This work presents a novel control method for multi-megawatt photovoltaic (PV) plants that is able to regulate each plant inverter and the battery system to mitigate PV power fluctuations. The proposed control method makes it possible to implement different PV ramp-rate control strategies based on the use of batteries and the limitation of inverters during positive fluctuations, which have been conceptually proposed in the specialized bibliography, but have. This work presents a novel control method for multi-megawatt photovoltaic (PV) plants that is able to regulate each plant inverter and the battery system to mitigate PV power fluctuations. The proposed control method makes it possible to implement different PV ramp-rate control strategies based on the use of batteries and the limitation of inverters during positive fluctuations, which have been conceptually proposed in the specialized bibliography, but have omitted how to perform the coordination between PV generators. The dynamic model and the tuning of the control parameters are presented and the method is used to correctly implement different inverter-limitation strategies using 5-second data from a real 45 MWp PV plant. Furthermore, a new control strategy is proposed. This strategy reduces curtailment losses to negligible values and takes into account and addresses the intrinsic asymmetry in the battery charging and discharging capability, an issue that has been overlooked in the specialized bibliography. The results show that the proposed control method can effectively control each of the multiple inverters in order to obtain the desired PV plant operation to regulate the battery charging power, even during highly fluctuating scenarios.••••New method, Smart Limitation Loop (SLL), to perform PV curtailment at plant level••SLL demonstrated simulating benchmark PV smoothing strategies through a year.••New strategy, Smart Coordinated Inverter Limitation (SCIL), allows safer battery operation.••SCIL reduces annual curtailment losses below 1 % independently on ramp-rate restriction.Inverter limitationBattery energy storage system (BESS)PV smoothingPV-battery integrationNowadays, photovoltaic (PV) power is one of the generation technologies with the lowest levelized cost of energy (LCOE) [,, ]. Consequently, and due to the urgent need to reduce greenhouse emissions, the worldwide PV generation capacity has dramatically increased over the last few years [4,5]. Despite the benefits of this extraordinary growth in PV installations, this situation can also challenge the quality and stability of the power grid: the intrinsic intermittency of the solar resource and the increase in the proportion of power electronic-based generators to the detriment of synchronous ones (thereby reducing the inertia of the electrical system). The potential concerns associated with severe PV power fluctuations, particularly in scenarios of weak grids or high PV penetration, are voltage fluctuations [6,7] and frequency deviations [8,9]. These concerns have led some transmission system operators (TSO) to request ramp-rate limitations on power dispatched from PV and other renewable sources (Puerto Rico, Ireland, China, Denmark, Australia or South Africa, among others).In order to meet these requirements, PV projects must deal with the excess or lack of energy caused by power fluctuations. A number of strategies have been proposed, the vast majority of which require energy storage systems (ESS), mainly Lithium-ion batteries, to maintain the dispatched power within the requi. Over a one-year period, data were recorded at the Amareleja PV plant (Fig. 1(a)) in southern Portugal (38°11′20″N, 07°12′08″0W). The 250-hectare PV plant consists of 2520 vertical solar trackers, with a tilt angle of 45° and a ground cover ratio (GCR) of 0.162. Each group of 36 solar trackers are grouped together and connected to a 550 kW DC/AC inverter. With a total number of 70 inverters, the plant peak power is 45.6 MW while the inverter rated power (PN) is 38.6 MW.The synchronized series of each inverter were obtained with a 5-second sampling period. In post-processing, the series for the 70 inverters were grouped into pairs to obtain data with a nameplate capacity of 1.1 MW, which is currently more realistic than the original 550 kW, and to reduce the computational effort during simulation. In this way, the PV plant analyzed would consist of 35 equivalent inverters.
