What is it about?
To help reduce CO₂ emissions, Double Stage Heat Transformers (DSHTs) have emerged as a promising solution for recovering thermal energy using a lithium bromide-water mixture. These systems can deliver heat at higher temperatures than they receive, making them highly efficient. To optimize their performance in real time, researchers compared two AI-based control methods—Fuzzy Logic and Artificial Neural Networks (ANN). While both showed high accuracy (R² > 0.98), ANN proved more effective, especially when managing refrigerant flow in the second evaporator. A 30-neuron ANN model delivered the most precise results, suggesting that future DSHT controllers should focus on this approach, particularly using evaporator flow data tied to a power level of 3.9×10⁻⁴ kg/kJ.
Featured Image
Photo by BoliviaInteligente on Unsplash
Why is it important?
The Artificial Neural Network (ANN) demonstrated superior accuracy (R² = 0.9990) compared to Fuzzy Logic (R² = 0.9847) for controlling the evaporator’s thermal power in a Double-Stage Heat Transformer (DSHT), making it the preferred method for calculating optimal operating conditions—including thermal flows and maximizing performance through a novel 30-neuron, 300-weight, 40-bias model.
Perspectives
The next step for this work is to move from simulations to the real world. This includes installing sensors, running the model in real-time, and verifying that it can accurately control key variables, such as evaporator flow and temperature. It will also be important to connect the model’s outputs to real physical measurements, especially in the second stage, and to see how well it handles changes in operating conditions. Finally, analyzing the energy savings and economic benefits will help determine its practical value and potential for wider use in energy-efficient heating systems.
Professor Rosenberg J Romero
Universidad Autonoma del Estado de Morelos
Read the Original
This page is a summary of: Control Strategy Based on Artificial Intelligence for a Double-Stage Absorption Heat Transformer, Processes, May 2023, MDPI AG,
DOI: 10.3390/pr11061632.
You can read the full text:
Contributors
The following have contributed to this page
