Development of a Mechanistic and Data-Driven Model for Multiphase Flow Leak Detection in Pipeline

The detection and location of pipeline leakage can be deduced from the time arrival leak signals measured by acoustic sensors placed at the pipe. Ongoing research in this field is primarily focused on refining techniques for accurately estimating the time delays. This enhancement predominantly revolves around the application of advanced signal processing methods. Additionally, researchers are actively immersed in the utilization of machine learning approaches on vibro-acoustic data files, to determine the presence or absence of leaks. Less attention has been given to evaluating the sensitivity, performance, and overall effectiveness of these sensors in leak detection; although acoustic methods have been successfully used for leak detection in metallic pipes, they are less effective in plastic pipes due to the high attenuation of leak noise signals. The primary thrust of this research centers on identifying sensors that not only possess sensitivity but also exhibit high efficiency. To accomplish this goal, we conducted an exhaustive evaluation of the performance of three distinct categories of acoustic sensors employed for detecting water leaks in plastic pipes: specifically, lead zirconate titanate (PZT) sensors, polyvinylidene fluoride (PVDF) sensors, and surface acoustic wave (SAW) sensors. Our evaluation encompassed the performance of PVDF and SAW sensors in leak detection, comparing them to PZT sensors under a variety of conditions, including different leak sizes, flow rates, and distances from the leak. The results showed that all three sensors, when they were placed in the same position, were able to detect water leaks in plastic pipes with different sensitivities. For small leaks (1 mm, 2 mm), the PVDF sensor showed the greatest sensitivity (0.4 dB/L/h, 0.33 dB/L/h), followed by the SAW sensor (0.16 dB/L/h, 0.14 dB/L/h), and finally the PZT (0.13 dB/L/h, 0.12 dB/L/h). Similarly, for larger leaks (4 mm, 10 mm), the PVDF sensor continued to show superior sensitivity (0.2 dB/L/h, 0.17 dB/L/h), followed by the SAW sensor (0.13 dB/L/h, 0.11), and finally the PZT sensor (0.12 dB/L/h, 0.1 dB/L/h), outperforming the PZT sensor. This suggests that SAW and PVDF sensors, have the potential to serve as valuable, cost-effective alternatives to traditional commercial leak noise transducers. The outcomes of this comparative study involving three acoustic sensors hold the potential to advance the development of robust and dependable systems for the detection of water leaks in plastic pipelines.

Digital twins are transforming the digitalization and automation of offshore gas pipeline systems by enabling realtime monitoring, predictive maintenance, and operational efficiency. This study introduces a novel adaptive digital twin framework designed for leak detection and localization in offshore gas pipelines. The framework integrates OLGA-generated synthetic data, validated experimental results, and advanced machine learning (ML) techniques, including transfer learning and ensemble models. The proposed framework achieves a classification accuracy of 98.2% for leak detection, with a mean absolute error (MAE) of 0.11 cm for leak size prediction and a mean absolute percentage error (MAPE) of 3.8% for leak localization. A core innovation of this framework is the calibration methodology, which recalibrates dimensionless nomographs and leak detection correlations for seamless adaptation to new pipeline geometries and operating conditions. Through systematic steps, the calibrated correlations predict leak size and location with high accuracy, leveraging pressure drop and mass flow difference data. Additionally, ML-driven models enable efficient generation of new nomographs for pipelines with varying configurations, enhancing scalability and reducing computational effort. The real-time implementation enables predictions with a latency of less than 2 seconds, significantly outperforming conventional methods in speed and accuracy. Also, the framework’s adaptability, supported by its digital twin visualization and real-time feedback mechanisms, significantly improves pipeline integrity management, operational safety, and environmental protection. The study demonstrates the framework’s robustness in handling complex flow dynamics and offers a scalable solution to enhance the digital transformation of offshore oil and gas operations.

The effective detection and prevention of CO2 leakage in active injection wells are paramount for safe carbon capture and storage (CCS) initiatives. This study assesses five fundamental machine learning algorithms, namely, Support Vector Regression (SVR), K-Nearest Neighbor Regression (KNNR), Decision Tree Regression (DTR), Random Forest Regression (RFR), and Artificial Neural Network (ANN), for use in developing a robust data-driven model to predict potential CO2 leakage incidents in injection wells. Leveraging wellhead and bottom-hole pressure and temperature data, the models aim to simultaneously predict the location and size of leaks. A representative dataset simulating various leak scenarios in a saline aquifer reservoir was utilized. The findings reveal crucial insights into the relationships between the variables considered and leakage characteristics. With its positive linear correlation with depth of leak, wellhead pressure could be a pivotal indicator of leak location, while the negative linear relationship with well bottom-hole pressure demonstrated the strongest association with leak size. Among the predictive models examined, the highest prediction accuracy was achieved by the KNNR model for both leak localization and sizing. This model displayed exceptional sensitivity to leak size, and was able to identify leak magnitudes representing as little as 0.0158% of the total main flow with relatively high levels of accuracy. Nonetheless, the study underscored that accurate leak sizing posed a greater challenge for the models compared to leak localization. Overall, the findings obtained can provide valuable insights into the development of efficient data-driven well-bore leak detection systems.

