SDH-FCOS:An Efficient Neural Network for Defect Detection in Urban Underground Pipelines

Urban underground pipelines are an important infrastructure in cities,and timely investigation of problems in underground pipelines can help ensure the normal operation of cities.Owing to the growing demand for defect detection in urban underground pipelines,this study developed an improved defect detection method for urban underground pipelines based on fully convolutional one-stage object detector(FCOS),called spatial pyramid pooling-fast(SPPF)feature fusion and dual detection heads based on FCOS(SDH-FCOS)model.This study improved the feature fusion component of the model network based on FCOS,introduced an SPPF network structure behind the last output feature layer of the backbone network,fused the local and global features,added a top-down path to accelerate the circulation of shallowinformation,and enriched the semantic information acquired by shallow features.The ability of the model to detect objects with multiple morphologies was strengthened by introducing dual detection heads.The experimental results using an open dataset of underground pipes show that the proposed SDH-FCOS model can recognize underground pipe defects more accurately;the average accuracy was improved by 2.7% compared with the original FCOS model,reducing the leakage rate to a large extent and achieving real-time detection.Also,our model achieved a good trade-off between accuracy and speed compared with other mainstream methods.This proved the effectiveness of the proposed model.

Video-Based Deception Detection with Non-Contact Heart Rate Monitoring and Multi-Modal Feature Selection

Deception detection plays a crucial role in criminal investigation.Videos contain a wealth of information regarding apparent and physiological changes in individuals,and thus can serve as an effective means of deception detection.In this paper,we investigate video-based deception detection considering both apparent visual features such as eye gaze,head pose and facial action unit(AU),and non-contact heart rate detected by remote photoplethysmography(rPPG)technique.Multiple wrapper-based feature selection methods combined with the K-nearest neighbor(KNN)and support vector machine(SVM)classifiers are employed to screen the most effective features for deception detection.We evaluate the performance of the proposed method on both a self-collected physiological-assisted visual deception detection(PV3D)dataset and a public bag-oflies(BOL)dataset.Experimental results demonstrate that the SVM classifier with symbiotic organisms search(SOS)feature selection yields the best overall performance,with an area under the curve(AUC)of 83.27%and accuracy(ACC)of 83.33%for PV3D,and an AUC of 71.18%and ACC of 70.33%for BOL.This demonstrates the stability and effectiveness of the proposed method in video-based deception detection tasks.

A Novel Parameter-Optimized Recurrent Attention Network for Pipeline Leakage Detection

Accurate detection of pipeline leakage is essential to maintain the safety of pipeline transportation.Recently,deep learning(DL)has emerged as a promising tool for pipeline leakage detection(PLD).However,most existing DL methods have difficulty in achieving good performance in identifying leakage types due to the complex time dynamics of pipeline data.On the other hand,the initial parameter selection in the detection model is generally random,which may lead to unstable recognition performance.For this reason,a hybrid DL framework referred to as parameter-optimized recurrent attention network(PRAN)is presented in this paper to improve the accuracy of PLD.First,a parameter-optimized long short-term memory(LSTM)network is introduced to extract effective and robust features,which exploits a particle swarm optimization(PSO)algorithm with cross-entropy fitness function to search for globally optimal parameters.With this framework,the learning representation capability of the model is improved and the convergence rate is accelerated.Moreover,an anomaly-attention mechanism(AM)is proposed to discover class discriminative information by weighting the hidden states,which contributes to amplifying the normalabnormal distinguishable discrepancy,further improving the accuracy of PLD.After that,the proposed PRAN not only implements the adaptive optimization of network parameters,but also enlarges the contribution of normal-abnormal discrepancy,thereby overcoming the drawbacks of instability and poor generalization.Finally,the experimental results demonstrate the effectiveness and superiority of the proposed PRAN for PLD.

