Original Article

Co-simulating photovoltaics and environments: a proof-of-concept for solar forecasting operation

¹ CEREA – Ecole nationale des ponts et chaussées, EDF R&D, 9 Rue de la Physique, Marne la Vallée 77455, France
² EDF R&D – Dpt. Technology and Research for Energy Efficiency, 1 Avenue des Renardières, Écuelles 77250, France
³ INSA-Lyon, CNRS, CETHIL, UMR5008, Villeurbanne 69621, France
⁴ EDF R&D − Dpt. Fluid Mechanics Energy and Environment, 6 Quai Watier, Chatou 78401, France

^* e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Received: 30 June 2025
Accepted: 16 October 2025
Published online: 9 January 2026

Abstract

This work proposes a co-simulation framework designed to simulate the relation between operational quantities for photovoltaic (PV) systems and local environmental quantities. It is based on the Functional Mock-up Interface (FMI) and a simulation environment that instantiates a simple irradiance-to-power chain and a microclimate model. A generalisable data-mapping scheme is introduced to facilitate effective communication between numerical instances. This scheme relies on the precise definition of thermal boundary conditions in each functional mock-up unit. In addition, the irradiance-to-power chain includes a conversion stage (e.g., DC/DC converter, maximum power point tracking controller, ideal battery) that allows one to mimic a basic unit commitment schedule. The framework is then tested for a specific floating photovoltaic array in the context of day-ahead solar forecasting. In this context, operational numerical weather prediction (NWP) is used to initialise the co-simulation framework and the microclimate instance recomputes the downward solar and thermal irradiation based on the atmospheric profiles of temperature and water content (vapour, liquid), and takes into account the optical effects from aerosols. In doing so and focussing only on the clear-sky situation, the co-simulation framework is found to refine energy forecasting by approximately 3% when compared to standalone simulation fed by NWP data. The co-simulation also allows us to predict the increase in air temperature due to the PV / atmosphere feedback, which remains limited in the studied PV configuration (less than 1 °C at 2.5 m height). Finally, the co-simulation is applied for various fictive scenarios mimicking the change in operational conditions (total energy curtailment and evolving convective transfer efficiency). Reproducing the heat production of the PV module from the maximum power point voltage to the open-current voltage, the impact on the air temperature is found to be approximately 0.3 °C, so the increase in the magnitude of the heat island effect would not be significant. However, the change in downward thermal radiation during energy curtailment scenario can affect the surrounding environment. This is particularly the case for eco-photovoltaic systems such as agrivoltaics or floating PV, for which the proposed co-simulation should be able to better anticipate the physical forcings on the ecosystems.