STATISTICAL MODEL-BASED SENSOR DIAGNOSTICS FOR AUTOMATION SYSTEMS

A method for the automatic detection of abnormal sensor values in automation systems is described. The method is based on statistical generative models of sensor behavior. A model of normal sensor behavior is automatically constructed. Model parameters are optimized using an on-line maximum-likelihood algorithm. Incoming sensor values are then compared to the model, and an alarm is generated when the sensor value has a low probability under the model. Model parameters are continuously adapted on-line. The system automatically adapts to changing conditions and sensor drift, as well as detecting isolated abnormal sensor values.