Urban Geological 3D Modeling Based on Papery Borehole Log

Abstract

Borehole log is important data for urban geological 3D modeling. Most of the current borehole logs are stored in a papery form. The construction of a smart city puts forward requirements for the automatic and intelligent 3D modeling of urban geology. However, it is difficult to extract the information from the papery borehole log quickly. What is more, it is unreliable to rely entirely on automated algorithms for modeling without artificial participation, but there is no effective way to integrate geological knowledge into 3D geological modeling currently. Therefore, it is necessary to research how to use existing papery borehole logs efficiently. To overcome the above obstacles, we designed a method that combines structural analysis and layout understanding to extract information from the borehole log. Then, the knowledge-driven three-dimensional geological modeling is proposed based on dynamic profiles. With these methods, the papery borehole log can be converted into structured data which can be used for data analysis directly, and geological knowledge can be integrated into the process of 3D geological modeling. The 3D geological modeling of Xinyang City based on a papery borehole log has been taken as an example to verify the feasibility of the method.

Method and Application of Urban 3D Rapid Modeling of Geology Based on CAD Borehole Logs

“…Liu et al designed strata division methods based on a convolutional neural network, but the neural network retains incorrect training weight during the classification process, which leads to inaccurate results when the sample size is small [27]. Zhang et al (2020) proposed a process of prestoring geological knowledge in a geological data corpus and then classifying the process, but the specific recognition rate of geological symbols is minimal. The above method solves the problem of standardizing stratum to a certain extent but still has defects of requiring merging and generalization, such as complex lithology and stratigraphic cross repetition.…”

Application of Multilayer Perceptron Neural Network in Geological Modeling of Categorical Variables: A Case Study in Peru

“…The construction of three-dimensional (3D) geologic models is essential for the management, visual representation, amalgamation, and deduction of findings from geologic studies [6][7][8][9]. Its core lies in the generation of 3D representations from geological measurements and logs, with the aim of understanding the spatial arrangement, depositional connections, and nature of the stony subsurface [10][11][12][13].…”

Application of Machine Learning and 3D Modelling to Improve Geotechnical Site Characterization

“…Although 1D representation gives accurate geotechnical investigations at the sampled pit/borehole locations, it does not show proper indication of subsurface conditions in between sampled pits/boreholes. Adequate subsurface characterization of a construction site is better 2 represented with a 3D than 1D viewpoint (Caballero et al, 2022;Zhang et al, 2020). 3D modelling gives more effective visualization and comprehension of the relationship between vertical and horizontal continuum of subsurface conditions than in 1D or 2D.…”