Jul 2026· Electric power systems research· 0 citations· 38 references
Computer Science
TL;DR
This work develops a unified one-day-ahead probabilistic forecasting framework that aligns temporal resolution, reconstructs the unavailable inputs, and derives causal features, and compares a modular post-hoc residual-quantile scheme with an integrated in-model quantile-learning scheme.
Abstract
Smart-building load forecasters are often trained offline on dense, multivariate, high-frequency data, but deployment may provide only hourly, feature-limited inputs. Missing features must then be reconstructed, and their errors can propagate through the model. If this input uncertainty is not reflected, prediction intervals may become miscalibrated, affecting demand-response scheduling. Our work examines where uncertainty should be placed once inference inputs are reconstructed. We develop a unified one-day-ahead probabilistic forecasting framework that aligns temporal resolution, reconstructs the unavailable inputs, and derives causal features, and we compare a modular post-hoc residual-quantile scheme with an integrated in-model quantile-learning scheme. The comparison uses three mid-scale Deep Learning (DL) backbones: recurrent, hybrid recurrent, and attention-based Temporal Fusion Transformer (TFT) models, under identical inputs, forecasting horizon, preprocessing rules, and training budgets. Results show that uncertainty placement is backbone-dependent. Integrated quantile learning is most reliable with the TFT, yielding 2.2-3.6% MAPE and 28-83W RMSE on the labeled test window, while producing intervals about 5x narrower than the modular intervals at the closest-to-nominal coverage level. Diebold-Mariano tests support the TFT ranking and the mixed behavior of the recurrent backbones. A reconstruction-sensitivity test shows that reconstructed inputs increase the Quantile Score (QS) by 106% while interval width remains nearly unchanged, indicating that the model does not automatically absorb reconstruction-induced uncertainty. Robustness checks against non-DL baselines and seasonal hold-out weeks support this ranking. Our results expose the limits of post-hoc residual quantiles when inference depends on reconstructed inputs.
A novel probabilistic framework for net load forecasting employs Dual-coefficient network (DUCET) to model both input-space and output-space distribution characteristics, thereby enhancing robustness against distribution shifts and resolving the quantile crossing problem commonly observed in quantile regression.
Chaojin Cao, Yaoyao He· International Conference on...· 0 citations
Accurate transformer oil-temperature forecasting is important for thermal-risk assessment and operational planning. However, reported gains from complex forecasting models may be affected by future information leakage, weak seasonal baselines, inconsistent target periods, and test-based model selection. This study establishes a leakage-free, target-aligned framework for direct forecasting at 6, 12, and 24 h, integrating controlled model comparison, input-utility analysis, exact temporal interpretation, and cross-dataset confirmation. Only information available at or before the forecast origin is used, and identical validation and test target periods are maintained across horizons and lookback lengths. Naive predictors, regularized autoregression, ensemble methods, XGBoost variants, deep sequence models, and linear-nonlinear hybrids are evaluated using expanding-window validation and moving-block bootstrap analysis. OT-only Ridge regression with a 72 h lookback and $\lambda = 10^{-4}$ was selected for all three horizons, achieving ETTh2 RMSEs of 3.1568, 4.0042, and 4.2920. After ETTh1-specific refitting, the corresponding RMSEs were 1.3701, 1.7532, and 2.0976. Ridge significantly outperformed the daily-seasonal baseline at 6 and 12 h, while the 24 h gain was not statistically distinguishishable. The six historical load channels provided no robust incremental value. Exact Ridge contributions showed a shift from recent thermal persistence at 6 h to dominant daily-cycle dependence at 24 h. Rapid cooling was overpredicted and rapid heating was underpredicted. Overall, increased model complexity did not provide a consistent advantage under a controlled protocol, while the combined evaluation, interpretation, and cross-dataset confirmation offer reproducible empirical guidance for transformer oil-temperature forecasting.
The experimental results show that weekly scale look-back windows (LBW) provide an optimal balance between accuracy and computational cost for day-ahead horizons and the DL model NHiTS, when integrating exogenous features and a 7-day LBW, improves forecasting accuracy by 50.8%, outperforming all other deep learning, machine learning, and statistical forecasting models.
Rubaiath E. Ulfath, Chi-Tsun Cheng, Toh Yen Pang et al.· Scientific Reports· 0 citations
A deployment-oriented environmental-AI pipeline for day-ahead hourly PV forecasting that corrects timestamp conventions, constructs leakage-safe solar-geometry and clearness-index features, adds short-term atmospheric context, and combines complementary predictors through validation-learned stacking is developed.
Fariba Dehghan, Sebastian Stein, V. Yazdanpanah et al.· 0 citations
An integrated probabilistic forecasting framework with three linked stages: extreme-weather load identification, TimeGAN-based sample augmentation, and conformal quantile forecasting, which improves forecasting accuracy under extreme-weather conditions.
Hao Zhang, Xiyang Liu, Ruotian Gao et al.· International journal of pat...· 0 citations
For smart grid power system planning and operation, short-term load forecasting is crucial. Important decisions including determining system safety, scheduling fuel, economically dispatching electricity, and selling energy can be aided by accurate day-ahead estimates. However, due to its reliance on external variables like weather, the process is intricate and computationally intensive. In order to address this issue, the paper proposes an LSTGR-based architecture that systematically enhances STLF in Smart Grids. The input characteristics are first scaled correctly using data normalisation. A hybrid feature selection method combining XGB and RF is utilised to determine the most essential features. Afterwards, RFE is employed to eliminate superfluous attributes. To facilitate learning, a hybrid deep learning model is trained using the updated dataset. This model combines LSTM and GRU. Using assessment criteria such as MAPE, MAE, MSE, and RMSE, the results demonstrate that the LSTGR model outperforms other models. With an RMSE of only 1.8%, the model clearly excels at producing accurate predictions. All things considered, the model successfully improves the reliability of forecasts while being computationally efficient. Because of this, it is an excellent option for smart grid applications in the actual world.
L. Jayavani, Banoth Ashwini, Kolkur Swabhavika et al.· 2026 7th International Confe...· 0 citations