Pipe Inspection Research Articles

A Rapid and Accurate Technique With Updating Strategy for Surface Defect Inspection of Pipelines

Defect inspection in pipes at the early stage is of crucial importance to maintain the ongoing safety and suitability of the equipment before it presents an unacceptable risk. Due to the nature of detection methods being costly or complex, the efficiency and accuracy of results obtained hardly meet the requirements from industries. To explore a rapid and accurate technique for surface defects detection, a novel approach QDFT (Quantitative Detection of Fourier Transform) has been recently proposed by authors to efficiently reconstruct defects. However, the accuracy of this approach needs to be further improved. In this paper, a modified QDFT method with integration of an integral coefficient updating strategy, called QDFTU (quantitative detection of Fourier transform of updating), is developed to reconstruct the defect profile with a high level of accuracy throughout iterative calculations of integral coefficients from the reference model updated by a termination criteria (RMSE, root mean square error). Moreover, dispersion equations of circumferential guided waves in pipes are derived in the helical coordinate to accommodate the stress and displacement calculations in the scattered field using hybrid FEM. To demonstrate the superiority of the developed QDFTU in terms of accuracy and efficiency, four types of defect profiles, i.e., a rectangular flaw, a multi-step flaw, a double-rectangular flaw, and a triple-rectangular flaw, are examined. Results show the fast convergence of QDFTU can be identified by no more than three updates for each case and its high accuracy is observed by a smallest difference between the predicted defect profile and the real one in terms of mean absolute percentage error (MSPE) value, which is 6.69% in the rectangular-flaw detection example.

Excitation and Propagation of Longitudinal L (0, 2) Mode Ultrasonic Guided Waves for the Detection of Damages in Hexagonal Pipes: Numerical and Experimental Studies

The hexagonal pipe is a special kind of tube structure. Its inner surface of the cross section is in the shape of circle, while the outer surface is hexagonal. It has functioned as an essential and critical part of a drill stem in a high‐torque drill machine used in various resource exploitation fields. The inspection of a hexagonal pipe to avoid its failure and thus to ensure safe operation of a drilling machine is becoming increasingly urgent and important. In this study, the excitation and propagation of ultrasonic guided waves for the purpose of detecting defects in hexagonal pipes are proposed. Dispersion curves of hexagonal pipes are firstly derived by using semianalytical finite element method. Based on these dispersion curves, longitudinal L (0, 2) mode at 100 kHz is selected to inspect hexagonal pipes. A ring of piezoelectric transducers (PZTs) with the size of 25 mm × 5 mm ×0.5 mm is able to maximize the amplitude of L (0, 2) mode and successfully suppress the undesired L (0, 1) mode in the experiments. Numerical and experimental studies have shown that the displacement field of L (0, 2) mode at 100 kHz is almost uniformly distributed along the circumferential direction. Furthermore, L (0, 2) mode ultrasonic guided waves at 100 kHz are capable of detecting circular through‐hole damages located in the plane and near the edge in a hexagonal pipe. Our study results have demonstrated that the use of longitudinal L (0, 2) mode ultrasonic guided wave provides a promising and effective alternative for the detection of defects in hexagonal pipe structures.

Development of an Inline Pipe Inspection Robot for the Oil and Gas Industry

Development of an Inline Pipe Inspection Robot for the Oil and Gas Industry

Studies of the traversing mechanism for pipe inspection using mecanum wheel

Studies of the traversing mechanism for pipe inspection using mecanum wheel

In-Pipe Inspection Robot Capable of Actively Exerting Propulsive and Tractive Forces With Linear Antagonistic Mechanism

In this study, the authors comprehensively evaluated the traveling performance of a flexible pipe inspection robot (PI-RO) that can actively output both propulsion and traction forces. Household pipes must be frequently inspected for cracks and corrosion to ensure continuous provision of water and gas. As these pipes are essential for our everyday life, it is difficult to suspend their usage for long periods; hence, inspection should be conducted for short durations. In addition, inspecting entire pipelines using existing methods is challenging because the pipes are constructed over long distances and have complex structures. Therefore, we developed a flexible robot that can actively output propulsion and traction forces to inspect long-distance complex pipelines, such as household pipelines. Herein, we describe the flexible structural design of a PI-RO that inspects thin one-inch pipes (inner diameter of approximately 28 mm) typically present in household piping. We also evaluated the characteristics of each PI-RO unit. Finally, we confirmed that the PI-RO is not significantly affected by the traveling environment, such as long distances and complex small-diameter pipes, by conducting traveling experiments on small pipes (inner diameter 28 mm) and long-distance (up to approximately 11.7 m) complex pipes composed of horizontal, vertical, and 90° bend pipes.

