arXiv:2601. 10494v3 Announce Type: replace-cross Abstract: With grid operators confronting rising uncertainty from renewable integration and a broader push toward electrification, Demand-Side Management (DSM) -- particularly Demand Response (DR) -- has attracted significant attention as a cost-effective mechanism for balancing modern electricity systems.
The paper presents a two-part workflow for analyzing smart‑meter data to uncover patterns of residential co‑adoption of photovoltaic (PV) systems and electric vehicles (EVs). First, dynamic time warping k‑means clustering identifies distinct daily import/export archetypes for PV‑only, EV‑only, co‑adopters, and neither groups, revealing a midday‑centered export pattern for many co‑adopters. Second, a bidirectional LSTM model trained on 21‑day windows achieves high detection performance (AUROC 0.991, macro‑F1 0.906) for PV/EV activity, outperforming tabular baselines and remaining robust across labeling rules and temporal splits.
SmartMeterFM is a conditional flow matching model that, once trained on monthly 15‑minute smart meter data, can generate realistic load and generation profiles, impute missing values, and perform super‑resolution without task‑specific retraining. By conditioning on customer and calendar attributes and using projection‑based guidance during inference, the model preserves temporal patterns and extreme values while matching supplied observations. In large‑scale case studies on Dutch consumption and photovoltaic data, SmartMeterFM achieved lower Continuous Ranked Probability Scores than task‑specific baselines for both imputation and 16‑fold super‑resolution, and matched or outperformed a task‑trained diffusion model on a public residential meter dataset.
The paper presents a method for exact machine unlearning in federated socially fair k‑means clustering, aiming to maintain fairness across demographic groups while allowing data deletion. It introduces separate local summaries per group so that unaffected institutions can keep valid global fairness weights, and demonstrates that reusing cached computations—especially when an entire client withdraws—can dramatically speed up retraining. Experiments on three datasets show significant speedups for initialization and moderate gains for individual record deletions, highlighting when exact unlearning is most beneficial.
Ireland’s smart metering programme records electricity use at 30‑minute intervals, which is too coarse to capture domestic appliance use. The authors present a label‑free disaggregation system that splits usage data into nine appliance categories by combining event detection for high‑power loads with questionnaire‑guided estimation. Evaluated on four datasets—including a large Irish smart‑meter dataset of over 4,800 years of use—the hybrid method achieves the lowest whole‑decomposition error and better month‑level performance than two independently developed systems.
arXiv:2607. 24237v1 Announce Type: new Abstract: Many existing clustering methods are designed based on a set-oriented definition---a cluster is a set of similar points---relying a point-to-point similarity function to find similar points.
arXiv:2607. 04949v1 Announce Type: new Abstract: We study the problem of k-means clustering on large datasets.
arXiv:2610.08538v1 Announce Type: cross Abstract: Probabilistic load forecasting has been widely studied for power-system operation and planning, but customer- and transformer-level forecasting intro...
LoaDiff is a diffusion-based generative model that produces year-long, sub-hourly smart‑meter electricity consumption time series. It can be conditioned on static household attributes like appliance ownership and dynamic factors such as calendar dates and outdoor temperature. Evaluations on three residential datasets show that LoaDiff generates realistic, diverse load profiles, limits memorization, retains useful information for downstream tasks, and responds coherently to conditioning changes.
arXiv:2609.06468v1 Announce Type: new Abstract: K-Means is one of the most widely used clustering algorithms, but its susceptibility to initial centroid selection remains a primary bottleneck for its...
arXiv:2609.06060v1 Announce Type: cross Abstract: Long-term electricity-consumption profiles exhibit several simultaneous periodic structures, including daily, weekly, and annual cycles. This work in...
The paper evaluates short‑term load forecasting methods that are sensitive to high‑demand periods across three distribution grid aggregation levels—area codes, secondary substations, and low‑voltage feeders—using UK and Swiss datasets. It compares statistical baselines, LightGBM, XGBoost, and foundation models Chronos Bolt and Chronos‑2, finding that Chronos‑2 delivers the best high‑demand performance and remains competitive overall. The study also shows that foundation model inference is fast enough for deployment and that peak‑aware evaluation and aggregation‑specific quantile selection can improve operational relevance.