Leak Location Research Articles

Intelligent identification method for pipeline leakage based on GPR time–frequency features and deep learning

ABSTRACT The detection and localization of water pipeline leakages based on ground penetrating radar (GPR) technology are gradually becoming a research hotspot. Current methods mostly focus on exploring the patterns of B-Scan images, heavily relying on the subjective experience of detection personnel, which can lead to misjudgments. Moreover, the large amount of data makes it difficult for manual processing. Therefore, a method based on wavelet transform (WT) and ResNet-50 is proposed to identify the time-frequency characteristics of GPR data, thereby achieving intelligent localization of pipeline leakages. The B-Scan images from GPR are transformed into time–frequency scale images using WT, and the features in both time and frequency domains are combined to enhance the representation of leakages. Subsequently, ResNet-50 is employed for feature extraction and leakage identification. Additionally, a deviation correction mechanism is proposed to improve the clarity of the prediction results. Experimental results demonstrate that ResNet-50 achieves an accuracy of 0.917 and a recall of 0.998 on the time-frequency dataset, almost detecting all leakages, with a detection efficiency of 0.0165 s per data trace. The comprehensive method is validated in the field, indicating its capability to accurately identify and localize pipeline leakages.

A stepwise fast leakage localization method applying the strategy of dynamic area narrowing down for large-scale water distribution network

ABSTRACT Urban water distribution networks (WDNs) are facing serious leakage problems. As cities expand, the leakage localization burden on large WDNs gradually increases. Although methods have been widely researched, there is a lack of studies and successful applications for large-scale WDNs. To deal with this, a stepwise fast leakage localization Method, SFLLM, utilizing the `dynamic area narrowing down (DAND)' strategy and coupled leakage features (CLF) is proposed. The SFLLM includes a fast-and-dynamic stage using the DAND strategy to reduce the potential leakage area and an accurate localization stage. Only partial representative candidate locations are required to simulate leakages by DAND, and meantime CLF is used to analyze the leakage similarities so that the localization accuracy and efficiency can be improved. SFLLM was tested on a benchmark WDN, saving more than 88% of simulation by DAND strategy and achieving localization in 11 seconds. The results also proved the enhanced performance of CLF in ensuring the stability of the accuracy against various types of uncertainties that may occur in real WDNs. Moreover, three real burst leaks in an actual large-scale WDN were localized within 205 m in about 22 minutes by SFLLM, showing the method's reliable applicability in guiding field leak exploration.

Multivariate analysis and construction and validation of a nomogram model from data of 1610 patients with non-tumor-related anastomotic stenosis after rectal cancer surgery

Objective: To assess the risk factors affecting development of non-tumor- related anastomotic stenosis after rectal cancer and to construct a nomogram prediction model. Methods: This was a retrospective study of data of patients who had undergone excision with one-stage intestinal anastomosis for rectal cancer between January 2003 and September 2018 in Nanfang Hospital of Southern Medical University. The exclusion criteria were as follows: (1) pathological examination of the operative specimen revealed residual tumor on the incision margin of the anastomosis; (2) pathological examination of postoperative colonoscopy specimens revealed tumor recurrence at the anastomotic stenosis, or postoperative imaging evaluation and tumor marker monitoring indicated tumor recurrence; (3) follow-up time <3 months; and (4) simultaneous multiple primary cancers. Univariate analysis using the χ2 or Fisher's exact test was performed to assess the study patients' baseline characteristics and variables such as tumor-related factors and surgical approach (P<0.05). Multivariate analysis using binary logistic regression was then performed to identify independent risk factors for development of non-tumor-related anastomotic stenosis after rectal cancer. Finally, a nomogram model for predicting non-tumor-related anastomotic stenosis after rectal cancer surgery was constructed using R software. The reliability and accuracy of this prediction model was evaluated using internal validation and calculation of the area under the curve of the model's receiver characteristic curve (ROC). Results: The study cohort comprised 1,610 patients, including 1,008 men and 602 women of median age 59 (50, 67) years and median body mass index 22.4 (20.2, 24.5) kg/m². Non-tumor-related anastomotic stenosis developed in 121 (7.5%) of these patients. The incidence of non-tumor-related anastomotic stenosis in patients who had undergone neoadjuvant chemotherapy, neoadjuvant radiotherapy, and surgery alone was 11.2% (10/89), 26.4% (47/178), and 4.8% (64/1,343), respectively. Neoadjuvant treatment (neoadjuvant chemotherapy: OR=2.455, 95%CI: 1.148-5.253, P=0.021; neoadjuvant chemoradiotherapy, OR=3.882, 95%CI: 2.425-6.216, P<0.001), anastomotic leakage (OR=7.960, 95%CI: 4.550-13.926, P<0.001), open laparotomy (OR=3.412, 95%CI: 1.772-6.571, P<0.001), and tumor location (distance of tumor from the anal verge 5-10 cm: OR=2.381, 95%CI:1.227-4.691, P<0.001; distance of tumor from the anal verge <5 cm: OR=5.985,95% CI: 3.039-11.787, P<0.001) were identified as independent risk factors for non-tumor-related anastomotic stenosis. Thereafter, a nomogram prediction model incorporating the four identified risk factors for development of anastomotic stenosis after rectal cancer was developed. The area under the curve of the model ROC was 0.815 (0.773-0.857, P<0.001), and the C-index of the predictive model was 0.815, indicating that the model's calibration curve fitted well with the ideal curve. Conclusion: Non-tumor-related anastomotic stenosis after rectal cancer surgery is significantly associated with neoadjuvant treatment, anastomotic leakage, surgical procedure, and tumor location. A nomogram based on these four factors demonstrated good discrimination and calibration, and would therefore be useful for screening individuals at risk of anastomotic stenosis after rectal cancer surgery.

