UFPB and UNI project uses artificial intelligence to optimize green hydrogen production and expand advances in renewable energy.
The global search for clean energy sources continues to accelerate investments in technologies focused on green hydrogen. In this scenario, researchers from the Federal University of Paraíba (UFPB) and the National University of Engineering of Peru (UNI) have developed a digital twin model capable of simulating real-time operations in an experimental hydrogen production plant.
According to a publication by UFPB on May 26, 2026, the system uses artificial intelligence and artificial neural networks to monitor the plant’s performance, predict failures, and continuously update operational data. The advancement highlights UFPB, UNI, and Latin American researchers in one of the most strategic areas of the global energy transition.
UFPB and UNI bet on technology to accelerate green hydrogen
The project developed by the researchers creates a kind of virtual replica of UNI’s green hydrogen production plant in Peru. This digital copy can reproduce the behavior of the physical installation in real time, allowing for continuous analysis, forecasting, and monitoring of operations.
-
At 4,400 meters of altitude in the Himalayas, where the air is so thin it makes breathing difficult, India’s state oil company drilled a thousand meters of rock to reach a 240-degree subsoil and set up the country’s first geothermal plant.
-
An “invisible” plant can transform rivers meeting the sea into electricity 24 hours a day: French startup uses nanotechnology to extract energy from the salt difference between fresh and saltwater, without sun, wind, dam, or fuel.
-
An “invisible” plant can transform rivers meeting the sea into electricity 24 hours a day: French startup uses nanotechnology to extract energy from the salt difference between fresh and saltwater, without sun, wind, dam, or fuel.
-
Scientists develop a low-cost alternative to generate green hydrogen on a large scale using an innovative chemical process; discovery could revolutionize energy infrastructure, heavy industry, and sustainable transportation systems around the world.
In practice, the digital twin functions as an intelligent virtual environment. It receives operational data from the real plant and can identify patterns, variations, and potential failures before major problems occur.
According to the study’s authors, the model showed high accuracy in reproducing the hydrogen production of the experimental plant. This makes the technology a relevant tool for increasing operational efficiency and reducing waste.
Furthermore, the initiative reinforces the potential of UFPB and UNI in research focused on renewable energy and industrial innovation.
Researchers use artificial intelligence to predict plant performance
The system created by the researchers integrates different types of operational information from UNI’s experimental plant. Among the data used are electrical, thermal, and hydraulic variables collected directly from the physical structure.
To process this information, the model uses artificial neural network algorithms. The goal is to predict the yield of green hydrogen and assess the overall efficiency of the plant in different operational scenarios.
The applied methodology includes important steps, such as:
- Acquisition of operational data
- Pre-processing of information
- Data normalization
- Incremental system learning
- Continuous update of the virtual model
Another differentiator is the so-called batch retraining, a process of continuous relearning. In this method, new data is gradually incorporated into the system to improve the accuracy of predictions.
According to researchers, this constant update helps the digital twin keep up with real plant changes and reduce differences between predicted and experimentally observed results.
How green hydrogen gains strength in the energy transition
Green hydrogen is produced from the electrolysis of water using electricity generated by clean sources, such as solar, wind, and hydroelectric power. Unlike hydrogen produced with fossil fuels, the process practically does not generate carbon emissions.
In recent years, several countries have invested billions of dollars in projects related to green hydrogen. The fuel is seen as an important alternative to decarbonize industrial sectors considered difficult to electrify.
Among the segments that can use the technology are:
- Steel industry
- Heavy transport
- Fertilizer production
- Sustainable aviation
- Renewable energy storage
In this context, research developed by UFPB, UNI, and other academic centers gains strategic relevance for the future of renewable energy in Latin America.
Digital twin can reduce industrial failures and waste
One of the most important aspects of the study is the digital twin’s ability to anticipate operational problems. This can help companies and research centers reduce maintenance costs and avoid unexpected interruptions.
The technology also allows for testing virtual scenarios without the need to interrupt the operation of the physical plant. This increases operational safety and facilitates adjustments in industrial systems.
The researchers highlight that the developed model can eventually be integrated into more advanced operational control strategies and explainable artificial intelligence.
Furthermore, the methodology can be applied to larger plants and other electrolysis technologies aimed at green hydrogen production.
UFPB, UNI, and researchers strengthen international cooperation
The work was led by Professor Juan M. Mauricio Villanueva from the Center for Alternative and Renewable Energies at UFPB, in partnership with researchers from UNI.
The study also included researchers Rubén Aquize Palacios, Aurelio Morales-Villanueva, Cesar Briceño Aranda, Oswaldo A. Waters Torres, and James Erick Vílchez García.
The cooperation between Brazilian and Peruvian universities shows how international projects are becoming fundamental to accelerating technological advances related to renewable energy.
In addition to the exchange of scientific knowledge, these partnerships allow for sharing infrastructure, operational experiences, and new methodologies applied to the energy sector.
Renewable energy drives new projects in Latin America
The advancement of green hydrogen also opens economic opportunities for Latin American countries. Regions with great potential for solar and wind generation can become strategic producers of sustainable fuels in the coming decades.
Brazil, Chile, and Peru frequently appear in international studies as countries with favorable conditions to expand projects related to renewable energy and low-emission hydrogen production.
In this scenario, technological solutions like digital twins can help make industrial plants more efficient, safe, and competitive. Experts point out that the combination of artificial intelligence, industrial automation, and renewable energy is expected to profoundly transform the energy sector in the coming years.
What this advancement represents for the future of green hydrogen
The project developed by researchers from UFPB and UNI shows how Latin American universities are directly participating in the new global technological race for clean energy.
The digital twin created for the green hydrogen experimental plant can simulate real-time operations, predict failures, and continuously monitor operational changes. This enhances the system’s reliability and strengthens future industrial applications.
In addition to boosting research in renewable energy, the initiative demonstrates how artificial intelligence and smart monitoring can help accelerate the energy transition and make green hydrogen more efficient and accessible in the future.
With information from UFPB.
Source link
