Pipeline Defect Detection Research Articles

Research on Pipeline Defect Detection Based on Optimized Faster R-cnn Algorithm

Aiming at the problem of low efficiency and high labor intensity in manual inspection of drainage pipes, a method of defect inspection of drainage pipes based on optimized Faster-Rcnn algorithm is proposed. Faster R-CNN is an algorithm proposed for target detection this year, which is based on the deep learning network model of region-based recommendation network (RPN). This paper analyzes the implementation of RPN network in Faster R-CNN algorithm, optimizes the network and introduces K-Means clustering method. By clustering all Anchors in the training set and inputting the clustering results into the RPN network, the training of the network can be accelerated and the recognition accuracy of the algorithm can be improved. The experimental results show that the accuracy of the algorithm is up to 92.4%, which has great application value. This research has important reference significance to promote the automatic detection of sewage pipeline defects.

Pipeline Defect Detection Cloud System Using Role Encryption and Hybrid Information

Pipeline defect detection systems collect the videos from cameras of pipeline robots, however the systems always analyzed these videos by offline systems or humans to detect the defects of potential security threats. The existing systems tend to reach the limit in terms of data access anywhere, access security and video processing on cloud. There is in need of studying on a pipeline defect detection cloud system for automatic pipeline inspection. In this paper, we deploy the framework of a cloud based pipeline defect detection system, including the user management module, pipeline robot control module, system service module, and defect detection module. In the system, we use a role encryption scheme for video collection, data uploading, and access security, and propose a hybrid information method for defect detection. The experimental results show that our approach is a scalable and efficient defection detection cloud system.

Formulation of consistent finite volume schemes for hydraulic transients

ABSTRACTIntentionally generated transient waves are used in the detection of defects in pipelines. Experience to date in these applications shows that the wider the band of injected wave frequencies the better is the detection effectiveness. Thus, as these technologies develop further, models that can handle high frequency waves will be required. Finite volume (FV) methods are well suited for high frequency wave phenomena, but their use in the water hammer field has been limited to date compared with other fields such as open channel flow. A major reason for this is that FV methods are formulated for simple boundary conditions such as no-flux or no-slip, but have not been well developed for the typically more complex boundary conditions found in pressurized pipeline systems (e.g. junctions, control valves, orifices, tanks and reservoirs). In those few instances in the literature where FV has been applied to water hammer problems, the approach has been to use FV for internal sections and method of characteristics (MOC) at the boundaries. The global order of accuracy of FV–MOC is governed by the MOC solution. To our knowledge, this paper constitutes a first attempt at handling boundary conditions within the FV framework. The approach places the boundary element within a FV to enforce mass and momentum conservation within this volume. The fluxes between the FV and the adjacent elements are then formulated in the usual manner. The approach is illustrated for the case of a valve, a reservoir and a junction. The finite volume method used is the Boltzmann-type scheme. The accuracy and efficiency of all schemes with the proposed non-iterative FV formulation of the boundary conditions are demonstrated through the following test cases: (i) water hammer wave interactions with a junction boundary characterized by a geometric discontinuity, (ii) water hammer wave interactions with a junction boundary characterized by a discontinuity in the value of wave speed, and (iii) water hammer wave interactions with a junction characterized by a flow rate discontinuity. The pure FV formulation guarantees mass and momentum conservation. The Boltzmann-based FV schemes capture discontinuity as well as wave interactions with boundary elements accurately. The stability of the proposed FV schemes is satisfied when .

Defect mapping in pipes by ultrasonic wavefield cross-correlation: A synthetic verification

This work presents a reverse-time imaging technique by cross-correlating the forward wavefield with the reverse wavefield for the detection, localization, and sizing of defects in pipelines. The presented technique allows to capture the wavefield reflectivity at the places of ultrasonic wave scattering and reflections. Thus, the method is suitable for detecting pipe defects of either point-like or finite-size types using data from a pulse-echo setup. By using synthetic data generated by 3D spectral element pipe models, we show that the 3D wavefield cross-correlation imaging is capable in the case of cylindrical guided ultrasonic waves. With a ring setup of transducers, we analyze the imaging results obtained from the synthetic single-transducer and all-transducer firings. The presented pipe flaw imaging method is straightforward to carry out using a suitable wave equation solver. Also, the method does not suffer from long iterative runs and numerical convergence issues commonly connected with imaging methods based on either deterministic optimization or statistical inference. The imaging procedure can be fully baseline-free by performing data processing to remove direct arrivals from the ultrasound data.

