In the heart of China’s Shaanxi province, researchers are tackling a longstanding challenge in the construction industry: the efficient reuse of coal gangue, a waste material from coal mining. Chaowei Zhang, a researcher at the School of Architecture and Civil Engineering and the Road Engineering Research Centre of Xi’an University of Science and Technology, has led a groundbreaking study that could revolutionize how we utilize industrial solid waste in construction materials.
Coal gangue, a byproduct of coal mining, has long been an environmental challenge due to its large-scale production and limited reuse. “The compressive strength of coal gangue mixtures is notoriously difficult to predict due to complex, nonlinear interactions among multiple composition factors,” Zhang explains. This unpredictability has hindered its efficient utilization in construction materials, leaving a significant gap in the market for sustainable and cost-effective solutions.
Zhang’s study, published in *Case Studies in Construction Materials* (translated as *典型案例研究* in Chinese), proposes a comprehensive ensemble machine learning (ML) framework to address this gap. The research team developed and compared 14 ML models based on 142 experimental datasets, with the top four models—CatBoost, GBDT, XGBoost, and ET—undergoing further optimization via a random search methodology.
The team employed SHapley Additive exPlanations (SHAP) analysis to interpret the models, revealing the underlying relationships between mixture proportions and compressive strength. This interpretability is crucial for practical applications, as it allows engineers to understand and trust the model’s predictions.
One of the most innovative aspects of this study is the establishment of an inverse analysis approach for the intelligent design of optimal mix proportions. “This approach enables us to not only predict the compressive strength of coal gangue mixtures but also to design the optimal mix proportions for desired performance,” Zhang says. This capability has significant commercial implications for the energy sector, as it can lead to more efficient and cost-effective use of coal gangue in construction materials.
The practical applicability of the inverse model was confirmed through field tests on a highway section. The deviations between predicted and measured 7-day compressive strengths were within 10%, demonstrating the model’s reliability and accuracy.
The study provides a reliable, interpretable, and data-driven solution for the performance prediction and mix design of coal gangue mixtures. This research contributes to the sustainable reuse of industrial solid waste and has the potential to shape future developments in the field. As the construction industry continues to seek sustainable and cost-effective solutions, the insights gained from this study could pave the way for innovative applications of industrial byproducts in construction materials.
The commercial impacts of this research are substantial. By enabling the efficient reuse of coal gangue, the study offers a viable solution for reducing waste and lowering costs in the construction industry. This, in turn, can contribute to the circular economy, where waste materials are repurposed, reducing the need for virgin resources and minimizing environmental impact.
As the energy sector continues to evolve, the demand for sustainable and innovative solutions will only grow. Zhang’s research provides a promising avenue for addressing these challenges, offering a glimpse into the future of construction materials and waste management. With further development and application, this technology could become a game-changer in the industry, driving progress towards a more sustainable and efficient future.