Search

In order to reduce the probability of leakage and improve detection accuracy, multiple radar equipment is required to perform multiple detections on the same airspace. Due to the overlap of airspace, there are multiple local tracks reported by various radar equipment that belong to the same target. The local tracks reported by each radar require track fusion to form a system track, and track association is a prerequisite for track fusion. This paper proposes a multi-target and multi-sensor track association analysis method, which can distinguish whether local tracks come from the same target and associate local tracks with the same target, providing conditions for track fusion to form system tracks. This method divides track association into two steps: Coarse association and fine association, which greatly improves the accuracy of track association. [ABSTRACT FROM AUTHOR]

Leaks are a constant problem in water distribution systems, resulting in wasted resources, environmental impacts, and financial losses. Thus, it is crucial to develop effective and agile methods to detect network leaks. In this context, this study proposes a leak detection methodology using three different processes. The first consists of treating monitoring data through independent component analysis, and the other two detection processes use the interquartile range (IQR) and matrix profile (MP) techniques, respectively. The methodology is evaluated based on a set of benchmark data. The results indicate that the proposed approach is effective in detecting leaks, with some cases being detected in a few minutes after the beginning of the leak. It is worth mentioning that the IQR method presented better performance in detecting leaks with abrupt onset, whereas the MP method was more efficient in leaks with gradual increase in flow. In summary, the proposed methodology offers a robust and promising approach for fast and accurate leak detection in water distribution networks. [ABSTRACT FROM AUTHOR]

The problem of localizing two leaks using only flow rate and pressure measurements at the boundaries of a pipeline, and under steady-state flow conditions, is ill-posed due to the undetermined nature of the inverse problem, which involves two coupled equations with four unknowns associated with the presence of the leaks: two emitter coefficients and two locations. Therefore, attempting to solve this problem using any method without imposing additional constraints leads to an unbounded solution space, which contains solutions that may be physically meaningless. In this article, we propose a four-algorithm method that incorporates both spatial and hydraulic constraints to filter out physically infeasible solutions. The proposed method is based on Monte Carlo simulations, which use input data from hydraulic instruments installed at the boundaries of the pipeline, as well as random values with predefined probabilities that are bounded by the hydraulic-spatial constraints. The outputs of the method are probability distributions for the four unknowns. To demonstrate the feasibility of the method, results obtained through simulations and experimental testing on a test bed are presented. • A four-algorithm method based on Monte Carlo simulations to localize two leaks in a pipeline. • The method incorporates spatial and hydraulic constraints to eliminate physically infeasible localization estimations. • Tests results using both synthetic and experimental data from a test bed. [ABSTRACT FROM AUTHOR]

The quickly developing drone technology can be used efficiently in the field of pipeline leak detection. The aim of this article is to provide drone mission concepts for detecting releases from pipelines. It provides an overview of the current applications of natural gas pipeline surveys, it considers environmental conditions by plume modelling, it discusses suitable commercially available sensors, and develops concepts for routine monitoring of pipelines and short term missions for localising and identifying a known leakage. Suitable platforms depend on the particular mission and requirements concerning sensors and legislation. As an illustration, a feasibility study during a release experiment is introduced. The main challenge of this study was the variability of wind direction on a time scale of minutes, which produces considerable differences to the plume simulations. Nevertheless, the leakage rates derived from the observations are in the same order of magnitude as the emission rates. Finally the results from the modeling, the release experiment and possible drone scenarios are combined and requirements for future application derived. [ABSTRACT FROM AUTHOR]

Hydrogen (H2) is positioned as a key solution to the decarbonization challenge in both the energy and transportation sectors. While hydrogen is a clean and versatile energy carrier, it poses significant safety risks due to its wide flammability range and high detonation potential. Hydrogen leaks can occur throughout the hydrogen value chain, including production, storage, transportation, and utilization. Thus, effective leak detection systems are essential for the safe handling, storage, and transportation of hydrogen. This review aims to survey relevant codes and standards governing hydrogen-leak detection and evaluate various sensing technologies based on their working principles and effectiveness. Our analysis highlights the strengths and limitations of the current detection technologies, emphasizing the challenges in achieving sensitive and specific hydrogen detection. The results of this review provide critical insights into the existing technologies and regulatory frameworks, informing future advancements in hydrogen safety protocols. [ABSTRACT FROM AUTHOR]

