30 March 2023 in IEEE Transactions on Smart Grid
Electricity theft has caused enormous damage to grids safety and economy globally, bringing plentiful attention to electricity theft detection. However, the inherent problems of data imbalance, data sparsity, and data shift due to the residents will or the seasons remain challenges. Besides, there is a lack of appliance-level information provided by non-intrusive load monitoring to improve the data graininess in electricity theft detection. To address these problems, an electricity theft detection algorithm based on contrastive learning and non-intrusive load monitoring is proposed. Firstly, a semi-supervised learning architecture composed of Gramian angular field encoding for sequence visualization and contrastive learning architecture featuring few-shot learning is established for load monitoring and initial detection considering the inherent operating characteristics of appliances. Furthermore, after the filtration of typical regular-switched appliances by Kendalls coefficient of concordance, in-depth detection of abnormal operation routines of appliances is conducted for suspicious residents, aiming to improve the fault-tolerant ability and resist the emergence of unknown electricity theft methods. Finally, the electricity theft probability is computed to confirm fraudulent users. Both the public and practical datasets are utilized to verify the effectiveness of the proposed study, and the results show an overall better performance compared to other state-of-the-art algorithms.
3 June 2024 in IEEE Transactions on Smart Grid
Precise estimation of flexible loads with different elasticity characteristics from the substation scale is crucial for demand-side management, especially considering the facility cost and privacy-preserving problem. However, owing to the non-independent and identical distribution resulting from randomness and variation in residential load profiles, as well as the load signature overlapping from power consumption aggregation, substation-level deferrable load exhibits non-periodicity with limited correlation with external factors, posing considerable challenges for disaggregation. To bridge the research gap, this paper proposes a hierarchical federated perception algorithm to disaggregate both deferrable and interruptible loads from the substation level following an edge-cloud collaborative framework. Firstly, residential load profiles are clustered with Affinity Propagation to exploit the hidden structural features from substation data. Secondly, the bottom-level residential load perception model is established based on FedPer to extract shared characteristics from representative residents through a lightweight federated sequence-to-point structure in a privacy-preserving manner. Subsequently, after the layerwise model transfer from the residential level, the top-level substation load disaggregation model is established, capable of precisely tracking transient fluctuations in complex deferrable loads. The proposed algorithm is verified on both the simulation platform and public datasets, demonstrating its accuracy, robustness, and efficiency for substation-level flexible load disaggregation.
1 January 2026 in IEEE Internet of Things Journal
1 January 2026 in IEEE Transactions on Geoscience and Remote Sensing