Accurate 3D models of underground cable wells play an important role in the operation and management of cities; however, the current modeling of industrial wells suffers from the problems of low modeling efficiency and incomplete 3D models. In this paper, we propose a method for modeling underground industrial wells based on the automated extraction of skeleton line and contour line features from 3D laser point cloud data. In the method, the external point cloud inside a well chamber is first separated and the skeleton points of a cable line are then extracted using the L1 median skeleton extraction algorithm with the maximum tangent sphere for the internal cable point cloud. For problems where the α value in the extraction of the profile line based on the conventional alpha shape algorithm is difficult to estimate, we propose a method that combines the Delaunay triangularization and the alpha shape algorithm for extracting the profile line features of the bottom surface of the well chamber. Boundary condition checking and curvature complexity analysis are proposed to adaptively obtain the α values of different shapes of industrial wells; finally, the cables and well chambers are modeled according to the acquired skeleton points and contour lines, respectively. In the experiments, the L1 median skeleton extraction algorithm is used to extract cable skeleton points for four types of cable with different levels of point cloud completeness, different levels of sparseness, different cable bending degrees, and the existence of missing point clouds, which are prominent in cable bending and turning areas. Boundary condition checking and curvature complexity analysis are carried out to calculate four typical cable wells, and the proposed α value is obtained to extract the bottom surface contour line features. The height parameter of the underground cable wells is combined with the acquired contour lines to reconstruct the underground cable wells in three dimensions. The underground cable well model is highly consistent with the real object in terms of geometry and dimensions and performs well in generating details of the cables. Underground cable well modeling is crucial to improving urban infrastructure management, ensuring safety, optimizing maintenance, supporting emergency response, and improving decision-making efficiency, which helps achieve smart city development.