Sinkhole Risk-Based Sensor Placement for Leakage Localization in Water Distribution Networks with a Data-Driven Approach

Sinkhole Risk-Based Sensor Placement for Leakage Localization in Water Distribution Networks with a Data-Driven Approach

Application of Distributed Acoustic Sensing Technology in Pipeline Leakage Monitoring

Pipeline leak monitoring is an important industrial safety measure designed to ensure the safety of liquids or gases during transportation. Distributed acoustic sensing (DAS) technology is based on the reverse Rayleigh scattering inside the fiber to reflect the change of the measured physical quantity, and has great advantages in monitoring range, environmental adaptability, transmission loss control and system stability. In this paper, the pipeline leakage monitoring technology based on distributed acoustic sensing fiber is used to study the leakage signal of small leak aperture. In order to improve the sensitivity of leakage monitoring, the optical fiber is spiral wound on the pipe section. The identification method of pipeline leakage signal based on fast Fourier transform is proposed. By analyzing the vibration of the optical fiber in the time domain and the frequency domain, the leakage signal can be accurately monitored. Pipeline leakage tests with different leak apertures were carried out, and the leakage locations were studied by energy attenuation and cross-correlation techniques. The experimental results show that the time-domain signal fluctuates obviously and the full-band energy of the frequency-domain signal increases after pipeline leakage. The increase of leakage diameter will gradually increase the signal energy, and the leakage energy will gradually move from high frequency to low frequency. The energy attenuation positioning technique can locate the leakage within the range of a single sensing unit, and determine the leakage location through cross-correlation analysis with an error of less than 3 m.

Leak detection and localization in underground water supply system using thermal imaging and geophone signals through machine learning

The underground water pipeline system is a crucial infrastructure that largely remains out of sight. However, it is the source of a clean and uninterrupted flow of water for our everyday lives. Various factors, including corrosion, material degradation, ground movement, and improper maintenance, cause pipe leaks, a silent crisis that causes an estimated 39 billion dollars of loss every year. Prompt leakage detection and localization can help reduce the loss. This research investigates the potential of two machine learning models as supporting tools for surveying extensive areas to identify and pinpoint the location of underground leaks. The presented combined approach ensures the speed and accuracy of the leakage survey. The first machine learning model is a hybrid ML model that employs thermal imaging to identify subterranean water leakage. It relies on detecting thermal anomalies and distinctive signatures associated with water leakage to identify and locate underground water leakage. The developed model can detect up to 750 mm underground leakage with 95.20 % accuracy. The second model uses binaural audio from geophones to localize the leakage position. The model utilizes interaural time difference and interaural phase difference for localization purposes, and the 1D-CNN network delivers an angle in twenty-degree increments with an accuracy of 88.19 %. Large-scale implementation of the proposed model could be a powerful catalyst to reduce water loss in the water supply system.