Magnetic flux leakage (MFL) inspection employs leakage magnetic fields to effectively detect and locate pipeline defects. The spacing between magnetic poles significantly affects the leakage magnetic field strength. While most detectors typically opt for moderate pole spacing for routine detection, this study investigates the propagation characteristics of MFL signals at small pole spacings (under specimen oversaturated magnetization) and their impact on MFL detection. Through finite element simulation and experiments, it reveals a new signal phenomenon in the radial MFL signal By at small pole spacings, the double peak–valley (DPV) phenomenon, characterized by outer and inner peaks and valleys. Theoretical analysis based on the simulation results elucidates the mechanisms for this DPV phenomenon. Based on this, the impact of defect size, pipe wall thickness, and magnetic pole and rigid brush height on MFL signals under small magnetic pole spacings is examined. It is demonstrated that, under a smaller magnetic pole spacing, a potent background magnetic field manifests in the air above the defect. This DPV phenomenon is generated by the magnetic diffusion and compression interactions between the background and defect leakage magnetic fields. Notably, the intensity of the background magnetic field can be mitigated by reducing the height of the rigid brush. In contrast, the pipe wall thickness and magnetic pole height exhibit a negligible influence on the DPV phenomenon. The emergence of the DPV precipitates a reduction in the peak-to-valley difference within the MFL signal, constricting the depth range of detectable defects. However, the presence of DPV increases the identification of defects with smaller opening sizes. These findings reveal the characterization of the MFL signal under small pole spacing, offering a preliminary study on identifying specific defects using unconventional signals. This study provides valuable guidance for MFL detection. [ABSTRACT FROM AUTHOR]

In modern industries, pipelines play a crucial role, both as an essential element in energy transportation (water, gas and electricity) and also in the distribution of these resources. The large size of piping infrastructures, their age and unpredictable external factors are the main difficulties in monitoring the piping system. In this context, the detection and the localization of leaks are challenging but essential, as leaks lead to substantial economic losses. Current methods have many limitations, involving invasive procedures, working only with short pipes or requiring a system shutdown. This paper presents a non-intrusive method based on acoustic signal processing. Leak detection is performed using matched filters, while localization is performed based on the phase diagram representation method and diagram-based entropy computation. Our continuous monitoring system was used for two months and a full comparison with the video inspection-based technique was conducted. The results indicate that this method has a high accuracy, regardless of the length of the pipe. [ABSTRACT FROM AUTHOR]

The detection of gas leaks using acoustic signals is often compromised by environmental noise, which significantly impacts the accuracy of subsequent leak identification. Current noise reduction algorithms based on non-negative matrix factorization (NMF) typically utilize the Euclidean distance as their objective function, which can exacerbate noise anomalies. Moreover, these algorithms predominantly rely on simple techniques like Wiener filtering to estimate the amplitude spectrum of pure signals. This approach, however, falls short in accurately estimating the amplitude spectrum of non-stationary signals. Consequently, this paper proposes an improved non-negative matrix factorization (INMF) noise reduction algorithm that enhances the traditional NMF by refining both the objective function and the amplitude spectrum estimation process for reconstructed signals. The improved algorithm replaces the conventional Euclidean distance with the Kullback–Leibler (KL) divergence and incorporates noise and sparse constraint terms into the objective function to mitigate the adverse effects of signal amplification. Unlike traditional methods such as Wiener filtering, the proposed algorithm employs an adaptive Minimum Mean-Square Error-Log Spectral Amplitude (MMSE-LSA) method to estimate the amplitude spectrum of non-stationary signals adaptively across varying signal-to-noise ratios. Comparative experiments demonstrate that the INMF algorithm significantly outperforms existing methods in denoising leakage acoustic signals. [ABSTRACT FROM AUTHOR]

