SYSTEM ANALYSIS OF THE PROCESS OF PRESERVING ENERGY INVARIANTS WHEN REDUCING THE MATHEMATICAL MODEL OF AN ELECTRIC VEHICLE WITH RECUPERATIVE BRAKING

Abstract

From the standpoint of systems analysis, the problem of reducing mathematical models of an electric vehicle with an AC traction drive and a regenerative braking system for application in driver decision-support systems is considered. A systems approach is applied to the decomposition of the electric vehicle energy chain into functional subsystems, followed by an analysis of energy flows between them. The necessity of preserving energy invariants when reducing the model order is substantiated, as this constitutes a critical condition for adequate prediction of driving range and regeneration efficiency. A method of structural-energy reduction is proposed, based on the principle of maintaining the power balance at nodal points of the transmission energy chain. A theorem is formulated on the necessary and sufficient conditions for preserving the energy invariant when transitioning from the full model to the reduced one. It is proven that classical modal reduction methods violate the energy balance of systems with regeneration, which leads to a systematic error in driving range estimation. A modified balanced truncation method with energy correction is developed, which guarantees the preservation of the integral energy balance over an arbitrary time interval. Computer modeling in the MATLAB environment is performed to verify the proposed method on typical driving cycles of an electric vehicle with intensive regenerative braking. The simulation results confirm that the proposed method ensures an energy consumption estimation error of no more than 1.5 percent while providing a fourfold acceleration of computations compared to the full model. A criterion for assessing reduction quality based on the deviation of the energy invariant is proposed. Prospects for applying the developed method to the construction of adaptive energy management systems for electric vehicles with forecasting of optimal regeneration modes are outlined.