In-pipe fibre optic pressure sensor array for hydraulic transient measurement with application to leak detection

Controlled hydraulic transient pressure waves can be used to actively assess the condition of pressurised pipelines (e.g. leak detection). One practical challenge to the application of the technique to buried water pipelines is in the pressure measurement – fast-response pressure transducers are costly and the deployment is limited to one sensor per access point (e.g. at air valves or fire hydrants). The measurement from a single sensor is always the superposition of the waves travelling upstream and downstream along a pipe. It is often difficult to determine the direction of the propagating waves, unless there are suitable locations nearby where additional measurements can be made. The research reported here develops a fibre optic pressure sensor array for in-pipe transient pressure measurement. The sensor array consists of five Fibre Bragg Gratings (FBGs), each formed into a custom designed pressure transducer. The cable containing the array has a diameter of approximately 4 mm, making it easy to insert into a pipeline through a single access point for distributed transient pressure measurement. The usefulness of the fibre optic pressure sensors for transient-based pipeline leak detection is validated through laboratory experiments. The measurement from each individual fibre optic sensor is generally consistent with conventional piezo-resistive pressure transducers over a wide frequency range. The sensor array configuration enables the determination of the directional information of waves, therefore enhancing the ability of pipeline defect detection and localisation.

Effect of internal surface damage on vibration behavior of a composite pipe conveying fluid

Composite pipes have become a viable alternative to metallic pipes in several applications. Flow-accelerated erosion and internal surface attack often result in thickness thinning that may compromise the structural integrity of the pipe and lead to its failure. In this paper, we investigate how the internal surface damage is reflected on the vibration behavior of a composite pipe conveying fluid. The defected pipe-fluid system is modeled using the extended Hamilton’s approach and discretized using the wavelet-based finite element method. The modal characteristics of the pipe vibrations have been obtained by solving the generalized eigenvalue problem. The developed model was validated and some benchmark solutions are presented to highlight the effects of the internal wall-thinning on the vibrational behavior of composite pipes conveying fluid. The obtained results facilitate the future research by revealing the potential of using the vibration signature as a basis for detection of erosion-induced internal defects in pipelines.

Fault location method for unexposed gas trunk line insulation at stray current constant effect area

For the purpose of gas trunk lines safe operation, two types of pipe wall metal anticorrosion protection are generally used - the passive (insulation coating) protection and the active (electrochemical) protection. In the process of a pipeline long-term operation, its insulation is subject to wear and damage. Electrochemical protection means of a certain potential value prevent metal dissolution in the soil. When insulation wear and tear attains a level of insufficiency of the protection potential value, the insulating coating needs repair which is a labor-consuming procedure. To reduce the risk of such situation, it is necessary to make inspection rounds to monitor the condition of pipe insulation. A method for pipeline insulation coating unexposed fault location based on Pearson method is considered, wherein a working cathodic protection station signal of 100 Hz frequency is used, which makes installation of a generator unnecessary, and also a specific generator signal of 1 kHz frequency is used at high noise immunity and sensitivity of the instrument complex. This method enables detection and sizing of unexposed pipeline defects within the zones of earth current permanent action. High noise immunity of selective indicators allows for operation in proximity to 110 kV, 220 kV, and 500 kV power transmission lines in action.

Detection and sizing of metal-loss defects in oil and gas pipelines using pattern-adapted wavelets and machine learning

Signals collected from the magnetic scans of metal-loss defects have distinct patterns. Experienced pipeline engineers are able to recognize those patterns in magnetic flux leakage (MFL) scans of pipelines, and use them to characterize defect types (e.g., corrosion, cracks, dents, etc.) and estimate their lengths and depths. This task, however, can be highly cumbersome to a human operator, because of the large amount of data to be analyzed. This paper proposes a solution to automate the analysis of MFL signals. The proposed solution uses pattern-adapted wavelets to detect and estimate the length of metal-loss defects. Once the parts of MFL signals corresponding to metal-loss defects are isolated, artificial neural networks are used to predict their depth. The proposed technique is computationally efficient, achieves high levels of accuracy, and works for a wide range of defect shapes.

