Researchers develop a system that identifies losses in solar panels caused by dirt

Studies by the IEA (International Energy Agency) indicate that, after solar radiation, dirt is the factor that most impacts solar panel efficiency, with estimated losses of between 3% and 5% per year. Furthermore, accumulated debris increases the temperature of the modules, which can compromise their durability and increase operating costs.

To address the challenge of efficiency losses in photovoltaic systems caused by the accumulation of dirt, researchers from Future Grid, a center that integrates the Embrapii Competence Center in Smart Grid and Electromobility at Lactec, under the coordination of Rodrigo Riella, developed Solo (Dirt Loss Detection System).

The project, led by Natália Menezes, responsible for the study, and Eduardo Massashi Yamao, technical coordinator of the project and AI specialist at Lactec, uses AI (artificial intelligence) to monitor the performance of photovoltaic systems in real time, automatically identifying when a drop in energy generation is caused by the accumulation of dirt.

Artificial intelligence for accurate monitoring

Solo's innovation relies on machine learning algorithms trained from data obtained in a highly instrumented experimental setup.

The system developed in the project takes into account seasonal variations and the impacts that climatic events, such as prolonged droughts or fires, can have on the efficiency of solar panels.

Although this feature is still in the validation phase, researchers use an artificial pollution methodology to simulate different levels of dirt on the photovoltaic modules.

This makes it possible to map the performance of the panels in different scenarios and build efficiency curves that accurately represent the behavior of the systems throughout the year, covering periods of high radiation to more critical conditions of waste accumulation.

Currently, Solo issues automatic alerts in binary format, indicating the need for cleaning the modules. The expectation is that, by the end of the project, the system will be able to classify dirt levels, allowing for more precise and efficient management.

An affordable and scalable system

Solo's operating system relies on the collection and analysis of a wide range of operational and environmental data. Various instruments, such as an EPE-standard solarimetric station, a pyranometer, anemometer, temperature and humidity sensors, and cameras, feed machine learning algorithms with detailed information about real-world operating conditions.

This data is cross-referenced with the systems' electrical parameters, including the energy production curve, allowing the software to accurately identify the levels of dirt that affect the panels' performance.

The acquisition of this data and the integration of the sensors used in the system are coordinated by Patryk Henrique da Fonseca, who works directly on the structuring and monitoring of the devices that make up the project's experimental setup.

Designed to be a flexible solution, the program can be installed in both residential systems and large solar plants. Its operation relies on external hardware, equipped with sensors that monitor the voltage, current, and temperature of the photovoltaic modules, all without relying on a direct connection to the inverters.

The system is designed to be interoperable, meaning it's compatible with different equipment configurations, with the potential for future direct integration into new devices in the industry. In addition to avoiding unnecessary or delayed cleaning, the system helps extend equipment lifespan and reduce operating costs, maximizing the energy efficiency of solar installations.

This becomes especially relevant in large projects, where each percentage point of loss can represent significant financial losses. For now, the Soil tests are being conducted in a controlled experimental environment that simulates the operation of solar power plants.

Project advances in research to become a commercial solution

SOLO is currently at an intermediate stage of technological maturity. Research began at level 2 on the TRL (Technology Readiness Level) scale and, at this stage, aims to reach TRL 4, a stage at which the technology has already been validated in a laboratory environment. To become commercially available, the project must still evolve to TRL 9, when the product is ready for market.

Development is made possible through the Embrapii Unit of Lactec, with support from RDI (Research, Development and Innovation) projects from ANEEL (National Electric Energy Agency) or other development programs

According to those responsible for the research, the results obtained so far indicate strong potential for SOLO to become a commercial solution, offering it as a piece of equipment, a built-in feature, or a specialized service. For now, the focus remains on technically enhancing the system.

The team is also evaluating the possibility of integrating Solo with automated cleaning systems, allowing alerts to generate direct orders for autonomous robots to wash the panels. This functionality, however, is planned for future phases of the research.

all the content of Canal Solar is protected by copyright law, and partial or total reproduction of this site in any medium is expressly prohibited. If you are interested in collaborating or reusing part of our material, please contact us by email: redacao@canalsolar.com.br.