Robust distance measures for on-line monitoring: Why use euclidean?

Traditionally, the calibration of safety critical nuclear instrumentation has been performed at each refueling cycle. However, many nuclear plants have moved toward condition-directed rather than time-directed calibration. This condition-directed calibration is accomplished through the use of on-line monitoring which commonly uses an autoassociative predictive modeling architecture to assess instrument channel performance. An autoassociative architecture predicts a group of correct sensor values when supplied with a group of sensor values that is corrupted with process and instrument noise, and could also contain faults such as sensor drift or complete failure. This paper introduces two robust distance measures for use in nonparametric, similarity based models, specifically the L-1-norm and the new robust Euclidean distance function. In this paper, representative autoassociative kernel regression (AAKR) models are developed for sensor calibration monitoring and tested with data from an operating nuclear power plant using the standard Euclidean (L-2-norm), L-1-norm, and robust Euclidean distance functions. It is shown that the alternative robust distance functions have performance advantages for the common task of sensor drift detection. In particular, it is shown that the L-1-norm produces small accuracy and robustness improvements, while the robust Euclidean distance function produces significant robustness improvements at the expense of accuracy.