The key to leak detection and location in water supply pipelines is signal denoising and feature extraction. First, in this paper, an improved spline-local mean decomposition (ISLMD) is proposed to eliminate noise interference. Based on the ISLMD decomposition of a signal, the cross-correlation function between the reference signal and the product functions component can be obtained. And then the PF component containing the leak information can be extracted reasonably. Compared with improved local mean decomposition, the ISLMD has higher accuracy in leak location. Second, an image recognition method using a convolutional neural network for leak detection is proposed, which can better address the problem that the features of different leak apertures or locations are highly similar to each other. The images from the conversion of the reconstructed signals are used as the input of the AlexNet model, which is capable of adaptive extraction of leak signal features. The trained AlexNet model can effectively detect different leak apertures. Finally, the signal time-delay between the upstream and downstream pressure transmitters caused by the leak and propagation of negative pressure wave is determined according to generalized cross-correlation analysis, and thereby, the leak location is obtained. The experimental results show that the proposed method is effective for leak detection and location.

In this paper, an approach to perform leak state detection and size identification for industrial fluid pipelines with an acoustic emission (AE) activity intensity index curve (AIIC), using b-value and a random forest (RF), is proposed. Initially, the b-value was calculated from pre-processed AE data, which was then utilized to construct AIICs. The AIIC presents a robust description of AE intensity, especially for detecting the leaking state, even with the complication of the multi-source problem of AE events (AEEs), in which there are other sources, rather than just leaking, contributing to the AE activity. In addition, it shows the capability to not just discriminate between normal and leaking states, but also to distinguish different leak sizes. To calculate the probability of a state change from normal condition to leakage, a changepoint detection method, using a Bayesian ensemble, was utilized. After the leak is detected, size identification is performed by feeding the AIIC to the RF. The experimental results were compared with two cutting-edge methods under different scenarios with various pressure levels and leak sizes, and the proposed method outperformed both the earlier algorithms in terms of accuracy.

Pipeline leak detection has always been a relatively difficult technical problem; especially in urban pipeline leak detection, there are still many problems to be solved. A multipoint leak detection and location method for urban gas pipelines based on variational mode decomposition and relative entropy was proposed. Firstly, the experiment pipeline system was built, and the original signal was collected by acoustic emission technology; then, a variational model method was used to decompose the signal to obtain multiple intrinsic mode function (IMF) components with different characteristic scales. According to the characteristics of relative entropy, each component was analyzed, the appropriate IMF component was selected, and the selected component was reconstructed to obtain the observation signal. The multipoint leakage location model of the urban gas pipeline was established. The number of source signals was estimated by singular value decomposition, and the leakage signals were separated; finally, the accurate location of leakage point was achieved by cross-correlation positioning. The results showed that the average relative error of the pipeline leak location results is 2.97%, and the leak location accuracy is significantly improved, achieving the purpose of precise location.

Abstract. In August and September 2020, three different measurement methods for quantifying methane (CH4) emissions from leaks in urban gas distribution networks were applied and compared in Hamburg, Germany: the “mobile”, “tracer release”, and “suction” methods. The mobile and tracer release methods determine emission rates to the atmosphere from measurements of CH4 mole fractions in the ambient air, and the tracer release method also includes measurement of a gaseous tracer. The suction method determines emission rates by pumping air out of the ground using soil probes that are placed above the suspected leak location. The quantitative intercomparison of the emission rates from the three methods at a small number of locations is challenging because of limitations of the different methods at different types of leak locations. The mobile method was designed to rapidly quantify the average or total emission rate of many gas leaks in a city, but it yields a large emission rate uncertainty for individual leak locations. Emission rates determined for individual leak locations with the tracer release technique are more precise because the simultaneous measurement of the tracer released at a known rate at the emission source eliminates many of the uncertainties encountered with the mobile method. Nevertheless, care must be taken to properly collocate the tracer release and the leak emission points to avoid biases in emission rate estimates. The suction method could not be completed or applied at locations with widespread subsurface CH4 accumulation or due to safety measures. While the number of gas leak locations in this study is small, we observe a correlation between leak emission rate and subsurface accumulation. Wide accumulation places leaks into a safety category that requires immediate repair so that the suction method cannot be applied to these larger leaks in routine operation. This introduces a sampling bias for the suction method in this study towards the low-emission leaks, which do not require immediate repair measures. Given that this study is based on random sampling, such a sampling bias may also exist for the suction method outside of this study. While an investigation of the causal relationship between safety category and leak size is beyond the scope of this study, on average higher emission rates were observed from all three measurement-based quantification methods for leaks with higher safety priority compared to the leaks with lower safety concern. The leak locations where the suction method could not be applied were the biggest emitters, as confirmed by the emission rate quantifications using mobile and tracer methods and an engineering method based on the leak's diameter, pipeline overpressure, and depth at which the pipeline is buried. The corresponding sampling bias for the suction technique led to a low bias in derived emission rates in this study. It is important that future studies using the suction method account for any leaks not quantifiable with this method in order to avoid biases, especially when used to inform emission inventories.