A one dimensional numerical approach for computing the eigenmodes of elastic waves in buried pipelines

Ultrasonic guided waves are often used in the detection of defects in oil and gas pipelines. It is common for these pipelines to be buried underground and this may restrict the length of the pipe that can be successfully tested. This is because acoustic energy travelling along the pipe walls may radiate out into the surrounding medium. Accordingly, it is important to develop a better understanding of the way in which elastic waves propagate along the walls of buried pipes, and so in this article a numerical model is developed that is suitable for computing the eigenmodes for uncoated and coated buried pipes. This is achieved by combining a one dimensional eigensolution based on the semi-analytic finite element (SAFE) method, with a perfectly matched layer (PML) for the infinite medium surrounding the pipe. This article also explores an alternative exponential complex coordinate stretching function for the PML in order to improve solution convergence. It is shown for buried pipelines that accurate solutions may be obtained over the entire frequency range typically used in long range ultrasonic testing (LRUT) using a PML layer with a thickness equal to the pipe wall thickness. This delivers a fast and computationally efficient method and it is shown for pipes buried in sand or soil that relevant eigenmodes can be computed and sorted in less than one second using relatively modest computer hardware. The method is also used to find eigenmodes for a buried pipe coated with the viscoelastic material bitumen. It was recently observed in the literature that a viscoelastic coating may effectively isolate particular eigenmodes so that energy does not radiate from these modes into the surrounding [elastic] medium. A similar effect is also observed in this article and it is shown that this occurs even for a relatively thin layer of bitumen, and when the shear impedance of the coating material is larger than that of the surrounding medium.

Estimation of Depth and Length of Defects from Magnetic Flux Leakage Measurements: Verification with Simulations, Experiments, and Pigging data

Magnetic flux leakage (MFL) method is the most widely used non-destructive testing techniques for detection and characterization of defects in product transmission pipelines. The maximum safe operating pressure is a crucial parameter in practice, which is predicted with respect to the length, width, and depth of defects. In a previous work, an algorithm is devised to detect and estimate the width of the defect based on MFL signals. In this paper, an efficient method based on axial MFL level contours is devised to estimate the length of the defect. This method uses the patterns of signal level contours in the region corresponding to the defect's area. In addition, a Gaussian radial basis function neural network (NN) is trained to approximate the depth of the defect. The NN is fed with the estimated length, width, and signal peak-to-peak values, and the output of the network is the estimated depth. The proposed detection and estimation method is applied to MFL measurements to detect and determine the sizing of metal loss defects. The efficacy and accuracy of the proposed methods are examined through a rich set of sample defects that contains simulated defects, designed defects that are carved on a real pipe by milling, and actual pigging data validated by several dig up verifications. Obtained results confirm the effectiveness and accuracy of the proposed depth and length estimation method along with the previously devised detection and width estimation methods, in characterization of metal loss defects.

Ultrasonic isolation of buried pipes

Long-range guided wave testing (GWT) is used routinely for the monitoring and detection of corrosion defects in above ground pipelines. The GWT test range in buried, coated pipelines is greatly reduced compared to above ground configurations due to energy leakage into the embedding soil. In this paper, the effect of pipe coatings on the guided wave attenuation is investigated with the aim of increasing test ranges for buried pipelines. The attenuation of the T(0,1) and L(0,2) guided wave modes is measured using a full-scale experimental apparatus in a fusion-bonded epoxy (FBE)-coated 8in. pipe, buried in loose and compacted sand. Tests are performed over a frequency range typically used in GWT of 10–35kHz and compared with model predictions. It is shown that the application of a low impedance coating between the FBE layer and the sand effectively decouples the influence of the sand on the ultrasound leakage from the buried pipe. Ultrasonic isolation of a buried pipe is demonstrated by coating the pipe with a Polyethylene (PE)-foam layer that has a smaller impedance than both the pipe and sand, and has the ability to withstand the overburden load from the sand. The measured attenuation in the buried PE-foam-FBE-coated pipe is found to be substantially reduced, in the range of 0.3–1.2dBm−1 for loose and compacted sand conditions, compared to measured attenuation of 1.7–4.7dBm−1 in the buried FBE-coated pipe without the PE-foam. The acoustic properties of the PE-foam are measured independently using ultrasonic interferometry and incorporated into model predictions of guided wave propagation in buried coated pipe. Good agreement is found between the experimental measurements and model predictions. The attenuation exhibits periodic peaks in the frequency domain corresponding to the through-thickness resonance frequencies of the coating layer. The large reduction in guided wave attenuation for PE-coated pipes would lead to greatly increased GWT test ranges; such coatings would be attractive for new pipeline installations.

