3D laser point cloud-based geometric digital twin for condition assessment of large diameter pipelines

Abstract

The increasing aging of underground pipe networks and the lack of effective inspection technologies present considerable challenges for whole life-cycle management of these infrastructures. Modern laser scanning technology offers a cost-effective and safe means to obtain dense and accurate 3D topographic data of the inner surface of pipelines. However, laser scanning point clouds contain substantial noise and outliers, and efficiently extracting valuable information for structural and functional mapping remains in its infancy. This paper presents an innovative method for fast processing point clouds data of large-diameter pipelines, enabling the accurate extraction of geometric features and efficient establishment of geometric digital twin using density-based clustering, fitting and region growing algorithm. Experimental tests were conducted to evaluate the accuracy, efficiency, and feasibility of the proposed method. The results demonstrate that the proposed approach not only robustly achieves high accuracy but also maintains high computational efficiency. Additionally, the geometric digital twin shows promise as tools for quantitatively assessing structural deformation and blockage defects.