Failure Precursors and Anomaly Detection in Complex Electrical Systems Using Symbolic Dynamics

Failures in a plant's electrical components are a major source of performance degradation and plant unavailability. In order to detect and monitor failure precursors and anomalies early in electrical systems, we propose to develop signal processing capabilities that can detect and map patterns in already existing, available signals to an anomaly measure. Toward this end Professor Asok Ray at Penn State University has pioneered an elaborate mathematical theory of "language measure" based on real analysis, finite state automaton, symbolic dynamics and information theory. Application of this theory for anomaly detection results in a robust statistical pattern recognition technique. This technique is superior to conventional pattern recognition techniques such as neural networks and principal component analysis for anomaly detection because it exploits a common physical fact underling most anomalies which conventional techniques do not. This superiority has recently been demonstrated on electrical circuits, lasers and in mechanical components. The objectives of the research proposed by Intelligent Automation Incorporated (IAI) and its subcontractor are: (i) to develop real-time anomaly sensing and monitoring systems for early detection of faults in avionic electrical systems; and (ii) to experimentally validate the proposed concept on an active nonlinear electrical circuit.