Wave Propagation Type Pipe Inspection Robot with the Linear Antagonistic Mechanism for Enhancing Traction Force

Wave Propagation Type Pipe Inspection Robot with the Linear Antagonistic Mechanism for Enhancing Traction Force

Time Series Analyses and Modeling of Environmental Systems: Case Studies in Kuwait

Background: Time series is a sequence of measurements made at regular time intervals. Tremendous work has been done on pipe break investigation and maintenance, using different models. However, the authors have not found any research work on pipe failures in water distribution networks using time series. Objective: The principal objective of this paper was to investigate the applications of time series analysis in modeling man-made (pipe breaks) and natural (rainfall) environmental systems. Methods: This objective was satisfied by employing the Time-Domain (Box-Jenkins) approach to assess two case studies in Kuwait of a water distribution network (pipe failures), and from rainfall rates measurement of a solid waste disposal site, respectively. The statistical analysis of these data was conducted with the aid of the BMDP and SAS computer software packages and by the use of the P2T programming procedure for the Box-Jenkins method. Results: The results showed a significant correlation between the data and the time series models provides good estimates of pipe reliability and rainfall measurements. Conclusion: Based on the significant results, time series analyses could and should be useful by water authorities in planning pipe maintenance and inspection. Moreover, it is recommended and it would be very useful to conduct similar studies on pipes carrying other fluids such as gas, air and oil.

Deep Learning for Ultrasonic Crack Characterization in NDE

Machine learning for nondestructive evaluation (NDE) has the potential to bring significant improvements in defect characterization accuracy due to its effectiveness in pattern recognition problems. However, the application of modern machine learning methods to NDE has been obstructed by the scarcity of real defect data to train on. This article demonstrates how an efficient, hybrid finite element (FE) and ray-based simulation can be used to train a convolutional neural network (CNN) to characterize real defects. To demonstrate this methodology, an inline pipe inspection application is considered. This uses four plane wave images from two arrays and is applied to the characterization of cracks of length 1-5 mm and inclined at angles of up to 20° from the vertical. A standard image-based sizing technique, the 6-dB drop method, is used as a comparison point. For the 6-dB drop method, the average absolute error in length and angle prediction is ±1.1 mm and ±8.6°, respectively, while the CNN is almost four times more accurate at ±0.29 mm and ±2.9°. To demonstrate the adaptability of the deep learning approach, an error in sound speed estimation is included in the training and test set. With a maximum error of 10% in shear and longitudinal sound speed, the 6-dB drop method has an average error of ±1.5 mmm and ±12°, while the CNN has ±0.45 mm and ±3.0°. This demonstrates far superior crack characterization accuracy by using deep learning rather than traditional image-based sizing.

Water Level Estimation in Sewer Pipes Using Deep Convolutional Neural Networks

Sewer pipe inspections are currently conducted by professionals who remotely control a robot from above ground. This expensive and slow approach is prone to human mistakes. Therefore, there is both an economic and scientific interest in automating the inspection process by creating systems able to recognize sewer defects. However, the extent of research put into automatic water level estimation in sewers has been limited despite being a prerequisite for further analysis of the pipe as only sections above the water level can be visually inspected. In this work, we utilize a dataset of still images obtained from over 5000 inspections carried out for three different Danish water utilities companies. This dataset is used for training and testing decision tree methods and convolutional neural networks (CNNs) for automatic water level estimation. We pose the estimation problem as a classification and regression problem, and compare the results of both approaches. Furthermore, we compare the effect of using different inspection standards for labeling the ground truth water level. By treating the problem as a classification task and using the 2015 Danish sewer inspection standard, where water levels are clustered based on visual appearance, we achieve an averaged F1 score of 79.29% using a fine-tuned ResNet-50 CNN. This shows the potential of using CNNs for water level estimation. We believe including temporal and contextual information will improve the results further.

