Researchers from UFPB and UNI develop digital twin for hydrogen production plant
- 4 hours ago
- 1 min read
A scientific partnership between the Federal University of Paraíba and the National University of Engineering of Peru has resulted in the development of an advanced "digital twin" model applied to an experimental hydrogen production plant. This initiative represents a strategic advancement in the use of artificial intelligence for the optimization of sustainable energy systems, consolidating the convergence between industrial digitalization and the energy transition based on renewable sources.
Model integrates artificial intelligence and real-time monitoring of hydrogen production
The study created a virtual replica of the Peruvian university's physical hydrogen plant, allowing for the monitoring, simulation, and prediction of the system's operational performance in real time. The model combines electrical, thermal, and hydraulic data collected at the experimental facilities with artificial neural network algorithms capable of estimating energy efficiency and hydrogen production yield with high precision. According to the researchers, the tool demonstrated strong adherence to the behavior of the real plant, establishing itself as a reliable solution for monitoring, operational analysis, and decision support.
One of the project's distinguishing features is the system's continuous learning capacity, through incremental updates and "relearning" based on new operational data. This dynamic adaptation reduces discrepancies between predictions and experimental results, increasing the model's reliability in real-world scenarios. Researchers also highlight the technology's future potential for applications in predictive control, early fault detection, energy optimization, and expansion to larger-scale industrial plants, reinforcing the role of digital twins as a strategic tool for consolidating the hydrogen economy and the smart energy industry in Latin America.