Detecting the backfill pipeline blockage and leakage through an LSTM-based deep learning model

Detecting a pipeline's abnormal status,which is typically a blockage and leakage accident,is important for the continuity and safety of mine backfill.The pipeline system for gravity-transport high-density backfill(GHB)is complex.Specifically designed,efficient,and accurate abnormal pipeline detection methods for GHB are rare.This work presents a long short-term memory-based deep learning(LSTM-DL)model for GHB pipeline blockage and leakage diagnosis.First,an industrial pipeline monitoring system was introduced using pressure and flow sensors.Second,blockage and leakage field experiments were designed to solve the problem of negative sample deficiency.The pipeline's statistical characteristics with different working statuses were analyzed to show their complexity.Third,the architecture of the LSTM-DL model was elaborated on and evaluated.Finally,the LSTM-DL model was compared with state-of-the-art(SOTA)learning algorithms.The results show that the backfilling cycle comprises multiple working phases and is intermittent.Although pressure and flow signals fluctuate stably in a normal cycle,their values are diverse in different cycles.Plugging causes a sudden change in interval signal features;leakage results in long variation duration and a wide fluctuation range.Among the SOTA models,the LSTM-DL model has the highest detection accuracy of98.31%for all states and the lowest misjudgment or false positive rate of 3.21%for blockage and leakage states.The proposed model can accurately recognize various pipeline statuses of complex GHB systems.

Current Trends and Perspectives of Detection and Location for Buried Non-Metallic Pipelines

Buried pipelines are an essential component of the urban infrastructure of modern cities.Traditional buried pipes are mainly made of metal materials.With the development of material science and technology in recent years,non-metallic pipes,such as plastic pipes,ceramic pipes,and concrete pipes,are increasingly taking the place of pipes made from metal in various pipeline networks such as water supply,drainage,heat,industry,oil,and gas.The location technologies for the location of the buried metal pipeline have become mature,but detection and location technologies for the non-metallic pipelines are still developing.In this paper,current trends and future perspectives of detection and location of buried non-metallic pipelines are summarized.Initially,this paper reviews and analyzes electromagnetic induction technologies,electromagnetic wave technologies,and other physics-based technologies.It then focuses on acoustic detection and location technologies,and finally introduces emerging technologies.Then the technical characteristics of each detection and location method have been compared,with their strengths and weaknesses identified.The current trends and future perspectives of each buried non-metallic pipeline detection and location technology have also been defined.Finally,some suggestions for the future development of buried non-metallic pipeline detection and location technologies are provided.

Quantitative risk assessment & leak detection criteria for a subsea oil export pipeline

A quantitative risk assessment （QRA） based on leak detection criteria （LDC） for the design of a proposed subsea oil export pipeline is presented in this paper. The objective of this QRA/LDC study was to determine if current leak detection methodologies were sufficient, based on QRA results, while excluding the use of statistical leak detection; if not, an appropriate LDC for the leak detection system would need to be established. The famous UK PARLOC database was used for the calculation of pipeline failure rates, and the software POSVCM from MMS was used for oil spill simulations. QRA results revealed that the installation of a statistically based leak detection system （LDS） can significantly reduce time to leak detection, thereby mitigating the consequences of leakage. A sound LDC has been defined based on QRA study results and comments from various LDS vendors to assist the emergency response team （ERT） to quickly identify and locate leakage and employ the most effective measures to contain damage.

Pipeline Potential Leak Detection Technologies: Assessment and Perspective in the Nigeria Niger Delta Region

This paper examines the advances in pipeline third party encroachment alert systems and leak control methods in the oil/gas industry. It also highlights the extent of spill/pollution issues in the Niger Delta region due to intended/unin- tended damages and suggests a possible method of control. It is believed that the best option to avoid pollution due to pipeline failure is to ensure that hydrocarbon does not exit from the pipeline. With the different methods considered in this review, acoustic monitoring of change in the operational sound generated from a given pipeline section is suggested to be practicable to identifying sound abnormalities of third party encroachments. One established challenge of the acoustic system for buried pipelines protection is attenuation of acoustic transmission. An attempt to check the performance of an acoustic transmission on steel pipelines submerged in water points to a similar research on plastic water pipelines that attenuation is small compared with pipe buried in soil. Fortunately, Niger Delta of Nigeria is made of wetland, swamps and shallow water and could therefore offer an opportunity to deploy acoustic system for the safety of pipelines against third party attacks in this region. However, the numerous configuration and quantity of oil installation in this region imply that cost of application will be enormous. It is therefore suggested that a combination of impressed alternating cycle current (IACC) which traces encroachment on the pipeline coating and an acoustic system be used to manage intended and unintended pipeline potential damages. The IACC should be used for flow lines and other short distance delivery lines within the oilfield, while the relatively large diameter and long length delivery, trunk and transmission lines should be considered for acoustic protection. It is, however, noted that further efforts are required to reduce cost and improve effectiveness of these systems.