Novel differential pumping system with extended sampling range: Design and application in the EAST tokamak

A novel differential pumping system (DPS) with an adjustable sampling rate has been developed and implemented in the Experimental Advanced Superconducting Tokamak (EAST). This system is equipped with a high-resolution quadrupole mass spectrometer (QMS), covering a mass range of 1–16 atomic mass units (AMU), alongside a conventional QMS with a broader range of 1–200 AMU. The DPS features a dynamic pressure sampling capability from 10−2 to 105 Pa, utilizing a gas dosing valve and an angle valve, enabling a wide gas sampling ratio range of 10−9 to 3 × 10−2. The high-resolution QMS (QMG700) is adept at directly quantifying the partial pressures of deuterium and helium within mixtures, while the conventional QMS (RGA200) offers indirect measurement of D2 partial pressures above 2 × 10−5 Pa, complemented by direct wide-range mass analysis. During the EAST tokamak campaigns of 2023 and 2024, the DPS was effectively utilized for glow discharge cleaning (GDC) at a pressure of 3 Pa and ion cyclotron wall conditioning (ICWC) at approximately 3 × 10−3 Pa, accurately distinguishing deuterium pressure from the helium environment in the wide pressure range of vacuum vessel. Moreover, the DPS was successfully employed to trace the vacuum leak location in EAST campaign with a pressure of up to hundreds Pa. This DPS is a key tool for gas ingredient analysis and leak detection in the EAST campaigns, providing a valuable reference for the vacuum operation of future fusion reactors.

Leakage detection applied to the Da Hedong Reservoir dam (China) based on the flow-field fitting method

Detecting leakage in concrete gravity dams presents a formidable challenge. The flow-field fitting (FFF) method is used to identify leaks in China’s Da Hedong Reservoir dam. We develop vertical- and horizontal-gradient approaches based on the transmitted signal electrical field. A plastic frame is developed to allow the probe to measure horizontal gradients. We determine the exact location of leaks at the reservoir’s bottom and upstream face, respectively, assuming that current leakages are located in hydraulic leakage zones. Numerical experiments and water pressure tests reveal that the conductive silt layer beneath the reservoir can lower the response of the potential values at the leakage inlet, and the effect of the silt layer should be considered and not ignored when interpreting measured data. In addition, with regard to the principle of superposition, the contribution associated with metal pipes in the dam body can be roughly determined by subtracting it from the original potential field data, thereby aiding in locating leaks around the metal pipes. In addition, the FFF measurements from the reservoir’s bottom also confirm the position of existing faults in the substratum at the reservoir site. Last, the results of the FFF method are tested against supporting evidence, such as ground penetrating radar and water pressure tests. In conclusion, these independent data confirm the effectiveness of the FFF method for leakage localization in the concrete gravity dam.

Arrangement of Nondestructive Tests for Technical Inspection Process of Electrofusion Weld in Gas Polyethylene Pipe

Due to the reduction of operational costs, Non-corrosiveness, and ease of implementation, polyethylene pipes are a suitable replacement for steel pipes in gas distribution networks. Considering the irreversible events of gas leakage, effective and feasible non-destructive testing of electrofusion welding is an essential factor to avoid the hazards of gas leakage. To control the quality of welds, the nature of the electrofusion structure leads us to manufacture vacuum boxes, internal pressure devices, and Tapping Saddle Connector for testing polyethylene joint leakages. This article focuses on the sequence of these non-destructive tests to monitor the health of polyethylene electrofusion joint structures under gas network project implementation conditions. To examine the testing chain and the proposed sequence of non-destructive tests, the data gathering process was performed, and with respect to the sensitivity of the electrofusion weld to the operator's skill, welding process, and the condition of the pipe and fittings, in each case the influential factors to find the key parameters were evaluated. The results of laboratory tests in cases of short tests and long-life tests using artificial defects such as slow crack growth with defective samples of on-site joints and defects reported in projects were compared. To validate leak detection, leak location, and leak measurement, weld simulation with finite element and CFD software analysis conforming to real conditions was considered. The test sequence, acceptance criteria, and calibration method of the tools according to EN 1555 standard were argued for performing the final inspection steps. In this perspective, for each leak testing technique, the average type, pressure level, test duration, and finally the acceptance criteria were proposed. The main outcome is the technical specifications for electrofusion weld assessment and baseline risk- based inspection scheme in gas supply projects. In this regard, a database software based on the proposed ITP (Inspection and Testing Plan) has been developed, which is fundamental for monitoring the whole project from the first stage of polyethylene granule control, pipe and fitting manufacture, and gas injection to the final stage in network and commissioning. Operations to determine failure causes and suitability for performance assessment. Another proposed idea is standardizing the recent technical specifications internationally. This idea will be implemented when management realizes that non-destructive testing can lead to cost savings, protection of humans and the environment, and when codes and standards encompass such requirements.