Fuel leakage detection is crucial for solid oxide fuel cells (SOFCs) as fuel leakage significantly reduces the reliability and lifetime of SOFC devices. In this study, fuel leakage detection using voltage signals is thoroughly investigated. Rectangular SOFCs with four segmented cathodes are prepared and tested with fuel leakage along the seals. Voltammetric tests show that the ring-shaped area on the current-voltage curve is primarily caused by a delay of gas diffusion. Such a ring-shaped area is exacerbated by fuel leakage but is not sufficient as an indicator for detecting fuel leakage. Additionally, the open-circuit voltages measured at all segmented cathodes are well below the theoretical potential of the fuel gas. This is because a localized steam electrolysis circuit is formed when fuel leakage occurs. The measured open-circuit voltage is the potential difference between the two interconnectors in the closed electrolysis circuit. This work contributes to the accurate detection of fuel leakage and the commercialization of SOFCs. [Display omitted] • OCVs tested with fuel leakage are well below equilibrium potential. • Fuel leakage causes localized steam electrolysis. • The ring-shaped area on I–V curves results from a delay in gas diffusion. • Ring-shaped areas are not sufficient as indicators of fuel leakage. [ABSTRACT FROM AUTHOR]

Creating an effective deep learning technique for accurately diagnosing leak signals across diverse environments is crucial for integrating artificial intelligence (AI) into the power plant industry. We propose an automatic weight redistribution ensemble model based on transfer learning (TL) for detecting leaks in diverse power plant environments, overcoming the challenges of site-specific AI methods. This innovative model processes time series acoustic data collected from multiple homogeneous sensors located at different positions into three-dimensional root-mean-square (RMS) and frequency volume features, enabling accurate leak detection. Utilizing a TL-driven, two-stage learning process, we first train residual-network-based models for each domain using these preprocessed features. Subsequently, these models are retrained in an ensemble for comprehensive leak detection across domains, with control weight ratios finely adjusted through a softmax score-based approach. The experiment results demonstrate that the proposed method effectively distinguishes low-level leaks and noise compared to existing techniques, even when the data available for model training are very limited. [ABSTRACT FROM AUTHOR]

Anomaly detection systems based on artificial intelligence (AI) have demonstrated high performance and efficiency in a wide range of applications such as power plants and smart factories. However, due to the inherent reliance of AI systems on the quality of training data, they still demonstrate poor performance in certain environments. Especially in hazardous facilities with constrained data collection, deploying these systems remains a challenge. In this paper, we propose Generative Anomaly Detection using Prototypical Networks (GAD-PN) designed to detect anomalies using only a limited number of normal samples. GAD-PN is a structure that integrates CycleGAN with Prototypical Networks (PNs), learning from metadata similar to the target environment. This approach enables the collection of data that are difficult to gather in real-world environments by using simulation or demonstration models, thus providing opportunities to learn a variety of environmental parameters under ideal and normal conditions. During the inference phase, PNs can classify normal and leak samples using only a small number of normal data from the target environment by prototypes that represent normal and abnormal features. We also complement the challenge of collecting anomaly data by generating anomaly data from normal data using CycleGAN trained on anomaly features. It can also be adapted to various environments that have similar anomalous scenarios, regardless of differences in environmental parameters. To validate the proposed structure, data were collected specifically targeting pipe leakage scenarios, which are significant problems in environments such as power plants. In addition, acoustic ultrasound signals were collected from the pipe nozzles in three different environments. As a result, the proposed model achieved a leak detection accuracy of over 90% in all environments, even with only a small number of normal data. This performance shows an average improvement of approximately 30% compared with traditional unsupervised learning models trained with a limited dataset. [ABSTRACT FROM AUTHOR]

