Detecting persistent gross errors by sequential analysis of principal components

Abstract Measurements such as flow rates from a chemical process are inherently inaccurate. They are contaminated by random errors and possibly gross errors such as process disturbances, leaks, departure from steady state, and biased instrumentation. These measurements violate conservation laws and other process constraints. Data reconciliation aims at estimating the true values of measured variables that are consistent with the constraints, detecting gross errors, and solving for unmeasured variables. This paper presents an approach to construct a sequential principal component test for detecting and identifying persistent gross errors in data reconciliation by combining Principal Component Analysis and Sequential Analysis. The test detects gross errors as early as possible with fewer measurements. It is sharper in detecting, and has a substantially greater power in correctly identifying, the gross errors than the currently used statistical tests in data reconciliation.