Pressure Interpolation in Water Distribution Networks by Using Gaussian Processes: Application to Leak Diagnosis

This work presents the reconstruction of the pressure head map of a water distribution system (WDS). This approach relies on historical data collected from a reduced number of sensors placed at some nodes of the WDS. Thus, a Gaussian regression process is then applied to estimate the pressure head at those nodes without a sensor, which allows the reconstruction of the pressure map for the whole network. Then, for leak diagnosis purposes, a dataset of pressure head maps of the WDN is created considering leaky scenarios, and a correlation method is applied to estimate the leak location. Then, for clarity, the Hanoi network is used to evaluate the performance of this leak diagnosis strategy in a simulation environment, assuming the availability of only three sensors. The results showed the potential for pressure head map reconstruction and leak localization.

Improving Acoustic Methods of Pipeline Leak Location with Distributed Sensing

Improving Acoustic Methods of Pipeline Leak Location with Distributed Sensing

Current dynamics to detect the spread of oil spills in coastal waters

Abstract Surface flow dynamics modelling is an important aspect of monitoring and detecting oil spills in marine environments, especially coastal waters. The model can predict the spatiotemporal distribution of oil spills in waters before remote sensing technologies such as radar and synthetic aperture radar (SAR) detect the oil spill distribution range. The purpose of this study is to model two-dimensional (2D) surface flow to predict the extent of oil spills, especially in shallow water or coastal areas. The accuracy of this ocean current distribution model was then verified using the case of an oil spill in East Aceh waters based on the results of the detection of synthetic aperture Interference images from the Sentinel-2B satellite Radar (InSAR) Technology from August 1st to September 20th, 2021. The results show that when the fluid dynamics used is residual flow, the oil distribution trajectory moves partly to the north and partly to the south toward the coast of the leak location. However, when mean current is applied, the distribution expands to the south toward the coast. Based on the state of the wind distribution when the oil spill occurred, it shows the same direction as the residual current model, but the influence of the wind is greater than the residual current. So, it can be concluded that oil spills in the waters of East Aceh are more influenced by a combination of mean currents and surface wind direction. These results were validated by the results obtained from satellite images. Therefore, the satellite image validation results indicate that mean flow provide a more accurate indication of oil spill distribution than the average flow in South Aceh waters.

Enhancing Leak Location in Buried Water Pipes using Array Signal Processing Techniques: the Effect of Wave Velocity Variation

Enhancing Leak Location in Buried Water Pipes using Array Signal Processing Techniques: the Effect of Wave Velocity Variation

Research on pipeline leakage localization method based on CZIM algorithm

The time difference of arrival is a common method to find the leakage point of water pipeline. The leakage point localization is achieved by calculating the time delay between the signals reaching different sensors. Mainstream time delay estimation algorithms based on signal correlation analysis are susceptible to the introduction of noise signals, low sampling rates, and signal clipping, resulting in inaccurate localization results. The article analyzed the impact of different interference factors and proposed a new time delay estimation algorithm based on signal cross-zero information modulation (CZIM) to address these problems. By normalizing the amplitude of the two signals at the detection points on both sides of the pipeline leakage position, two sets of sparse signal sequences with only two eigenvalues of 0 and 1 are obtained. The error coefficient function is calculated by a similar traversal method to finally index the time delay. In this paper, the principle and characteristics of the algorithm are analyzed and compared with the most commonly used GCC method. In both numerical simulations and actual pipe leakage localization experiments, the CZIM algorithm has shown its wide applicability, low impact by low sampling rate, and adaptability to low signal-to-noise ratios, etc. At the same time, the algorithm is simple in design and has a small amount of calculation and can meet the demand for real-time data processing, providing a new idea for the development of acoustic localization technology.

Air-Leak-Management Supported by Ventilation-SPECT/CT: A New Perspective on a Long-Known Problem

Abstract Background Prolonged air leakage (PAL) is a common problem after lung surgery. Localization of air leaks (AL) remains challenging in some cases especially in minimally invasive surgery. Ventilation-SPECT/Computertomography (vSPECT/CT) has been reported as potential aid to identify and localize AL. Aims To evaluate vSPECT/CT as potential aid for the localization of AL and explore its use as support for PAL management. Methods Retrospective analysis of a case series of seven patients with PAL after minimally invasive thoracic surgery at our institution in 2023 for which vSPECT/CT was applied. Results Six patients underwent oncologic lung resections, one patient underwent thymectomy for a thymoma. In 5/7 cases an AL was spotted with vSPECT/CT (maximal AL ranged from 40-2000 ml/min). In the remaining two cases (with a maximal AL of 1900ml/min for 1 day, and 10 ml/min after more than 4 weeks and continuous clinical signs of PAL), the AL could neither be localized during revision surgery. Nevertheless, the AL ceased in all patients after revision surgery. In two cases the detected AL supported the continuance of conservative treatment as it confirmed a leak within the prior resection site (maximal AL 40ml/min and 200ml/min). These results are depicted in the attached visual abstract. Conclusion The use of vSPECT/CT for localizing ALs proved to be precise in all instances where an imaging spot was observed (5/5 cases). Additionally, in the two cases where no spot was detected, no AL was found during subsequent revision surgery. These initial results are promising and support further prospective studies of a broader group of cases, reinforcing the role of vSPECT/CT as valuable aid in managing PAL.

