当前位置: X-MOL 学术Appl. Ocean Res. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Subsea pipeline leak inspection by autonomous underwater vehicle
Applied Ocean Research ( IF 4.3 ) Pub Date : 2020-12-29 , DOI: 10.1016/j.apor.2020.102321
Hongwei Zhang , Shitong Zhang , Yanhui Wang , Yuhong Liu , Yanan Yang , Tian Zhou , Hongyu Bian

Faced with the difficulty in finding and locating leakage points of submarine pipelines timely as well as the high cost of pipeline routine inspection, a scheme using autonomous underwater vehicle (AUV) equipped with multibeam echo sounder (MBES) and forward looking sonar (FLS) for automatic inspection of submarine pipelines was developed. In the process of inspection, the AUV maintains autonomous navigation along the pre-set pipeline route at a fixed height above the pipeline, while the MBES collects water column images. After extracting the outline characteristics of gas-filled bubbles and making a leakage risk judgment, it rises to the sea surface to issue an alarm to the shore-based command center via satellite. The results of sea trial verify the effectiveness of sub-sea pipeline leak detection and real-time obstacle avoidance of the vehicle proposed in this study. Additionally, a variable buoyancy system (VBS) is adopted, which enhances the navigation efficiency of the vehicle. On the other hand, considering the complex operation environment of the pipeline inspection, the online collision avoidance is indispensable. The traditional image segmentation algorithm is not ideal for the high noise and bottom reverberation in operation sea area. Based on the traditional algorithm, an improved Otsu algorithm is proposed to improve the operation speed and denoise effect. As a result, an improved Otsu algorithm is proposed to accurately identify obstacles. Moreover, Kalman filtering is also introduced to estimate dynamic obstacles.



中文翻译:

自主水下航行器检查海底管道泄漏

面对及时发现和定位海底管道泄漏点的困难以及管道常规检查的高昂成本,使用配备有多波束回声测深仪(MBES)和前向声纳(FLS)的自动水下航行器(AUV)的方案开发了海底管道自动检查系统。在检查过程中,AUV会沿着预设的管道路线在管道上方的固定高度上保持自主导航,而MBES会收集水柱图像。在提取出气泡的轮廓特征并做出泄漏风险判断后,它上升到海面,并通过卫星向岸基指挥中心发出警报。海上试验的结果验证了本研究中提出的海底管道泄漏检测和车辆实时避障的有效性。此外,采用了可变浮力系统(VBS),可提高车辆的导航效率。另一方面,考虑到管道检查的复杂操作环境,避免在线碰撞是必不可少的。传统的图像分割算法对于操作海域的高噪声和海底混响不理想。在传统算法的基础上,提出了一种改进的Otsu算法,以提高运算速度和降噪效果。结果,提出了一种改进的Otsu算法以准确地识别障碍物。此外,还引入了卡尔曼滤波来估计动态障碍。此外,采用了可变浮力系统(VBS),可提高车辆的导航效率。另一方面,考虑到管道检查的复杂操作环境,避免在线碰撞是必不可少的。传统的图像分割算法对于操作海域的高噪声和海底混响不理想。在传统算法的基础上,提出了一种改进的Otsu算法,以提高运算速度和降噪效果。结果,提出了一种改进的Otsu算法以准确地识别障碍物。此外,还引入了卡尔曼滤波来估计动态障碍。此外,采用了可变浮力系统(VBS),可提高车辆的导航效率。另一方面,考虑到管道检查的复杂操作环境,避免在线碰撞是必不可少的。传统的图像分割算法对于操作海域的高噪声和海底混响不理想。在传统算法的基础上,提出了一种改进的Otsu算法,以提高运算速度和降噪效果。结果,提出了一种改进的Otsu算法以准确地识别障碍物。此外,还引入了卡尔曼滤波来估计动态障碍。考虑到管道检查的复杂操作环境,避免在线碰撞是必不可少的。传统的图像分割算法对于操作海域的高噪声和海底混响不理想。在传统算法的基础上,提出了一种改进的Otsu算法,以提高运算速度和降噪效果。结果,提出了一种改进的Otsu算法以准确地识别障碍物。此外,还引入了卡尔曼滤波来估计动态障碍。考虑到管道检查的复杂操作环境,避免在线碰撞是必不可少的。传统的图像分割算法对于操作海域的高噪声和海底混响不理想。在传统算法的基础上,提出了一种改进的Otsu算法,以提高运算速度和降噪效果。结果,提出了一种改进的Otsu算法以准确地识别障碍物。此外,还引入了卡尔曼滤波来估计动态障碍。结果,提出了一种改进的Otsu算法以准确地识别障碍物。此外,还引入了卡尔曼滤波来估计动态障碍。结果,提出了一种改进的Otsu算法以准确地识别障碍物。此外,还引入了卡尔曼滤波来估计动态障碍。

更新日期:2020-12-29
down
wechat
bug