This study proposes a YOLOv5 inspection model based on polariser filtering (PF) to improve the recognition accuracy of the machine vision inspection model for tea leaf shoots when operating under intense outdoor light. To study the influence of the polariser parameters on the quality of the tea shoot image datasets, we improved the YOLOv5 algorithm module, inputted the results obtained from the spatial pyramid pooling structure in the backbone module into the neck module, set the up-sampling link of the neck module as a low-level feature alignment (LFA) structure, and used a bounding box similarity comparison metric based on the minimum point distance (mpdiou) to improve the accuracy of the YOLOv5 detection model. The mpdiou loss function is used to replace the original loss function. Experimental results show that the proposed method can effectively address the impact of intense outdoor light on tea identification, effectively solving the problem of poor detection accuracy of tea buds in the top view state. In the same identification environment, the model mAP50 value increased by 3.3% compared to that of the existing best mainstream detection model, and the mAP50-90 increased by 3.1%. Under an environment of light intensity greater than 5 × 10 4   L u x , the proposed YOLOv5s+LFA+mpdiou+PF model reduced the leakage detection rate by 35% and false detection rate by 10% compared to that with YOLOv5s alone. [ABSTRACT FROM AUTHOR]

Deep Learning-based Side-Channel Analysis (DL-SCA) has emerged as a powerful method in the field of side-channel analysis. Current works on DL-SCA primarily rely on publicly available datasets, which typically consist of well-organized and well-aligned training and attack sets. However, this disregards the challenges faced in real-world attacks, where the attack traces are not well-aligned with the training traces as attackers have different levels of control over profiling and attack devices. A network that is capable of identifying areas of leakage and subsequently predicting the leaked values can bypass such difficulty. Therefore, we proposed Arbitrary Trace Attacks, which are placed under the flexible scenario that provides training traces and attack traces with arbitrary sizes. To implement such attacks, we present the Arbitrary Convolutional Neural Network (ACNN), which scans the input trace of arbitrary sizes for leakage area identification and leakage value prediction using a sliding window. Experimental evaluation is conducted on two datasets DPAv4.2 and ASCAD to verify the effectiveness of our approach on unprotected and masked implementation respectively. As a result, the target leakage areas are detected with a significant frequency and the key recovery performance is on par with state-of-the-art. Moreover, the trained model shows the potential for detecting leakage in a general context, that is, detecting leakage of key bytes other than the target one. [ABSTRACT FROM AUTHOR]

Maximizing the net economic benefits of leak management programs provides the most value to water utilities and their customers. Leak management programs with alternative objectives (e.g., to "break even" or to maximize leak reduction) are not necessarily the most efficient use of resources. Combining hydraulic modeling with principles of economics and using a real distribution system to provide a case study, we present a methodology for maximizing the net economic benefits of a leak management program consisting of leak detection and repair activities and pressure management interventions. The optimization results include recommended leak detection and repair intervals for individual district metered areas, optimal siting of both pressure reducing valves and pumps-as-turbines (PATs), and modifications to the distribution system's connectivity to isolate high-pressure areas. The case study results indicate that 10% to 11% of the system's current energy use can be recovered by PATs and that the avoided leakage over the study period is equivalent to 13% to 23% of the system's current customer demand. While leak detection and repair and pressure management are more commonly considered independently, this analysis demonstrates the feasibility of including their interdependencies when optimizing a leak management program, resulting in more useful recommendations. [ABSTRACT FROM AUTHOR]

The use of temperature field to detect leakage in earth-rock dams has a low cost and high efficiency, which is crucial to the safety of dams. This paper briefly introduces the principle of temperature field leak detection, reviews the research results of temperature leak detection, and categorises the research on the detection of leakage through the temperature field of earth-rock dam into three levels: 1) technical research based on temperature field detection; 2) study on the theoretical model of temperature detection; 3) research on numerical calculations of multi-field coupling such as temperature field and seepage field. The main achievements and characteristics of various research levels are reviewed, revealing the development trends of the field of temperature field leakage detection in earth-rock dams. Consequently, UAVs mounted with infrared thermal imaging leak detection technology are integrated, and the technical scheme for patrol leak detection is noted. Finally, future research directions for the temperature field detection of earth-rock dam leakage are explored for the timely and accurate detection of the location of leakage channels and to establish an intuitive and effective judgement and identification method for dam leakage. [ABSTRACT FROM AUTHOR]