Pipe Crawler Robot

This pipe inspection robot aims at detecting the exact location of leakage and clearing the blockages, thus removing human factors from labor-intensive and dangerous work, thereby reducing the number of accidents that happen due to the lack of regular inspection. Detection Pipe leakage, water distribution system. An autonomous pipe crawler robot is designed to navigate through pipelines, offering a practical and efficient solution for inspecting and monitoring the condition of pipe interiors. This robot leverages advanced locomotion mechanisms and control strategies to maneuver through various pipe diameters and configurations, including straight sections, bends, and junctions. Equipped with sensors and cameras, it can detect anomalies such as cracks, corrosion, blockages, and leaks, providing real-time, high-resolution imagery and data. This technology significantly reduces the need for manual inspections, minimizes human exposure to hazardous environments, and improves the precision and efficiency of pipeline maintenance operations. The development of pipe crawler robots represents a significant advancement in the field of robotics and infrastructure management, promising enhanced safety, reliability, and cost-effectiveness for industries reliant on extensive pipeline networks. Key words: inspection, autonomous, locomotion, robot, crawler., etc.

Convolutional neural network for leak location in buried pipes of underground water supply

Convolutional neural network for leak location in buried pipes of underground water supply

Identification research of chemical process leakage based on deep learning and correlation-distance graph coding

As a leak identification method in chemical processes, deep learning is limited by inadequate sample labels and highly coupled, nonlinear process variables, therefore difficult to model with a high identification accuracy. To address these problems, a fusion correlation-distance graph (CDG) coding aided deep learning method is proposed. First, the mechanism model is built based on dynamic process data with different leakage labels. Then, the weighted degree and spatial distribution of the dynamic dataset are computed through innovatively designed correlation and distance coding, aided by sliding window fusion into a correlation-distance matrix. Finally, the well-trained long short-term memory (LSTM) is constructed to learn the deep features of the leakage process and identify them. The applications in two industrial cases, CO2 capture and ammonia synthesis, show that CDG-LSTM presents state-of-the-art identification performance with R2 scores of 0.994 and 0.986. 6.6% and 18.4% improvement over LSTM. In addition, the proposed method effectively categorizes the unknown leak samples into known labels with nearby locations, providing a valuable reference for operators to determine the unknown leak locations.

Detection of Dural Defect Localization Using 4-Dimensional Dynamic Computed Tomography Myelography for Patients with Superficial Siderosis

There are cases of superficial siderosis (SS) with spinal ventral fluid-filled collection in the spinal canal. In our previous study, the balanced steady-state free precession sequence magnetic resonance imaging is useful in identifying the location of dural defects. However, because of its narrow scan area and long scan time, it cannot easily detect the defect location in some patients with small dural defect. In this study, we applied 4-dimensional (4D) dynamic computed tomography (CT) imaging, including time-axis imaging, to myelography using the latest CT imaging equipment, which can perform short-time continuous imaging, to identify the dural defect site. Twenty SS patients with ventral fluid-filled collection in the spinal canal (9 males, 11 females; mean age 61.6years) underwent 4D dynamic CT myelography. A 192-row helical CT (SOMATOM Force, SIEMENS, Munich, Germany) with high-speed scanning capability was used to obtain 9-11 scans per minute at low dose while passing contrast medium into the subarachnoid space. Then, contrast leakage sites were identified. The contrast leakage sites could be identified in all 20 cases: C7/Th1, 2 cases; Th1/2, 5 cases; Th2/3, 9 cases; Th3/4, 1 case; Th5/6, 1 case; Th7/8, 1 case; and Th8/9, 1 case. Eighteen cases underwent surgical operation, and actual dural defects were confirmed at the contrast leakage sites. The mean±standard deviation of leakage time from contrast agent injection was 19.0±9.2s. The 4D dynamic CT myelography can be used to reliably identify the location of spinal fluid leakage. In SS cases, dural defects could be visualized in an average of 19seconds.

A Survey and Study of Signal and Data-Driven Approaches for Pipeline Leak Detection and Localization

A Survey and Study of Signal and Data-Driven Approaches for Pipeline Leak Detection and Localization