Asynchronous harmonic analysis based on out-of-sequence measurement for large-scale residential power network

Within the trend of smart grid, distributed and intelligent metering has gained its popularity among large scale residential power network. Supervising and assessing power quality in time is the method to guarantee secure and trustworthy energy for the demand side, and harmonics measurement and analysis are of necessity. Harmonics metering under the large-scale distributed metering architecture faces the out-of-sequence measurement (OOSM) problem which brings latency in communication and data fusion. This paper depicts a distributed measuring schedule for large scale harmonic analysis, in which compressive sensing (CS) is utilized to decrease the harmonic sampling length and transform packets, and Nonlinear Autoregressive model with Exogenous inputs (NARX) model is exploited to reorder the out-of-sequence measuring data at the meter data center. Experiments on a practical residential power utilization environment are implemented and harmonic identification accuracy is adopted to evaluate the measuring mechanism under distributed metering network. Results demonstrate a improvement of the harmonics analysis precision and validate the asynchronous measuring method in large scale residential power network.