In-Line Inspection of Pipeline Defects Detection Using Ring-Type Laser

Pipeline plays a vital role in transporting fluids like oils, water, and petrochemical substances for longer distances. Based on the materials they carry, prolonged usage may cause the initiation of defects in the pipeline. These defects occur due to the formed salt deposits, chemical reaction happens between the inner surface and the transferring substance, prevailing environmental conditions, etc. These defects, if not identified earlier may lead to significant losses to the industry. In this work, an in-line inspection system utilizes the nondestructive way for analyzing the internal defects in the petrochemical pipeline. This system consists of a pipeline inspection robot having two major units namely the visual inspection unit and the power carrier unit. The visual inspection unit makes use of a ring-type laser diode and the camera. The laser diode serves as a light source for capturing good quality images of inspection. This unit is controlled by the Arduino in the power carrier unit which provides the necessary movement throughout the pipe. The inspected images captured by the camera are further processed with the aid of NI vision assistant software. After applying the processing function parameters provided by this software, the defect location can be clearly visualized with high precision. Three sets of defects are introduced in a Polylactide (PLA) pipe based on its position and angle along the circumference of the pipe. Further, this robot system serves as a real-time interactive image synchronization system for acquiring the inspected images. By comparing the actual and calculated defect size, the error percentage obtained was less than 5%.

Design of motion control system of pipeline detection AUV

A great number of pipelines in China are in unsatisfactory condition and faced with problems of corrosion and cracking,but there are very few approaches for underwater pipeline detection.Pipeline detection autonomous underwater vehicle(PDAUV) is hereby designed to solve these problems when working with advanced optical,acoustical and electrical sensors for underwater pipeline detection.PDAUV is a test bed that not only examines the logical rationality of the program,effectiveness of the hardware architecture,accuracy of the software interface protocol as well as the reliability and stability of the control system but also verifies the effectiveness of the control system in tank experiments and sea trials.The motion control system of PDAUV,including both the hardware and software architectures,is introduced in this work.The software module and information flow of the motion control system of PDAUV and a novel neural network-based control(NNC) are also covered.Besides,a real-time identification method based on neural network is used to realize system identification.The tank experiments and sea trials are carried out to verify the feasibility and capability of PDAUV control system to complete underwater pipeline detection task.

Research on Oil Pipeline Leak Detection and Location system

Because pipeline has large pipe diameter, large throughout and high pressure, once pipeline leakage accident happens, the damage is quite serious. In addition, pipeline leakage accident caused by man-made drilling oil stolen every year results in huge economic losses on oilfield. Therefore, a real-time and accurate pipeline leak detection and location system not only can effectively decrease leakage loss and reduce the waste of manpower and material resources in patrolling work, but also is conductive to the management of oil pipeline and improvement of economic efficiency of enterprise. The paper determines leak detection and location project giving priority to negative pressure wave and supplemented by flow parameter analysis. The method not only can judge the accidence of leakage timely and accurately, but also can effectively avoid leakage false alarm caused by start or stop pumps in pipeline.

Using acoustic technique to detect leakage in city gas pipelines

For solving the difficult problem of leakage detection in city gas pipelines, a method using acoustic technique based on instantaneous energy (IE) distribution and correlation analysis was proposed. Firstly, the basic theory of leakage detection and location was introduced. Then the physical relationship between instantaneous energy and structural state variation of a system was analyzed theoretically. With HILBERT-HUANG transformation (HHT), the instantaneous energy distribution feature of an unstable acoustic signal was obtained. According to the relative contribution method of the instantaneous energy, the noise in signal was eliminated effectively. Furthermore, in order to judge the leakage, the typical characteristic of the instantaneous energy of signal in the input and output end was discussed using correlative analysis. A number of experiments were carried out to classify the leakage from normal operations, and the results show that the leakages are successfully detected and the average recognition rate reaches 93.3% among three group samples. It is shown that the method using acoustic technique with IED and correlative analysis is effective and it may be referred in other pipelines.

一种新的用于点目标检测与跟踪的闭环pipeline算法

提出了一种由滤波管道、检测管道构成的闭环管道算法用于检测与跟踪复杂云天背景下的点目标.引入了基于单帧局部对比度门限的预处理算法进行杂波抑制,设计了基于三帧图像相关性分析的连续滤波器进行噪声剔除.实验分析表明,该算法不仅能够检测出作任意轨迹运动的点目标,而且能够有效地预测跟踪窗.它在搜索空间的缩小、计算复杂性的降低以及杂波剔除方面优于传统的管道算法.