In this paper, a Random Forest classifier was used to predict leak locations for two differently sized water distribution networks based on pressure sensor measurements. The prediction model is trained on simulated leak scenarios with randomly chosen parameters - leak location, leak size, and base node demand uncertainty. Leak localization methods found in literature that rely on numerical simulations can only predict network nodes as leak nodes; however, since a leak can occur at any point along a pipe segment, additional spatial discretization of suspect pipe is proposed in this paper. It was observed that pipe segmentation of the whole network is a non-feasible approach since it rapidly increases the number of potential leak locations, consequently increasing the complexity of the prediction model. Therefore, a novel approach is proposed, in which a prediction model is trained on scenarios with leaks occurring in original network nodes only, but with its accuracy assessed against pressure sensor measurements from scenarios in which leaks occur in points between network nodes. It was observed that this approach can successfully narrow down the suspect leak area and, followed by additional segmentation of that network area and subsequent prediction, a precise leak localization can be achieved. The proposed approach enables incorporation of various uncertainties by simulating leak scenarios under different conditions. Investigation of leak size uncertainty and base demand variation showed that several different scenarios can produce similar sensor measurements which makes it difficult to unambiguously determine leak location using the prediction model. Therefore, future approaches of coupling prediction modeling with optimization methods are proposed.

To determine the influence of anastomotic crotch suture augmentation on leakage pressures and leakage location following intestinal functional end-to-end stapled anastomosis (FEESA) in dogs. Ex vivo, randomized, experimental. Chilled jejunal segments from 3 adult dogs. Jejunal specimens were tested within 24 hours of collection. A FEESA was performed and randomly assigned to 1 of 4 treatment groups (n=12/group): (1) no crotch suture (NCS); (2) simple interrupted crotch suture (SICS); (3) two simple interrupted crotch sutures (TCS) placed laterally on opposing jejunal limbs; (4) simple continuous crotch suture (SCCS) augmentation. Crotch sutures were performed using 3-0 USP polydioxanone. Initial (ILP) and maximal (MLP) leakage pressures (Mean ± SD mm Hg) and leakage location were recorded and compared between groups. Initial leakage pressure was greater after placement of TCS (37.8 ± 6.4, P < .039) and SCCS (47.6 ± 11.0, P < .002) than NCS (27.1 ± 2.5) and SICS (33.0 ± 6.0). Maximal leakage pressure was greater in specimens including SICS, TCS, and SCCS than those without crotch suture augmentation (P < .043). Leakage occurred at the anastomotic crotch in 8/12 NCS, 6/12 SICS, 11/12 TCS, and 12/12 SCCS constructs (P < .001). Augmentation of FEESA with TCS and SCCS increased ILP and decreased the occurrence of leakage from the anastomotic crotch, while all methods of anastomotic crotch augmentation increased MLP. Augmenting the FEESA with crotch suture(s) improved the resistance of the jejunal anastomosis to leakage in normal cadaveric segments. Placing 2 crotch sutures or use of a simple continuous pattern for anastomotic augmentation appeared to be superior to the placement of a single suture.

This paper presents a methodology for leak detection and location in a great longitude pipeline, which belongs to E.P. PETROECUADOR, through the combination of line balance and speed propagation of the negative pressure wave. Two different approaches, slope projection and a first order dead time (FODT) model, are used for computing the pressure change time, which in combination with speed propagation of the sound is used for estimating leak location. A better performance in leak location accuracy than a commercial installed system is obtained, whose results are compared.

