In the ever-evolving world of civil engineering, predicting the bearing capacity of driven piles—a critical component in deep foundations—has long been a complex challenge. Traditional methods rely heavily on physical assumptions and mathematical models, often leaving room for improvement in accuracy and efficiency. However, a groundbreaking study led by Yan Peng from Anhui Technical College of Industry and Economy in Hefei, China, is set to revolutionize this process.
Peng and his team have developed innovative hybrid adaptive neuro-fuzzy inference systems (ANFIS) optimized with three different algorithms: artificial rabbit optimization (ARO), cuckoo optimization algorithm (COA), and grey wolf optimization (GWO). These systems use experimental data to calculate the bearing capacity of driven piles (Qt) with remarkable precision. “The hybrid ANFIS models we’ve developed show a strong correlation between experimental and estimated Qt values,” Peng explains. “This means our models can provide highly accurate predictions, which is crucial for ensuring the safety and stability of deep foundations.”
The study, published in the *Journal of Applied Science and Engineering* (translated from Chinese as *Journal of Applied Sciences and Engineering*), also compared the performance of these hybrid models with other algorithms such as single ANFIS, support vector regression (SVR), M5P, multi-adaptive regression spline (MARS), random forests (RF), and random trees (RT). The results were impressive, with the ANFIS systems optimized with ARO, GWO, and COA achieving a minimum R^2 of 0.9285 in the learning dataset and 0.9313 in the examining dataset. “The ARO-ANFIS model, in particular, outperformed the others, demonstrating the highest performance in terms of correlation metrics and the lowest values of error-based metrics,” Peng notes.
The implications of this research for the energy sector are significant. Accurate prediction of bearing capacity is essential for the design and construction of deep foundations for various energy infrastructure projects, including wind farms, oil and gas platforms, and power plants. “By improving the accuracy of these predictions, we can enhance the safety and reliability of these structures, ultimately reducing the risk of failures and the associated costs,” Peng says.
The study also highlights the potential for these advanced modeling techniques to be applied in other areas of civil engineering and beyond. “The hybrid ANFIS models we’ve developed are not only applicable to driven piles but could also be adapted for use in other types of foundations and structural elements,” Peng explains. “This opens up new possibilities for improving the design and construction of a wide range of infrastructure projects.”
As the energy sector continues to evolve, the need for accurate and efficient modeling techniques will only grow. The research led by Yan Peng represents a significant step forward in this field, offering a powerful tool for engineers and researchers to enhance the safety, reliability, and cost-effectiveness of deep foundation projects. “We believe that our work will have a lasting impact on the industry and pave the way for future developments in this area,” Peng concludes.