Method for Identification of Structural Parameters of Electric Power Consumption of Industrial Enterprises Based on Phase Space Reconstruction

The main objectives of the study are to develop a new adaptive geometric method for real-time monitoring of industrial electricity consumption schedules to ensure timely detection of structural imbalances and minimization of financial risks for energy-intensive enterprises in competitive electricity markets. To achieve these objectives, the following tasks were accomplished: transition from conventional one-dimensional statistical analysis to multidimensional topology using the method of phase space reconstruction of discrete load time series with consideration of commercial metering intervals; development of an algorithm for spatial limitation of reference phase trajectories through the construction of bounding rectangles; calculation of geometric parameters of the formed daily quasi-cycles, in particular centroid coordinates and semi-perimeters, for accurate identification of the current state of the system; typologization of characteristic behavioral patterns of electricity consumption based on a system of algebraic inequalities. The most significant result is the mathematical confirmation of the regularity of the drift of the centers of mass of daily quasi-cycles within a narrow neighborhood of the first quadrant of the phase plane, which proves the high inertia of the process and its tendency toward autocorrelation. The significance of the obtained results lies in providing dispatch personnel of industrial enterprises with a transparent and fast tool for real-time monitoring of energy consumption without requiring substantial computational resources. This ensures the possibility of immediate managerial decision-making regarding the operational purchase or sale of corresponding volumes of electricity in the intraday market, thereby effectively avoiding financial penalties and optimizing overall production costs under conditions of unstable equipment operation schedules.