Side channels are unintended pathways within target systems that leak internal target information. Side-channel sensing (SCS) is the process of exploiting side channels to extract embedded target information. SCS is well established within the cybersecurity (CYB) domain, and has recently been proposed for medical diagnostics and monitoring (MDM). Remaining unrecognised is its applicability to human–computer interaction (HCI), among other domains (Misc). This article analyses literature demonstrating SCS examples across the MDM, HCI, Misc, and CYB domains. Despite their diversity, established fields of advanced sensing and signal processing underlie each example, enabling the unification of these currently otherwise isolated domains. Identified themes are collating under a proposed domain-agnostic SCS framework. This SCS framework enables a formalised and systematic approach to studying, detecting, and exploiting of side channels both within and between domains. Opportunities exist for modelling SCS as data structures, allowing for computation irrespective of domain. Future methodologies can take such data structures to enable cross- and intra-domain transferability of extraction techniques, perform side-channel leakage detection, and discover new side channels within target systems. [ABSTRACT FROM AUTHOR]

In this subsection, a simulation model of underground pipeline leaks was created. Modeling the flow fields of overhead pipelines and underground pipelines with different soil porosity was conducted. The influence of the underground environment and soil porosity on the pipeline leakage field was analyzed. The influence of changes in the underground environment and soil porosity on the loss of kinetic energy at gas leakage was defined. Thus, the law of change in the characteristics of the sound signal of a leak on the pipe wall was obtained. Based on the results of study, it was determined the area of damage to a gas pipeline which is depended on proportionality of the diameter of the gas jet. As the gas is released from a crack in the gas pipeline and with different porosity of the pipe material, the gas diffusion occurred. As a result, the explosion zone increased which ultimately the release of gases into the environment under the influence of pressure increased. [ABSTRACT FROM AUTHOR]

Poor sealing of gas extraction boreholes is one key to restrict gas extraction efficiency. In this paper, a novel borehole sealed quality detection device for the gas extraction of a coal mine is developed based on the theory of air leakage. By comparing the amount of gas extracted at different test points, it is possible to determine whether there is air leakage around the borehole, and the specific leakage position and leakage amount. Moreover, this device has the advantages of simple operation and a short test period. Based on the above analysis, a corresponding air leakage disposal method was proposed to handle the leaky boreholes. Field tests showed that the air–gas mixture flow in the test borehole was reduced by a factor of approximately 1.55 and the concentration of pure seam gas was increased approximately six times after the disposal of the air leaks. The combination of the leakage disposal method and the leakage detection device can accurately seal the borehole at the position of the leak, thus effectively ensuring the effectiveness of gas extraction from the borehole. The findings have important implications for improving the efficiency of gas extraction in coal mining operations. [ABSTRACT FROM AUTHOR]