Automated sewer pipe defect tracking in CCTV videos based on defect detection and metric learning

Computer vision techniques are widely studied for automating the interpretation of sewer pipe inspection videos, yet previous studies mainly focus on defect detection and segmentation of individual images, which cannot identify if the defect is the same one across consecutive video frames (i.e. track the defect). Nevertheless, the number of unique defects in the video is required for evaluating the pipe condition. This paper proposes a framework for tracking multiple sewer defects in CCTV videos based on defect detection and metric learning. First, a deep learning -based defect detection model and a metric learning model is developed and trained respectively using with our sewer datasets. Then, using the detections and their features from the trained models as inputs, the tracking module predicts tracks by Kalman filter and associates tracks based on defect motion, appearance features, and defect types. Our experiments demonstrate the framework is able to track sewer defects in CCTV videos with a decent IDF1 score of 57.4%. We notice that tracking performance can be influenced by the detection accuracy and configurations of the metric learning module. By analyzing the tracking results based on different weights of the distance metrics, we find that assigning larger weights to appearance and defect class distance metrics tends to increase IDF1 score, while larger motion distance weight may degrade tracking accuracy. The proposed framework contributes by tracking multiple sewer defects, which can assist with counting unique defects in inspection videos.

Microwave Holographic Imaging of Nonmetallic Concentric Pipes

Recently, nonmetallic materials, which are resilient to corrosion, low cost, and lightweight, have been exploited in many industrial sectors such as oil and gas field. A common application of them is in the form of pipes. Despite the advantages, these pipes can be damaged by delamination, defects, holes, and erosion, which may lead to major production failures or even environmental catastrophes. To prevent these issues, nondestructive testing (NDT) of these pipes is required to replace these components in a timely manner. Due to the fact that the traditional NDT methods are mostly effective for metallic pipes, microwave holographic imaging is proposed for inspection of multiple concentric nonmetallic pipes. Besides, standardized minimum norm (SMN) is used to mitigate the depth biasing problem in the proposed imaging technique. To reduce the complexity of the imaging system, we aim at using the narrowest possible frequency band using an array of receiver antennas. The performance of the proposed NDT method is validated through simulation and experimental data in two scenarios including the antennas placed outside the pipes and inside the pipes.

Ultimate bearing capacities of large diameter pipeline elbows with ellipsoidal pits

Defects significantly affect the carrying capacity of pipe elbows. Accurately estimating the capacity permits fewer pipe inspections and replacements, saving time and reducing costs. To study the effect of defects, a 90° pipe elbow model with a 1016 mm-diameter and 19.9 mm wall thickness was built. The material is API X80 steel and the curvature radius is 6D. The plastic failure criterion and twice elastic slope method were adopted to determine the limit internal pressure load. Using the commercial finite element software, Abaqus, the influences of defect location, defect depth, defect length, defect width and pipe wall thickness on the ultimate bearing capacity of an pipe elbow with ellipsoidal pits was studied. Through multiple nonlinear regression analysis and nonlinear fitting, formulas for calculating the ultimate internal pressure load of large-diameter elbows with defects under various influencing factors were obtained. The formulas can provide guidance and reference for evaluating the residual strength of large-diameter elbows with defects.

