Short-Term Industrial Power Load Forecasting Based on XGBoost and Multi-Dimensional Feature Engineering

Day-ahead power load curve forecasting based on multi-objective trade-off feature construction and selection

Against the backdrop of the rapid development of modern power systems, day-ahead load forecasting faces increasingly complex challenges. Current methods largely rely on a single evaluation criterion for feature selection, making it difficult to balance feature relevance and redundancy. Meanwhile, the introduction of attention mechanisms often lacks clear motivation and visual performance demonstrations, limiting their full potential and impeding improvements in prediction accuracy and curve smoothness. To address these issues, this paper proposes a deep learning forecasting method that integrates feature construction with multi-objective feature selection. First, an extended feature set combining time series structural features and external driving factors is constructed, and a fitness function based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is designed to achieve a multi-objective optimization balancing predictive relevance and feature redundancy. Second, dynamic weight allocation is implemented through an attention mechanism to focus on critical time periods, with visualizations intuitively demonstrating its effectiveness in suppressing prediction “sharp oscillations” and enhancing trend perception. Experimental results show that the proposed method significantly improves forecasting performance and curve smoothness across multiple real-world load datasets.

Short-Term Electrical Load Forecasting Based on IMBKA-BiGRU-Attention Model

This paper proposes a short-term electrical load forecasting method based on a BiGRU-Attention network optimized by an improved multi-strategy black-winged kite algorithm (IMBKA), which achieves favorable forecasting performance among the compared models.

Short-term wind power forecasting with lag-based feature engineering: a turbine-level machine learning comparison

Short-Term Wind Power Prediction Method Based on Multi-feature Adaptive Fusion and Extreme Value Theory Tail Calibration

Accurate short-term wind power prediction is a critical foundation for improving the accommodation capacity and dispatching security of power systems. Aiming at the problems of multicollinearity in traditional models and large prediction errors under extreme fluctuations, this paper proposes a prediction method combining multi-feature fusion ridge regression with extreme value theory. First, multiple ridge regression branches are constructed, and multi-dimensional electrical features such as voltage and reactive power are introduced. L2 regularization is applied to alleviate variable multicollinearity, and a difference strategy is adopted to adapt to the inertial characteristics of short-term power. Second, the optimal backbone model for different prediction horizons is adaptively matched based on validation set errors, and weighted fusion with the persistence model is performed to output point prediction results. Finally, the POT-GPD model is used to fit the tail of errors, and a 95% probabilistic prediction interval is constructed. Experiments show that the average error of the proposed model across multiple horizons outperforms comparison models, and the prediction interval coverage rate exceeds 95%, which balances both conventional prediction accuracy and early warning capability for extreme wind power fluctuation risks.

Short Term Load Forecasting Using Sliding-Window LSTM in DKI Jakarta Power Subsystems

Subsystem-level short-term load forecasting is important not only for capturing heterogeneous load characteristics, but also for maintaining system reliability and supporting accurate operational decision-making. While many existing studies focus on aggregated system-level forecasting, such approaches may may fail to capture localized load dynamics localized load dynamics in large interconnected systems. This paper proposes a validation-driven, subsystem-aware multi-horizon Stacked LSTM framework for seven power subsystems in DKI Jakarta using two years of 30-minute resolution operational data (August 2023-July 2025). The proposed framework integrates validation-driven sliding window optimization, architecture tuning, and weather feature evaluation within a structured experimental pipeline. A strict chronological split is applied, and final performance is evaluated on a fully unseen operational month (July 2025) to ensure outof-sample generalization. Across the seven subsystems, the optimized models achieve test MAPE values ranging from 3.34% to 9.24%, demonstrating reliable forecasting performance under heterogeneous load dynamics. Weather feature evaluation revealed subsystem-dependent responses. A Wilcoxon signed-rank test indicates no statistically significant difference between the Baseline and All-Weather configurations p-value 0.43, despite subsystem-dependent responses to meteorological variables. The results indicate that subsystemspecific parameter configurations, determined through validation, are essential for accurate subsystem-level short-term load forecasting in heterogeneous smart grid systems, while weather information should be incorporated selectively according to subsystem characteristics.

Multi-site Wind Energy Prediction System Based on Power Decomposition and Deep Model Integration

A multi-site wind power forecasting system based on power decomposition and deep model ensemble that applies Variational Mode Decomposition (VMD) to separate raw power sequences into high-frequency and low-frequency components, each directed into a structurally symmetric dual-branch framework.