Pipeline leakage seriously threatens the efficient and safe gas drainage in coal mines. To achieve the accurate detection and localization of gas drainage pipeline leakages, this study proposes a gas drainage pipeline leakage detection and localization approach based on the pressure gradient method. Firstly, the basic law of gas flow in the drainage pipeline was analyzed, and a pipeline network resistance correction formula was deduced based on the pressure gradient method. Then, a drainage pipeline model was established based on the realizable k - ε turbulence model, and the pressure and flow velocity distribution during pipeline leakage under different leakage degrees, leakage locations, and pipeline negative pressures were simulated and analyzed, thus verifying the feasibility of the pipeline leakage detection and localization method. It is concluded that the positioning errors of pipeline leakage points under different leakage degrees, different leakage positions, and different pipeline negative pressures were 0.88~1.08%, 0.88~1.49%, and 0.68~0.88%, respectively. Finally, field tests were conducted in the highly located drainage roadway 8421 of the Fifth Mine of Yangquan Coal Industry Group to verify the accuracy of the proposed pipeline leakage detection and localization method, and the relative error was about 8.2%. The results show that with increased pipeline leakage hole diameters, elevated pipeline negative pressures, and closer leakage positions to the pipeline center, the relative localization error was smaller, the localization accuracy was higher, and the stability was greater. The research results could lay the foundation for the fault diagnosis and localization of coal mine gas drainage pipeline networks and provide technical support for safe and efficient coal mine gas drainage. [ABSTRACT FROM AUTHOR]

This paper introduces a novel methodology that integrates low‐ and high‐frequency pressure monitoring in the parts of the urban water distribution network characterized by the largest number of repairing operations. This integrated approach aims to pinpoint the source and characterize potentially dangerous pressure variations. The actual behavior of a real operational water distribution network is considered as an example to verify the effectiveness of the proposed methodology. The analysis of the acquired pressure signals allows for mapping frequent pressure surges and identifying their source in the functioning conditions of a big user. Notwithstanding these pressure variations are not overly large, according to the literature, their near‐continuous nature poses a risk of inducing pipe breaks, and then leakage, due to fatigue. The proposed method integrating routine low‐frequency measurements with targeted high‐frequency surveys provides a cost‐effective, proactive management strategy. Key Points: A method is proposed for detecting transient source in Urban Water Distribution NetworksLow‐frequency pressure monitoring initially identifies the most pressure‐stressed section in a high‐maintenance areaHigh‐frequency pressure monitoring allows localizing the transient source [ABSTRACT FROM AUTHOR]

Wood surface defect detection is a challenging task due to the complexity and variability of defect types. To address these challenges, this paper introduces a novel deep learning approach named SiM-YOLO, which is built upon the YOLOv8 object detection framework. A fine-grained convolutional structure, SPD-Conv, is introduced with the aim of preserving detailed defect information during the feature extraction process, thus enabling the model to capture the subtle variations and complex details of wood surface defects. In the feature fusion stage, a SiAFF-PANet-based wood defect feature fusion module is designed to improve the model's ability to focus on local contextual information and enhance defect localization. For classification and regression tasks, the multi-attention detection head (MADH) is employed to capture cross-channel information and the accurate spatial localization of defects. In addition, MPDIoU is employed to optimize the loss function of the model to reduce the leakage of detection due to defect overlap. The experimental results show that SiM-YOLO achieves superior performance compared to the state-of-the-art YOLO algorithm, with a 9.3% improvement in mAP over YOLOX and a 4.3% improvement in mAP over YOLOv8. The Grad-CAM visualization further illustrates that SiM-YOLO provides more accurate defect localization and effectively reduces misdetection and omission issues. This study highlights the effectiveness of SiM-YOLO for wood surface defect detection and offers valuable insights for future research and practical applications in quality control. [ABSTRACT FROM AUTHOR]

Natural gas is widely used in many homes for cooking, but a lack of gas leak detection has led to large fires and accidents. This article presents the design and implementation of a natural gas detection and extraction system for domestic kitchens in Lima, Peru. The ESP32 microcontroller allowed remote circuit control, resulting in a more convenient setup than the Arduino UNO microcontroller. After calibration of the sensors and their corresponding programming, three actions were established in response to different gas levels: alarm activation, space ventilation and gas extraction, and thermal shutdown. Strategic sensor placement and improved physical presentation of the system were performed to ensure accurate readings and effective deployment. The results demonstrate the proper functioning of the circuit and its ability to prevent accidents related to gas leaks. The designed system offers the advantage of remote monitoring, providing access to the user from any location. In conclusion, this project offers a comprehensive solution to prevent accidents caused by gas leaks in home kitchens. With satisfactory results in terms of operation and rapid response, the project demonstrates its effectiveness in accident prevention. This design offers a practical and accessible solution, improving security and bringing peace of mind to homes. [ABSTRACT FROM AUTHOR]

