The beginning of the 21st century for the fuel and energy complexes of most countries is characterized as a period of active restructuring and a fundamental shift in developmental priorities. The basis of these changes is technological development. Industry 4.0 technologies have particular importance in achieving maximum optimization of production processes. In the same way, they are applicable in establishing effective interaction between the energy sector and other sectors of the economy. The authors outline an approach to assessing the country’s fuel and energy balance state through the selected properties: sustainability, accessibility, efficiency, adaptability and reliability. Hence, a model of the fuel and energy complex was created on the example of the Russian Federation, considering the country’s territorial and functional division. The methodology is based on scenario modeling of the influence level of external challenges in conjunction with the accompanying technological development. The mathematical model allowed forecasting changes in the properties of the energy system. The scientific significance of the work lies in the application of a consistent hybrid modeling approach to forecast the state of the fuel and energy balance. The results of the study are useful in compiling scenarios for the regional and entire development of the fuel and energy complex. Further model improvements should include an expansion of the number of counted industries and their relations.
Non-intrusive load monitoring (NILM) can provide actionable feedback for demand response (DR) when direct measurements of device states are unavailable. We propose a reproducible, engineering-oriented pipeline for detecting ON/OFF states of end-use load groups from an aggregated active power time series. The method uses robust hysteresis-based labeling with adaptive thresholds derived from the median and median absolute deviation, followed by compact feature engineering restricted to global active power (GAP). After removing collinear features (|r| > 0.98), permutation importance is used to retain informative predictors. Probabilistic binary classifiers (LGBM, Histogram-based Gradient Boosting, XGBoost, and CatBoost) are trained for each target load, and the decision threshold is optimized via Fβ to balance missed events and false alarms. A post-processing stage stabilizes predictions by smoothing probabilities and suppressing spurious triggers. Model quality is assessed with both sample-wise metrics and event-based metrics that credit the correct detection of switching intervals within a time tolerance. Experiments on the open “Individual Household Electric Power Consumption” dataset (1-min resolution, 2007–2010) demonstrate that lightweight gradient boosting models, particularly LGBM, deliver reliable and interpretable state estimates suitable for practical DR integration and edge deployment.