Research on 3D geological modeling method for UHV tower foundation based on machine learning

For a long time, the process of three-dimensional geological modeling has faced issues such as limited borehole data and non- smooth modeling results. This study proposes a machine learning-based approach for three-dimensional geological modeling, specifically applied to the foundations of ultra-high voltage transmission towers. The method leverages borehole data and a support vector machine model to predict subsurface stratigraphy, addressing limitations in traditional explicit modeling reliant on sparse, discrete data. By employing feature resampling, coordinate standardization, and label conversion, the model enhances compatibility with machine learning algorithms, achieving improved convergence and predictive accuracy. Experimental results demonstrate the model's effectiveness in complex geological environments. The 3D visualization using Unity3D aids in understanding stratigraphic distribution, providing an intuitive tool for engineering decision -making.