Solar tracker and agrivoltaic plant, how to improve the marriage
How can the solar PV tracker be even more efficient, mono-axis or double-ass? How can we increase their performance through the control algorithms in photovoltaic systems with two-faceted modules? And how can we best adapt this technology to the emerging segment of the ground-lifted agrivoltaic? A new German research project is answering all these questions. We are talking about DeepTrack, a state-funded initiative managed by Fraunhofer Institute for Solar Energy Systems ISE in partnership with Zimmermann PV-Tracker.
The project was created to apply the new artificial intelligence tools to solar PV trackers with a clearly defined objective: expand the use of diffuse light.
Solar PV tracker, types and operation
The solar tracker controls the arrangement of the photovoltaic modules in relation to the angle of incidence of the solar rays through an automatic tracking system. The objective is that the cells are always oriented during the day so that direct light hits their front surface perpendicularly. The advantage is intuitive but wanting to quantify it, according to the latest estimates, this technology makes possible a gain of efficiency from 20 to 30% compared to fixed plants on the ground.
Overall the PV trackers consist of a microprocessor that processes signals, light sensors, electromagnetic and mechanical motion control modules and a power supply system. Depending on the mode of their movement or directional flexibility, these devices can be classified as:
- Single-axis solar trailers.
- Dual-axis solar trailers.
At the technological level the most advanced systems are those classified Open Loop Trackers that employ advanced control algorithms.
IA at the service of solar traker
One of the requirements underlying the German project is to expand the criteria used to define the orientation of solar panels. So no longer just how to get the perfect impact on the front surface but, for example, modulate the light requirement for any crop underneath the modules. Zimmermann PV-Tracker and the Fraunhofer ISE are optimizing these tracking algorithms with a digital twin.
In detail, the company then built one of its photovoltaic systems with tracking on the test field of the Fraunhofer ISE to carry out measurements in natural environmental conditions. Based on these data, he developed a digital twin that combines photovoltaic monitoring and modelling tools with weather forecasts thanks to deep learning. This allows you to map the optimal tracking positions of the photovoltaic modules for different needs.
“In an initial phase we have developed control processes aimed at the optimal electrical yield of the two-factor solar modules or precisely to the requirements of a particular plant in agricultural photovoltaics,” explains Dr. Matthew Berwind, head of the Fraunhofer ISE. “The next step is to combine the two approaches to get the most out of both perspectives. Calculating this “optimal point” is challenging, but possible thanks to our concept based on artificial intelligence.”