of information generated by a sensor. This approach is known as the functional redundancy and it is more attractive than physical redundancy by duplicating sensors and using a voting logic to select the correct information. Several techniques based on statistics and system theory have been developed for validation of sensor information by functional redundancy. In most of these techniques, it is assumed that detailed process information is available a priori. Often, this knowledge is in the form of an accurate state-space model [39, 230]. In many cases, this type of accurate representation of a chemical process based on first principles is not available. 

This chapter introduces two sensor audit strategies that can detect and diagnose sensor faults. The first strategy (Section 8.1) focuses on sensor auditing by using calibration and test data sets that are processed either by developing PLS models (for data with low autocorrelation) or canoni- 

---