Vision-Based System of AUV for An Underwater Pipeline Tracker

This paper describes a new framework for detection and tracking of underwater pipeline, which includes software system and hardware system. It is designed for vision system of AUV based on monocular CCD camera. First, the real-time data flow from image capture card is pre-processed and pipeline features are extracted for navigation. The region saturation degree is advanced to remove false edge point group after Sobel operation. An appropriate way is proposed to clear the disturbance around the peak point in the process of Hough transform. Second, the continuity of pipeline layout is taken into account to improve the efficiency of line extraction. Once the line information has lbeen obtained, the reference zone is predicted by Kalman filter. It denotes the possible appearance position of the pipeiine in the image. Kalman filter is used to estimate this position in next frame so that the information of pipeline of each frame can be known in advance. Results obtained on real optic vision data in tank experiment are displayed and discussed. They show that the proposed system can detect and track the underwater pipeline online, and is effective and feasible.

DISTRIBUTED OPTICAL FIBER SENSOR FOR LONG-DISTANCE OIL PIPELINE HEALTH

A fully distributed optical fiber sensor （DOFS） for monitoring long-distance oil pipeline health is proposed based on optical time domain reflectometry （OTDR）. A smart and sensitive optical fiber cable is installed along the pipeline acting as a sensor, The experiments show that the cable swells when exposed to oil and induced additional bending losses inside the fiber, and the optical attenuation of the fiber coated by a thin skin with periodical hardness is sensitive to deformation and vibration caused by oil leakage, tampering, or mechanical impact. The region where the additional attenuation occurred is detected and located by DOFS based on OTDR, the types of pipeline accidents are identified according to the characteristics of transmitted optical power received by an optical power meter, Another prototype of DOFS based on a forward traveling frequency-modulated continuous-wave （FMCW） is also proposed to monitor pipeline. The advantages and disadvantages of DOFSs based on OTDR and FMCW are discussed. The experiments show that DOFSs are capable of detecting and locating distant oil pipeline leakages and damages in real time with an estimated precision of ten meters over tens of kilometers.

Oil/Gas Pipeline Leak Inspection and Repair in Underwater Poor Visibility Conditions: Challenges and Perspectives

Mechanical pressure clamps are examples of innovative tools commonly used in the oil and gas industry for arresting leaks from damaged oil and gas pipelines. However, if leaks result from pipeline rupture, clamps are not usually recommended. It is therefore obvious that inspection of the leaking pipeline is very crucial in deciding the strategy for repair. For subsea pipelines where underwater poor visibility is pronounced, this important aspect of the pipeline repair process becomes difficult to implement. The result is a repair-leak-repair cycle. This challenge is commonly found in repairs of old pipelines in unclear water conditions. Old pipelines and their vulnerability to fractures that often lead to ruptures are discussed. In this paper, the challenges and technologies available for visualisation and examination in such unclear water conditions are discussed. There appears to be a gap in the existing pipeline integrity management system with respect to inspection and repair of pipelines in unclear water conditions. This gap needs to be filled in order to minimise spills and pollution. For pipelines installed in unclear water condition, a perspective is suggested to extend the capability of existing remotely operated vehicles to employ the use of clear laminar water system or a related technique to provide integrity engineers and operators with close visual assess to inspect leaking pipelines and effect adequate repairs. This paper suggests that the use of optical eye as the main tool for examination remains valuable in managing the challenges in underwater pipeline repairs in unclear water condition.

Composite excitation multi-extension direction defect magnetic flux leakage detection technology

In the traditional pipeline magnetic flux leakage(MFL)detection technology,circumferential or axial excitation is mainly used to excite the magnetic field of defects.However,the domestic and foreign pipeline detection devices currently in operation are mainly axial excitation MFL detection tools,in which circumferential cracks can be clearly identified,but the detection sensitivity of axial cracks is not high,thus forming a detection blind zone.Therefore,a composite excitation multi-extension direction defect MFL detection method is proposed,which can realize the simultaneous detection of axial and circumferential defects.On the basis of the electromagnetic theory Maxwell equation and Biot Savart law,a mathematical model of circumferential and axial magnetization is firstly established.Then finite element simulation software is used to establish a model of a new type of magnetic flux leakage detection device,and a simulation analysis of crack detection in multiple extension directions is carried out.Finally,under the conditions of the relationship model between the change rate of leakage magnetic field and external excitation intensity under unsaturated magnetization and the multi-stage coil magnetization model,the sample vehicle towing experiment is carried out.The paper aims to analyze the feasibility and effectiveness of the new magnetic flux leakage detection device for detecting defects in different extension directions.Based on the final experimental results,the new composite excitation multi extension direction leakage magnetic field detector has a good detection effect for defects in the axial and circumferential extension directions.