Copyright of Wool Textile Journal is the property of National Wool Textile Science & Technology Information Center and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

AbstractThis study proposes an efficient and effective method to estimate the water temperature inside buried gas pipelines during hydrostatic tests. The goal is to develop appropriate correlations that can easily calculate the pipeline’s water temperature, taking into account the adjacent soil temperature variations, pipeline size, thickness, and coating properties. The study first presents a mathematical model of the problem, introducing relevant non-dimensional parameters. Then, the unsteady problem is numerically analyzed, considering all significant variables, and next, the effects of each non-dimensional parameter on the water temperature are carefully examined and evaluated. The numerical simulations show that the difference between the pipeline water temperature and the adjacent soil temperature can be up to 80%, particularly for large-diameter pipelines. The proposed correlations are able to estimate the pipeline water temperature during the hydrostatic test with around 1% deviation from the direct numerical solutions. Additionally, the results indicate that for typical test cases, the pipeline size is the main influential parameter, while the pipeline thickness and material coating with conventional properties do not significantly affect the water temperature variations. [ABSTRACT FROM AUTHOR]

AbstractProlonged air leakage (AL) following pulmonary resections leads to prolonged hospital stay and post-operative complications. Intra- and postoperative quantification of AL might be useful for improving treatment decisions, but these measurements have not been characterised. AL calculations based on inspiratory and expiratory tidal volumes were investigated in an Intensive Care Unit mechanical ventilator circuit (Servo-I). AL was also measured by a digital chest drainage system. This study shows that AL measurements increase in accuracy when corrected for baseline deviations (R: 0.904 > 0.997, p < 0.001). Bland-Altman analysis revealed a funnel-shape, indicative of a detection threshhold. Corrected measurements were most accurate when averaged over five breaths and AL was >500 mL/min, with an estimated mean systemic bias of 7.4% (95%-limits of agreement [LoA]: 1.1%-13.7%) at 500 mL/min air leak. Breath-by-breath analysis showed most accurate results at AL >20 mL/breath (R: 0.989–0.991, p < 0.001) at tidal volumes between 350–600 mL. The digital drain had a mean systemic bias of −11.1% (95%-LoA: −18.9% to −3.3%) with homogenous scatter in Bland-Altman analysis and a strong correlation to the control measurement over a large range (0–2000mL/min, R: 0.999, p < 0.001). This study indicates that the Servo-I can be used for air leak quantification in clinically relevant ranges (>500 mL/min), but is unsuited for small leak detection due to a detection threshold. Researchers and clinicians should be aware of varying accuracy and interoperability characteristics between AL measurement devices. [ABSTRACT FROM AUTHOR]

In large public places such as railway stations and airports, dense pedestrian detection is important for safety and security. Deep learning methods provide relatively effective solutions but still face problems such as feature extraction difficulties, image multi-scale variations, and high leakage detection rates, which bring great challenges to the research in this field. In this paper, we propose an improved dense pedestrian detection algorithm GR-yolo based on Yolov8. GR-yolo introduces the repc3 module to optimize the backbone network, which enhances the ability of feature extraction, adopts the aggregation–distribution mechanism to reconstruct the yolov8 neck structure, fuses multi-level information, achieves a more efficient exchange of information, and enhances the detection ability of the model. Meanwhile, the Giou loss calculation is used to help GR-yolo converge better, improve the detection accuracy of the target position, and reduce missed detection. Experiments show that GR-yolo has improved detection performance over yolov8, with a 3.1% improvement in detection means accuracy on the wider people dataset, 7.2% on the crowd human dataset, and 11.7% on the people detection images dataset. Therefore, the proposed GR-yolo algorithm is suitable for dense, multi-scale, and scene-variable pedestrian detection, and the improvement also provides a new idea to solve dense pedestrian detection in real scenes. [ABSTRACT FROM AUTHOR]

