支持地下排水管道三维建模的声呐检测数据处理方法

宁锐

摘 要

地下排水管网系统的运营安全，关系到城市的环境质量以及民生安全。随着社会的快速发展，绝大多数排水管道处于高水位甚至满水运行状态，检测工作相对困难。一旦出现淤堵或者破裂等状况，将会影响城市道路交通安全，造成严重危害。传统的检测技术需要大量人工投入，存在效率低、工作量大、成本高等问题。因此，利用先进的检测技术收集管道的原始数据，通过数据处理技术和三维模型重塑技术使管道数据形成可视化三维模型，对于实现地下管网的安全运维具有重要意义。

本文首先分析城市地下排水管道的发展现状，研究声呐检测技术原理以及点云数据处理方法。其次，以武汉市某区域的排水管道为例，采用水下声呐检测设备获取管道原始数据，利用改进的滤波算法对点云数据进行去噪处理，结合聚类算法以及自适应蚁群优化算法提取管道的外形轮廓，拟合管道中心轴线以及淤积曲线函数。通过曲线插值放样算法建立管道三维模型以及淤积物模型，分析管道模型的精度，并计算管道内部淤积量。最后，根据功能需求搭建系统的总体架构，设计出城市排水管网综合管养平台，实现对地下排水管网系统的综合管理。

研究结果表明，利用本文改进的点云数据处理技术和三维模型重塑技术，对声呐数据进行有效去噪，建立出排水管道的三维模型。通过精度分析，计算出模型的体积值为15.27 m³，与实际管道体积误差为0.0367%。此外，建立出管道内部的淤积物模型，计算出淤积量为2.57 m³，约占管道体积16.83%。最后，将排水管道三维模型应用到城市排水管网综合管养平台中，对管道的模型信息进行有效管理，实现管道的智能诊断以及科学治理功能，为城市地下排水管网的科学管理提供决策支持。

关键词： 排水管道；滤波去噪；蚁群优化算法；三维重建；管养平台

Abstract

The operational safety of the underground drainage network system is related to the environmental quality of the city as well as the safety of people's livelihood. With the rapid development of society, most of the drainage pipes are in high water level or even full water operation, which makes the inspection work relatively difficult. Once there is blockage or rupture and other conditions, it will affect the safety of urban road traffic and cause serious harm. Traditional detection technology requires a lot of manual input, there are low efficiency, workload, high cost and other problems. Therefore, the use of advanced detection technology to collect the original data of the pipeline, through data processing technology and three-dimensional model reshaping technology to make the pipeline data to form a visual three-dimensional model, is of great significance to achieve the safe operation and maintenance of underground pipeline networks.

The thesis first analyzes the development status of urban underground drainage pipes, and studies the principle of sonar detection technology and point cloud data processing methods. Secondly, taking a drainage pipe in an area of Wuhan city as an example, we use underwater sonar detection equipment to obtain the original data of the pipe, use the improved filtering algorithm to denoise the point cloud data, combine the clustering algorithm and adaptive ant colony optimization algorithm to extract the shape contour of the pipe, fit the central axis of the pipe and the siltation curve function. The three-dimensional model of the pipeline and the siltation model are established by the curve interpolation and release algorithm, and the accuracy of the pipeline model is analyzed and the siltation volume inside the pipeline is calculated. Finally, the overall architecture of the system is built according to the functional requirements, and the integrated management and maintenance platform of urban drainage network is designed to realize the integrated management of the underground drainage network system.

The research results show that using the improved point cloud data processing technology and three-dimensional model reshaping technology in this thesis, the sonar data can be effectively denoised, and a three-dimensional model of the drainage pipeline can be established. Through the accuracy analysis, the calculated volume value of the model is 15.27 m³, and the error with the actual pipeline volume is 0.0367%. In addition, the sludge model inside the pipeline was established, and the calculated sludge volume was 2.57 m³, accounting for about 16.83% of the pipeline volume. Finally, the three-dimensional model of the drainage pipeline is applied to the comprehensive management platform of the urban drainage network to effectively manage the model information of the pipeline, realize the intelligent diagnosis and scientific management functions of the pipeline, and provide decision support for the scientific management of the urban underground drainage network.

Key words: Drainage pipes, Filtering and denoising, Ant colony optimization algorithm, Three-dimensional reconstruction, Management and maintenance platform