A Sequential Bayesian Approach to Online Power Quality Anomaly Segmentation

Abstract

Increased observability on power distribution networks can reveal signs of incipient faults which can develop into costly and unexpected plant failures. While low-cost sensing and communications infrastructure is facilitating this, it is also highlighting the complex nature of fault signals, a challenge which entails precisely extracting anomalous regions from continuous data streams before classifying the underlying fault signature. Doing this incorrectly will result in capture of uninformative data. Extraction processes can be confounded by operational noise on the network including harmonics produced by embedded generation. In this paper, an online model is proposed. Our Bayesian Changepoint Power Quality anomaly Segmentation allows automated segmentation of anomalies from continuous current waveforms, irrespective of noise. Demonstration of the effectiveness of the proposed technique is carried out with operational field data as well as a challenging simulated network, highlighting the ability to accommodate noise from typical network penetration levels of power electronic devices.

Incipient Fault Detection in Power Distribution System: A Time–Frequency Embedded Deep-Learning-Based Approach

“…For example, in [14], a method based on human-level concept learning is proposed by selecting waveform features of current and voltage and decomposing it into primitives to detect faults. In [15], an online model based on sequential Bayesian approach is proposed by splitting power quality abnormalities of continuous current. This kind of methods are easy to implement, however, human selected features and thresholds rely heavily on expertise knowledge, and are not sufficiently capable of characterizing complex non-stationary signals to be well applied to incipient fault detection in distribution system.…”