This paper presents a novel flexibility management system to handle short-term power deviations generated by intrinsic variability in generation and demand, and behind-the-meter flexibility provision in a close-to-real-time framework (30-second intervals). The proposed system employs an adaptive autoregression algorithm as a short-term forecast of energy exchange to the grid and a cost-benefit analysis to adjust the power setpoints of the energy devices by means of redispatching and unit reassignment modules. Experimental case studies are conducted in the PVZEN microgrid laboratory to validate the proposed system, testing critical scenarios and diverse flexibility requests. The results confirm the optimal management of the power deviations and the techno-economic feasibility of flexibility supply with real-world equipment, reducing the total operation cost from 11% to 85% compared to the baseline scenario. Finally, findings of previous studies based on simulations are verified experimentally, validating the influence of variables such as sample frequency, aging cost models, and flexibility price.
Forero-Quintero et al. (Fri,) studied this question.