Investigation of guided wave propagation and attenuation in pipe buried in sand

Long-range guided wave testing is a well-established method for detection of corrosion defects in pipelines. The method is currently used routinely for above ground pipelines in a variety of industries, e.g. petrochemical and energy. When the method is applied to pipes buried in soil, test ranges tend to be significantly compromised and unpredictable due to attenuation of the guided wave resulting from energy leakage into the embedding soil. The attenuation characteristics of guided wave propagation in an 8in. pipe buried in sand are investigated using a laboratory full-scale experimental rig and model predictions. We report measurements of attenuation of the T(0,1) and L(0,2) guided wave modes over a range of sand conditions, including loose, compacted, mechanically compacted, water saturated and drained. Attenuation values are found to be in the range of 1.65–5.5dB/m and 0.98–3.2dB/m for the torsional and longitudinal modes, respectively, over the frequency of 11–34kHz. The application of overburden pressure modifies the compaction of the sand and increases the attenuation. Mechanical compaction of the sand yields similar attenuation values to those obtained with applied overburden pressure. The attenuation decreases in the fully water-saturated sand, and increases in drained sand to values comparable with those obtained for compacted sand. Attenuation measurements are compared with Disperse software model predictions and confirm that the attenuation phenomenon in buried pipes is essentially governed by the bulk shear velocity in the sand. The attenuation behaviour of the torsional guided wave mode is found not to be captured by a uniform soil model; comparison with predictions obtained with the Disperse software suggest that this is likely to be due to a layer of sand adhering to the surface of the pipe.

Scanning System Non-destructive Inspection of Weld Connections

In this paper a system for branch pipes welding connections is described. One of the important problems for lots of production facilities is defect detection in pipelines. Often the defect detection is both dangerous and a matter of high liability. In such case robotic system are preferable to substitute humans as they grant higher precision, unbiased analyzing, good repeatability, etc. The system presented here satisfies all the requirements while being equipped with modern sensors for non-destructive defectoscopy and simple yet effective and unified design.

Error Factor and Mathematical Model of Positioning with Odometer Wheel

High positioning accuracy with odometer wheel is of great importance for internal pipeline defect detection. In current research, the qualitative analysis is focused on and the accumulation characteristics of positioning errors are not shown. Based on this, the models and the accumulation characteristics of positioning errors are investigated in this paper. The errors are classified into systematic errors and random errors. Systematic errors consist of difference between actual and nominal diameters, diameter difference among wheels, and terminal error. Random errors consist of initial point error, weld error, and oversized clearance. The mathematical models of the errors are obtained by the methods of quantitative analysis and contrastive analysis. Through calculation examples, the results show the relationship of the errors and the effect of each error on total error. Diameter calibration and detection in sections are proposed to improve the positioning accuracy.

Defect Detection in Pipelines during Operation Using Magnetic Flux Leakage and Phased Array Ultrasonic Method

The present article is focused on the non-destructive testing (NDT) inspection of pipelines during operation namely Magnetic Flux Leakage (MFL) method and Phased Array ultrasonic (PA) method. MFL inspection technique is electromagnetic test method primarily used to detect flaws or defects in high-permeability of ferromagnetic metals such as carbon steel tubing, plate, wire rope and tubular parts. PA ultrasonic method is an advanced NDT method that is used to detect component failures i.e. cracks or flaws and thereby determine component quality. Due to the possibility to control parameters such as beam angle and focal distance, this method is very efficient regarding to the defect detection and speed of testing. In this article real pipeline defect was identified by MFL method in the internal pipe inspection. This defect was fully mapped by Phased Array ultrasonic method in the terrain. 3D model of defect in the tested material was created from measured data and obtained by PA method. The real dimensions of the defect determined from measurements by the method of MFL and PA are compared.

Application of morphological segmentation to leaking defect detection in sewer pipelines.

As one of major underground pipelines, sewerage is an important infrastructure in any modern city. The most common problem occurring in sewerage is leaking, whose position and failure level is typically idengified through closed circuit television (CCTV) inspection in order to facilitate rehabilitation process. This paper proposes a novel method of computer vision, morphological segmentation based on edge detection (MSED), to assist inspectors in detecting pipeline defects in CCTV inspection images. In addition to MSED, other mathematical morphology-based image segmentation methods, including opening top-hat operation (OTHO) and closing bottom-hat operation (CBHO), were also applied to the defect detection in vitrified clay sewer pipelines. The CCTV inspection images of the sewer system in the 9th district, Taichung City, Taiwan were selected as the experimental materials. The segmentation results demonstrate that MSED and OTHO are useful for the detection of cracks and open joints, respectively, which are the typical leakage defects found in sewer pipelines.

