The SRPSE dataset simulates load fluctuations on 3 load nodes during a time period of 60 days. The load nodes are assumed to contain 90000, 100000, and 125000 households, respectively. By generating the SRPSE dataset, the load consists of 900, 1000, and 1250 households’ electricity with 100 times magnification afterwards in our paper. The household’s electricity usage behavior follows a given mode. The appliance’s waveform in each household comes from the PLAID [1] dataset. The PLAID samples 11 types of appliances at 30 kHz which is down-sampled to 100 Hz in our dataset. Other sampling frequency data used in our paper is based on the 100 Hz SRPSE dataset.

Notice: Those wishing to use the dataset in academic work should cite the following paper as the reference.

Liang, Gaoqi, et al. "Super resolution perception for improving data completeness in smart grid state estimation." Engineering 6.7 (2020): 789-800.

1. Released: Different household behavior description

2. Released: SRPSE dataset (values before the 100 times magnification)

[1] J. Gao, S. Giri, E.C. Kara, and M. Berges, PLAID: a public dataset of high-resolution electrical appliance measurements for load identification research: demo abstract, Proceedings of the 1st ACM Conf. Embedded Systems for Energy-Efficient Buildings, 2014: 198-199