Energy management control of grid-tied PV-battery system based on intelligent proportional allocation

Abstract

Most existing grid-tied PV-battery systems rely on static charge-discharge schedules, typically involving daytime battery charging and nighttime discharging, which lead to inefficient energy storage utilization and reduced overall economic performance. Traditional energy management techniques, primarily designed in large timescales, are hard to deal with the rapid and stochastic fluctuations of PV-battery systems, which reduces the modeling accuracy and economic efficiency of system operation. Furthermore, conventional centralized or hierarchical optimization frameworks often overlook the control requirements of underlying power converters, which makes it challenging to simultaneously achieve both high economic efficiency and operational stability. This paper aims to improve the economic efficiency and operational stability of PV-battery systems considering the control requirements of power converters. A novel energy management control strategy for grid-tied PV-battery systems is proposed, structured within a dual-layer architecture that integrates upper-layer intelligent optimization scheduling with lower-layer proportional coordinated converter control. Specifically, a robust optimization-based multi-timescale scheduling framework is formulated, integrating day-ahead and intraday scheduling to effectively mitigate system uncertainties and improve economic performance. On this basis, an optimized proportional allocation-based control strategy is proposed at the converter level. This method ensures precise tracking of upper-layer dispatch commands while substantially enhancing dynamic response and operational stability of the system. The correctness of the theories has been verified by simulations and experiments. To address the current challenges of insufficient robustness in master–slave control and limited accuracy in droop control, a new converter cooperative control method is needed that balances command tracking precision with operational robustness. This paper therefore investigates two levels, i.e. scheduling and control. At the scheduling level, a multi-timescale optimization model combining two-stage robust optimization with intraday rolling correction is developed to accommodate highly volatile operating scenarios. At the control level, an optimized proportional allocation strategy for energy management control is proposed to enhance system dynamic stability. Finally, simulations and experiments validate the effectiveness of the proposed method.