МОДЕЛЬ І МЕТОД НАВЧАННЯ КЛАСИФІКАТОРА КОНТЕКСТІВ СПОСТЕРЕЖЕННЯ НА ЗОБРАЖЕННЯХ ВІДЕОІНСПЕКЦІЇ СТІЧНИХ ТРУБ

Video inspection is often used to diagnose sewer pipe defects. To correctly encode founded defects according to existing standards, it is necessary to consider a lot of contextual information about the orientation and location of the camera from sewer pipe video inspection. A model for the classification of context on frames during observations in the video inspection of sewer pipes and a five-stage method of machine learning is proposed. The main idea of the proposed approach is to combine the methods of deep machine learning with the principles of information maximization and coding with self-correcting Hamming codes. The proposed model consists of a deep convolutional neural network with a sigmoid layer followed by the rounding output layer and information-extreme decision rules. The first stages of the method are data augmentation and training of the feature extractor in the Siamese model with softmax triplet loss function. The next steps involve calculating a binary code for each class of recognition that is used as a label in learning with a binary cross-entropy loss function to increase the compactness of the distribution of each class's observations in the Hamming binary space. At the last stage of the training method, it is supposed to optimize the parameters of radial-basis decision rules in the Hamming space for each class according to the existing information-extreme criterion. The information criterion, expressed as a logarithmic function of the accuracy characteristics of the decision rules, provides the maximum generalization and reliability of the model under the most difficult conditions in the statistical sense. The effectiveness of this approach was tested on data provided by Ace Pipe Cleaning (Kansas City, USA) and MPWiK (Wroclaw, Poland) by comparing learning results according to the proposed and traditional models and training schemes. The obtained model of the image frame classifier provides acceptable for practical use classification accuracy on the test sample, which is 96.8 % and exceeds the result of the traditional scheme of training with the softmax output layer by 6.8 %.

Design of a dual-port, side-incident microwave probe for detection of in-pipe damage

This study reports on a dual-port, side-incident microwave probe dedicated to the non-destructive inspection of metal pipes. Two types of side-incident probes (LJ and JL) are proposed to emit TM01- and TM02-mode microwaves into the pipe under test, and each type of probe has two ports used to inspect in two opposite directions. Numerical simulations are conducted to study the dimensional parameters affecting the transmission characteristics of the probe and optimize the probe to obtain better mode purity and transmission directivity. The simulation results also suggest that the optimal probe dimensions for one inner pipe diameter can be applied to another diameter by multiplying the dimensions by a factor that is determined by the ratio of two diameters. Two LJ-type, side-incident probes, one with an inner diameter of 19 mm and one of 39 mm, are fabricated according to the simulation results. Experimental verification is subsequently carried out to test the detection directivity of the probes by detecting the short pipes with partially milled damage situated on both sides of both probes. Experimental results show that the proposed dual-port, side-incident probe can effectively detect the in-pipe damage on either side of the probe using the corresponding port and thus realize the directional pipe inspection. Furthermore, this method shows a potential to be applied to various inner pipe diameters.

Hammering Acoustic Analysis Using Machine Learning Techniques for Piping Inspection

In Japan, the deterioration of industrial plants built during the period of high economic growth in the middle of the 20th century has recently become a social concern. Corrosion under insulation (CUI) of piping in such plants is a pressing problem. X-ray and ultrasound inspections are conventional methods for detecting CUI; however, these methods are time-consuming and expensive. Therefore, rapid and low-cost screening techniques for CUI are required. We develop a hammering-type inspection robot system that moves inside the piping and records hammering sounds. Furthermore, we propose an acoustic analysis method to identify anomalous parts from the hammering sound using machine learning techniques. Using three testing pipes, we can successfully identify anomalous parts through acoustic analysis using a deep neural network as a supervised learning method. However, in practical piping inspections, the detection of anomalies without training data is required for further applications. Therefore, we investigate unsupervised learning anomaly detection using an autoencoder and a variational autoencoder and report the results.

Development of Inchworm Type Pipe Inspection Robot using Extension Type Flexible Pneumatic Actuators

In the case of damp and wet pipes, pipe inspection robots using pneumatic actuators offer advantages such as no electrical leakage and short circuit. In the previous study, a robot consisting of sliding/bending mechanisms using parallel arranged three extension type flexible pneumatic actuators and two holding mechanisms was successfully developed. In order to use the robot in thinner pipe, a novel and simpler propulsion mechanism utilising the difference of frictional force moving forward and backward are proposed and tested in this work. There are two mechanisms, which are “wriggling type” and “cilia type”. The “wriggling type” mechanism moves forward by wriggling its body while the “cilia type” mechanism moves by using plate type cilia that covered on the mechanism. Both mechanisms have been tested in the pipeline. As a result, it can be confirmed that the cilia type propulsion mechanism can travel in the pipe with accumulated water. It can be found that the mechanism can easily travel through corners while twisting its body by giving bending motion toward any direction.

