In the quest for more durable and sustainable construction materials, a team of researchers led by Hossein Khosravi from the Department of Civil Engineering at Hakim Sabzevari University has made a significant stride. Their work, published in the journal *Scientific Reports* (translated to English as *Scientific Reports*), focuses on predicting the effectiveness of self-healing concrete using advanced machine learning techniques. This research could have profound implications for the energy sector, where infrastructure durability is paramount.
Concrete, the most widely used construction material, is notoriously prone to micro-crack formation. These tiny fissures can compromise the material’s durability and structural performance, leading to costly repairs and reduced lifespan of buildings and infrastructure. Self-healing concrete, which can autonomously repair these cracks, offers a promising solution. However, evaluating its effectiveness has traditionally been a time-consuming and expensive process, lacking standardized methods.
Khosravi and his team addressed this challenge by developing three hybrid predictive models that combine artificial neural networks (ANNs) with optimization algorithms. These models—ANN optimized with genetic algorithm (GA), particle swarm optimization (PSO), and the Levenberg–Marquardt (LM) algorithm—aim to predict the percentage of crack healing in self-healing concrete accurately.
“The integration of optimization algorithms with ANN provides a robust and reliable framework for predicting the crack healing percentage in self-healing concrete,” Khosravi explained. This approach not only accelerates the evaluation process but also offers valuable insights for designing more durable and sustainable cement-based materials.
The team assessed the performance of these models using multiple statistical indices and found that all three hybrid models outperformed previous studies. Notably, the ANN–LM model achieved the highest prediction accuracy, highlighting the potential of this method to revolutionize the field.
For the energy sector, where infrastructure often operates in harsh conditions, the ability to predict and enhance the self-healing capabilities of concrete could lead to significant cost savings and improved safety. “This research offers a pathway to more resilient infrastructure, which is crucial for the energy sector,” Khosravi added. By reducing the need for frequent repairs and extending the lifespan of structures, self-healing concrete could contribute to more sustainable and efficient energy systems.
The findings of this study not only advance our understanding of self-healing concrete but also demonstrate the power of machine learning in modeling complex nonlinear behaviors. As the construction industry continues to embrace digital transformation, such innovative approaches are likely to shape the future of material science and engineering.
In the broader context, this research underscores the importance of interdisciplinary collaboration, combining civil engineering expertise with cutting-edge computational techniques. As Khosravi and his team continue to refine their models, the potential applications of their work extend beyond the energy sector, promising to impact various industries that rely on durable and sustainable construction materials.