There is an increasing need for timely pipeline leak detection and localization methods, pipeline leak could lead to not only the loss of the goods but also considerable environmental and economic problems. With the rapid development of hardware and software, the pipeline leak detection and localization algorithms have been widely researched and applied in many Fields. However, traditional methods are usually limited by extracting features manually, which is inefficient and time-consuming. Convolutional neuron network is an effective method to extract features automatically. In this paper, a novel ensemble transfer learning one-dimension convolutional neural network (TL1DCNN) for the pipeline leak detection and localization is proposed. The TL1DCNN plays the role of base learner. The results of a set of obtained base learners are integrated to achieve the task of pipeline leak detection and localization. Firstly, one-dimension convolutional neural network (1DCNN) models with different parameters are pretrained with source domain data. A small learning rate is set to retrain the above 1DCNN models for target task with target domain data in order to obtain TL1DCNN base learners. Then, the four ensemble strategies with different number base learners whose ensemble weights are optimized by particle swarm optimization algorithm are obtained by minimizing the sum of similarity. The dataset simulated by pipeline network model is used to evaluate the effectiveness of the proposed approach. The indicators such as classification accuracy, precision, recall, F_score and confusion matrix are used to compare the proposed approach with traditional methods and other deep learning methods. The experimental results show that the performance of the proposed approach is superior to other compared methods.

Flow instability in parallel-channel systems with ultra-supercritical water

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. Negative pressure wave method is the method which is used widely on pipeline leakage detection and location internationally at present. It is based on signal processing. And the method doesn't need to establish hydraulic and mathematical model of pipeline, and most only use pressure signal. So the method adapts to the application of pipeline in China(1). Because oil pipeline has the function of guiding wave, the negative pressure wave caused by leakage can transmit more than tens of kilometers along the pipeline. Its propagation velocity is equal to the propagation velocity of sound wave transferring fluid in pipeline. And the propagation velocity of negative pressure wave in the original oil pipeline is between 1000meters and 1200 meters. As negative pressure wave has the characteristics of remote transmission, pressure transducer installed at the end and first of the pipeline can be used to capture negative pressure wave containing leakage information and to detect the leakage. After identifying leakage happens, leakage location can be made according to time difference of negative pressure wave caused by leakage transmitting to the head and end of the pipeline and negative pressure wave velocity(2).

Numerical and experimental analyses were carried out to investigate the static characteristics of liquid annular seals with helical grooves in a seal stator. In the numerical analysis, the momentum equations with turbulent coefficients and the continuity equation, which were averaged across the film thickness, were numerically solved to obtain the leakage flow rate and the pressure distributions in the seal clearance. To accurately define the location of the step between the groove and the land regions in the calculation domain, these governing equations were expressed using an oblique coordinate system in which the directions of coordinate axes coincided with the circumferential direction and the direction along the helical grooves. The numerical analysis included the effects of both fluid inertia and energy loss due to expansion during the passage of fluid from the land region to the helical groove region and that due to contraction from the groove region to the land region. In the experimental analysis, the leakage flow rate and the fluid-film pressure distributions in the seal clearance were measured for the helically grooved seals with different helix angles of the helical groove. The numerical results of leakage flow rate and pressure distributions agree reasonably with the experimental results, which demonstrates the validity of the numerical analysis. The leakage flow rate of the helically grooved seals was influenced by two factors: fluid energy loss during passage through the step between the groove and the land, and the pumping effect by which the spinning motion of the rotor pushes the flow back upstream along the helical grooves. Under a low range of rotor spinning velocity, the leakage flow rate decreased with helix angle because the effect of fluid energy loss in the steps was significant. By contrast, under a high range of spinning velocity, the quantitative difference in the leakage flow rate due to the helix angle decreased compared to that under a low range because the reduction in the leakage flow rate due to the pumping effect was pronounced for a larger helix angle. The effects of helix angle and rotor spinning velocity on the leakage flow rate are explained qualitatively using a simplified model.

Pipeline leak detection using an acoustic emission (AE) method requires highly sensitive transducers responding to less attenuative and dispersive wave motion in order to place the discrete transducer spacing in an acceptable approach. In this paper, a new piezoelectric transducer geometry made of PZT-5A is introduced to increase the transducer sensitivity to the tangential direction. The finite element analysis of the transducer geometry is modeled in the frequency domain to identify the resonant frequency, targeting 60 kHz, and the loss factor. The numerical results are compared with the electromechanical characterization tests. The transducer response to wave motion generated in different directions is studied using a multiphysics model that couples mechanical and electrical responses of structural and piezoelectric properties. The directional dependence and the sensitivity of the transducer response are identified using the laser-induced load function. The transducer response is compared with a conventional thickness mode AE transducer under simulations and leak localization in a laboratory scale steel pipe.