DFTMicroagg: a dual-level anonymization algorithm for smart grid data

Abstract The introduction of advanced metering infrastructure (AMI) smart meters has given rise to fine-grained electricity usage data at different levels time granularity. AMI collects high-frequency daily energy consumption that enables utility companies and aggregators perform a rich set grid operations such as demand response, monitoring, load forecasting many more. However, the privacy concerns associated with been raised. Existing studies on anonymization for focused direct application perturbation algorithms, microaggregation, protect consumers. In this paper, we empirically show reliance microaggregation alone is not sufficient data. Therefore, propose DFTMicroagg algorithm provides dual level improve privacy. leverages benefits discrete Fourier transform (DFT) provide additional layer protection. We evaluated our two publicly available datasets millions readings. Experimental results based clustering analysis using k-Means, classification via k-nearest neighbor (kNN) mean hourly forecast Seasonal Auto-Regressive Integrated Moving Average eXogenous (SARIMAX) factors model further proved applicability proposed method. Our approach more flexibility control protection their published