Application of the Methods of Mechanical Resonance for the Detection of Defects in Steels of Steam Pipelines After Operation

We use linear and nonlinear methods of mechanical resonance to detect defects in steels after operation for ~7 · 104 − 2.45 · 105 h in steam pipelines of a thermoelectric plant. We determine the sound velocity, vibration decrement, its dependence on the amplitude of maximum cyclic stresses, and the dependence of the resonance frequency of a cantilevered rod specimen on the strain amplitude. It is shown that the vibration decrement of materials after operation is higher than in the as-delivered state and its dependence on the stress amplitude becomes stronger. It is found that the material after operation is characterized by the nonlinear elastic behavior, which manifests itself as a distortion of the resonance curve and in the form of dependence of the resonance frequency on the amplitude. We establish a linear dependence of the resonance frequency on the strain amplitude, which is in good agreement with a similar dependence of the vibration decrement. We confirm that the nonlinear methods are promising for the detection of defects in steam pipelines.

Pulsed eddy current differential probe to detect the defects in a stainless steel pipe

Pulsed eddy current (PEC) is an electromagnetic nondestructive technique widely used to detect and quantify the flaws in conducting materials. In the present study a differential Hall-sensor probe which is used in the PEC system has been fabricated for the detection of defects in stainless steel pipelines. The differential probe has an exciting coil with two Hall-sensors. A stainless steel test sample with electrical discharge machining (EDM) notches under different depths of 1–5 mm was made and the sample was laminated by plastic insulation having uniform thickness to simulate the pipelines in nuclear power plants (NPPs). The driving coil in the probe is excited by a rectangular current pulse and the resultant response, which is the difference of the two Hall–sensors, has been detected as the PEC probe signal. The discriminating time domain features of the detected pulse such as peak value and time to zero are used to interpret the experimental results with the defects in the test sample. A feature extraction technique such as spectral power density has been devised to infer the PEC response.

Intelligent Monitoring Approach for Pipeline Defect Detection from MFL Inspection

Artificial Neural Networks(ANNS) have top level of capability to progress the estimation of cracks in metal tubes. The aim of this paper is to propose an algorithm to identify modeled cracks by magnetic flux leakage inspection in Non Destructive Testing (NDT) [1, 2, 3, 4, 5, and 6]. The analysis is carried out with a simulated database of signals in which the depth of the crack, its width, shape, And geometric dimension of the detection process, is allowed to change. The simulated signal is input to the network, after a reduction process in which the main features of the signal are extracted. Feature extractors are used in pattern recognition area due to their advantages in representing data. With this approach classifier’s job became easier and more effective. The main goal of the feature extractor is to reflect the characteristics of an object in a given dataset. In this way feature extractor simplify the amount of resources required to describe a large dataset accurately. This paper presents the results of employing different kinds of feature extraction functions and classification and provides compression between them. As the output of ANN, we shall justify if any care in meta lto indicate whether the input signal is crack or not. The analysis based on the neural network and feature extractor functions is shown to be quite top probability of detection.

The application of synthetic focusing for imaging crack-like defects in pipelines using guided waves

This paper deals with quantifying the performance of a technique for detection, location, and sizing of circumferential crack-like defects in pipelines using synthetically focused guided waves. The system employs a circumferential array of piezoelectric transducer elements. A torsional probing guided wave is excited using the array, which subsequently interacts with the reflecting features of the pipe, such as defects or weld caps. The recorded backscattered signals are synthetically focused to every point of interest in the pipe wall, to form an image of the reflecting features of the pipe. The defect image amplitude is used to estimate the defect depth, and the full width at half maximum of the defect image circumferential profile is used to estimate the circumferential extent of the defect. The imaging system is tested with data from finite element simulations and from laboratory experiments. It is found that reliable sizing of circumferential cracks in finite element simulations and experiments can be achieved if the circumferential extent of the defect is greater than 1.5 lambda(S), where lambda(S) is the shear wavelength at the frequency of inspection. This result is theoretically valid for any pipe size, any axial defect location, and any inspection frequency. Amplitude gains of around 18 dB over an unfocused system have been observed experimentally in an 8-inch pipe with a 9 dB SNR improvement.