Ayesa helps bring artificial intelligence to vineyards to combat the effects of climate change

Through the Ibermática, the company is developing a computational digital-twin framework to optimise the flow of solar power at vineyards.

Ayesa, a global provider of technology and engineering services, is participating in a project designed to optimise crop yield on vineyards with solar arrays through the use of digital twins. The solution, which is based on artificial intelligence, adjusts the flow of solar power through complex, agrivoltaic systems with rapidly changing conditions using data from sensors.

This is a particularly important project given the threat climate change poses to Spain’s wine industry. Warmer temperatures as well as changes to rainfall and radiation have had a significant impact on vineyards, resulting in changes to the organoleptic properties of the grapes, including their taste. This is a particularly serious problem for wines with protected denominations of origin, the grapes for which must be grown within specific geographical areas.

It is within this context that VidVolt 4.0 arose, a project led by INNOVI – Clúster Vitivinícola Català, which Ayesa is participating in through the Ibermática Foundation, its R&D division. Its aim is to develop a computational digital-twin framework to track and optimise the flow of solar power on vineyards, and thus maximise their yield at their current locations.

The model will carry out a complex and multipurpose analysis based on data received from sensors on the ground in order to optimise the agrophotovoltaic (APV) system in question. APV systems symbiotically cohabitate power generation facilities and agricultural production systems.

Digital twins

The digital-twin model allows a solar installation to be tested virtually from multiple source directions in different scenarios, using a genomic-based machine-learning algorithm to optimise the system. The configuration which offers the greatest yield will then be used.

With this method, solar power flow is rapidly computed with a reduced order model of Maxwell’s equations, based on a high-frequency decomposition of the irradiance into multiple rays, which are propagated forward in time to ascertain multiple reflections and absorption for various source-system configurations, varying multi-panel inclination, panel refractive indices, sizes, shapes, heights, ground refractive properties, etc.

VidVolt 4.0 will provide a suite of software and machine learning models able to analyse the complex and changing conditions on vineyards with APV systems.

In order to bring artificial intelligence to these systems, the project focuses on the following:

• Researching and developing systems able to collect and process high volumes of varied data, and work in accordance with all the relevant protocols.
• Effectively identifying anomalies and optimising the position of solar panels as crops grow.
• Interaction with users through a human-in-the-loop approach. Particularly important is ensuring users are able to effectively interpret and learn from the results provided.

The project is supported by Next Generation EU funds through the Spanish Ministry of Industry, Trade and Tourism. In addition to the Ayesa Foundation and INNOVI – Clúster Vitivinícola Català, Clúster de l’Energia Eficient de Catalunya, Km0 Energy, Tamic and INCAVI are also part of this project.