A flexible and noncontact guided-wave transducer based on coils-only EMAT for pipe inspection

Ultrasonic guided wave testing technique, with its advantages of long propagation distance, low attenuation and high testing efficiency, has been widely used in nondestructive testing and structure health monitoring for pipes. In this paper, a flexible and noncontact guided-wave transducer based on a novel and coils-only electromagnetic acoustic transducer (EMAT) is proposed. In this new transducer, a planar magnetic coil of racetrack shape fed with rectangular-pulsed current is used to produce a quasi-static magnetic field, while a planar meander coil fed with RF-pulse current is used for pulsing and receiving guided-wave signals. Based on the magnetostrictive mechanism, the proposed transducer can generate and receive guided wave of L(0,2) mode in ferromagnetic metal pipes without coupling medium. Electromagnetic fields for the two coils and the induced ultrasonic waves are studied by simulation. Both the simulation and experimental results show that the proposed transducer is effective for generating and receiving ultrasonic guided wave of L(0,2) mode in ferromagnetic pipes, and can be applied on detection for corrosion. Due to the flexible, noncontact feature, and compact configuration of the proposed transducer, it can be conveniently applied for pipes in different diameter with rusty surface or coating in narrow space.

Registration method for maintenance-work support based on augmented-reality-model generation from drawing data

Abstract The importance of operations and maintenance (O&M) and piping inspection in shipbuilding and offshore plant industries has significantly increased recently. Therefore, this study proposes a system that uses augmented reality (AR) to support these operations. AR in O&M and inspection systems can increase work comprehension and efficiency by utilizing 3D graphics, instead of drawings, to describe specific work functions. To realize this improvement, the augmented model should correspond to the reality; if accurate registration is not achieved, it can disrupt the work functions. Therefore, marker-based AR is used to generate specific recognition objects and correct the location of the augmented model. However, owing to certain characteristics of the shipbuilding and offshore plant industries, the markers are likely to be damaged, and thus the cameras fail to clearly detect them. In this study, a 3D camera was used to generate a point cloud based on 3D image information. Accordingly, the area around the model to be detected was designated as the region of interest (ROI). Furthermore, it is difficult to support a high-end device environment because it is tested as a portable device that can be used in a working environment; therefore, superfluous data were removed by detecting the 3D edges in the ROI, thereby minimizing data operation. The scalability of this work was enhanced using the computer-aided design (CAD) file information extracted from the CAD tool (used in the shipbuilding industry) and converting it into a point cloud. The proposed system is expected to eliminate the issues related to understanding the O&M work by addressing registration errors that might occur in constrained environments and to improve the torsional phenomenon of the augmented models.

Visual Inspection of Welding Zone by Boundary-Aware Semantic Segmentation Algorithm

The segmentation effect of welding zone is vulnerable to image quality and target shape, which brings difficulty for the radiographic inspection of water-cooled pipe. A boundary-aware semantic segmentation algorithm is proposed to improve the precision and robustness of the welding zone inspection system. First, a convolutional neural network and atrous spatial pyramid pooling are employed to generate the feature map of the input image. Second, a boundary distance field (BDF) regression module is designed to predict the distance fields of the inner and outer boundaries, which can reflect the continuous boundary information of welding zone. Then, spatial attention weight is calculated according to the regressed boundary distance fields, and the calculated weight is fused into the feature map to enhance the significance of welding zone. Finally, the segmentation mask is predicted based on the fused feature map. Experiments indicate that the guidance of boundary information in the semantic segmentation algorithm can effectively reduce the interference of nonweld zones. The segmentation accuracy of the welding zone is 93.70%, and average detection time of a single image is less than 11 ms.

Design and Simulation of RC Pipe Inspection Robot

In recent days, most of the industries are facing many problems to find out damages inside the pipe line. The pipe inspection Robot was made to avoid this problem and to find out the exact place of damage and to reduce the time of evaluating, repairing and managing of interior surface pipes. One of the most important requirements in inspection, repairing and maintenance of interior pipelines is the ability to monitor and evaluate interior surface of pipes. This research work deals with the modelling of robot for inspection, cleaning, maintenance to evaluate the interior surface piping systems. The design of pipe inspection robot has been done in NX Cad (UNIGRAPHICS).the robot is designed in such way that it can be used for different pipe diameters. The design and development of pipe inspection robot involves a mechanism which in turn is connected to two opposite sides of frame of links, joints and wheels. So by using this pipe inspection robot one can find out the exact location of damage inside the pipe, so that it will be very easy to resolve the problem.