Abstract:Power distribution network is responsible for the important function of supply and distribution of electric energy, which is seriously threatened insulation faults and difficult to realize insulation condition monitoring. With the development of the intelligent sensing technology and communication technology, internet of things (IoT) technology provides a new opportunity to realize the all-round real-time rapid perception of the operation status of smart distribution network and timely push of early warning information. However, the feature quantity of big data and high frequency represented by partial discharge (PD) detection poses a great challenge to IoT-based intelligence sensing ability and data analysis ability. In this paper, IoT-based distributed PD monitoring network architecture is proposed for PD status detection, and the design principle for IoT-based distributed PD intellsense sensors is proposed. On this basis, to meet the data transmission capacity of the low-power and wide area network (LPWAN), a simplified feature set extraction method of phase resolved partial discharge (PRPD) patterns is proposed according to the PD simulation tests. This paper has a certain reference significance for feature extraction and pattern recognition of PD on the edge of the distributed IoTs.

Research on Intellisense and Simplified Feature Set Extraction of Partial Discharge for Power Distribution Network.