An online 2-D changepoint detection algorithm for sensor-based fault detection, is proposed. The algorithm consists of a differential detector and a standard detector and can detect anomalies and meaningful changepoints while maintaining a low false-alarm rate. The efficiency of the algorithm is validated by two industrial examples. It is thereby shown that the proposed algorithm can be used as an early warning indicator and prevent impending unit failures.
History
School affiliated with
School of Engineering (Research Outputs)
Publication Title
International Journal of Advances in Computer Science & Its Applications