Characteristics of Vibrational Wave Propagation and Attenuation in Submarine Fluid-Filled Pipelines

As an important part of lifeline engineering in the development and utilization of marine resources, the submarine fluid-filled pipeline is a complex coupling system which is subjected to both internal and external flow fields. By utilizing Kennard＇s shell equations and combining with Helmholtz equations of flow field, the coupling equations of submarine fluid-filled pipeline for n=0 axisymmetrical wave motion are set up. Analytical expressions of wave speed are obtained for both s=1 and s=2 waves, which correspond to a fluid-dominated wave and an axial shell wave, respectively. The numerical results for wave speed and wave attenuation are obtained and discussed subsequently. It shows that the frequency depends on phase velocity, and the attenuation of this mode depends strongly on material parameters of the pipe and the internal and the external fluid fields. The characteristics of PVC pipe are studied for a comparison. The effects of shell thickness/radius ratio and density of the contained fluid on the model are also discussed. The study provides a theoretical basis and helps to accurately predict the situation of submarine pipelines, which also has practical application prospect in the field of pipeline leakage detection.

基于45 nm SOI CMOS工艺的10 bit、125 MS/s过零检测Pipeline-SAR ADC设计

基于45nm SOI CMOS工艺,设计了一款两级流水线级联型逐次逼近ADC(Pipeline-SAR ADC).摒弃了传统流水线结构中大功耗级间运算放大器,采用过零比较器和受控电流源完成级间余量放大功能,极大地减小了ADC的功耗.分析了子ADC中比较器失调对ADC精度的影响,提出了一种具有失调校准的动态比较器,满足了高精度、高速度的要求.此外,在设计逐次逼近结构时,采用共模切换、上极板采样和全定制控制逻辑等技术进一步降低了系统功耗.仿真结果显示,ADC在125 MS/s、奈圭斯特输入频率下,实现了60.46dB的信噪失真比和77.33dB的无杂散动态范围,有效位数为9.75bit,系统总功耗只有1mW.ADC的FoM值仅为9.29fJ/step,较其他工作有很大的提升.

Comparative Study on Perimeter Intrusion Detection System of High-speed Railway

The perimeter intrusion detection system is critical to China’s railway safety.An efficient intrusion detection system can effectively avoid human casualties and property damage.This article makes a comprehensive comparison of popular detection systems in recent years.It first outlines the characteristics and classification of intrusion detection systems,and then introducestherelevantliteratureofcontactandnon-contactsystemsaccordingtodifferenttypes,andalsointroducesthe principles and architecture of the models they use in detail.Finally,the detection performance and suitable environment under different system models are analyzed by comparison.

Performing Reliable Early Leak Detection on Storage Tank Terminals:Retrofitting of Existing Facilities and New Buildings

Liquid leak detection may represent a challenge for Oil&Gas operators,as indicated by operational feed-back and independent studies.Despite the availability of many different leak detection technologies,some systems may either fail to detect spills or generate frequent false alarms.In particular,possible soil contamination from pre-existing leaks and pollution carry-over by rain water is difficult to filter out by a leak sensing system.Typical case of false alarms relates to punctual sensors installed upstream the drain valve within the storage tank bunds,monitoring possible presence of leaks in rain water.Besides old soil contamination,other criteria should also be considered when selecting a spill detection technology,such as asset type to be monitored(storage tank,pipeline,…),system accuracy(minimum detectable quantity,ability to localize the leak),detection time,reliability over time,capital,installation and operating costs.The paper will include an evaluation of different external leak detection technologies with respect to the above-mentioned criteria,pointing out the capabilities and limitations of each system.Focus will be placed on reliability of leak monitoring systems in challenging environments.A new generation of digital,reusable sensing cables and probes,as well as the impact of sensitivity for different applications,will be discussed.Since leak sensor installation environment(positioning,adoption of special precautions,…)may significantly affect the system performance,different above ground and underground configurations will be presented,both for new builds and existing facilities.