Aiming at the complex characteristics of negative pressure waves in low-pressure pipelines inside of buildings, we proposed an estimation method of pressure fluctuation trends based on the robust Kalman filter and the improved VMD, which can be used for leakage detection. The reconstructed baseline signal can accurately describe the fluctuation trend of the negative pressure wave after the pressure drop, and quantitatively express the characteristic difference between the leakage condition and the gas usage condition. The robust Kalman filter was used to estimate the pressure fluctuations. The parameters of VMD were adaptively calculated based on the WAA and discrete scale space. The trend components contained in the IMFs were separated by a reconstruction based on the Fourier series. Based on the simulation signal, the method can accurately restore the trend component contained in the complex pressure signal. Based on the actual signals, the accuracy of small leakage detection is 96.7% and the accuracy of large leakage detection is 73.3%. [ABSTRACT FROM AUTHOR]

Aiming at the problems of low contrast of industrial defects and high false detection rate and leakage rate caused by the surrounding interference information, it proposes an industrial surface defect detection algorithm EML-YOLO based on the improvement of YOLOv8. By designing a high-efficiency large convolution module ELK, the model ' s feature extraction capability can be improved by providing a multi-scale feature representation while retaining the spatial information; by proposing a parallel multi-branch feature fusion module MCM, which enables the model to acquire rich feature information and global context information; and reducing the number of parameters and computation of the model by feature compression and streamlining in the Neck module, which makes the model more applicable to industrial scenarios with limited resources. Two industrial surface defect datasets, GC10-DET and DeepPCB, are used to validate the effectiveness of the improved EML-YOLO algorithm. The experimental results show that on the GC10-DET dataset and Deep- PCB dataset, the detection accuracy is improved by 4.3 percentage points and 2.9 percentage points, respectively, and the number of parametric quantities is only 2.7x106. The proposed algorithm can be better applied to industrial defect detection scenarios. [ABSTRACT FROM AUTHOR]

Urban water systems worldwide are confronted with the dual challenges of dwindling water resources and deteriorating infrastructure, emphasising the critical need to minimise water losses from leakage. Conventional methods for leak and burst detection often prove inadequate, leading to prolonged leak durations and heightened maintenance costs. This study investigates the efficacy of logic- and machine learning-based approaches in early leak detection and precise location identification within water distribution networks. By integrating hardware and software technologies, including sensor technology, data analysis, and study on the logic-based and machine learning algorithms, innovative solutions are proposed to optimise water distribution efficiency and minimise losses. In this research, we focus on a case study area in the Sunbury region of Victoria, Australia, evaluating a pumping main equipped with Supervisory Control and Data Acquisition (SCADA) sensor technology. We extract hydraulic characteristics from SCADA data and develop logic-based algorithms for leak and burst detection, alongside state-of-the-art machine learning techniques. These methodologies are applied to historical data initially and will be subsequently extended to live data, enabling the real-time detection of leaks and bursts. The findings underscore the complementary nature of logic-based and machine learning approaches. While logic-based algorithms excel in capturing straightforward anomalies based on predefined conditions, they may struggle with complex or evolving patterns. Machine learning algorithms enhance detection by learning from historical data, adapting to changing conditions, and capturing intricate patterns and outliers. The comparative analysis of machine learning models highlights the superiority of the local outlier factor (LOF) in anomaly detection, leading to its selection as the final model. Furthermore, a web-based platform has been developed for leak and burst detection using a selected machine learning model. The success of machine learning models over traditional logic-based approaches underscores the effectiveness of data-driven, probabilistic methods in handling complex data patterns and variations. Leveraging statistical and probabilistic techniques, machine learning models offer adaptability and superior performance in scenarios with intricate or dynamic relationships between variables. The findings demonstrate that the proposed methodology can significantly enhance the early detection of leaks and bursts, thereby minimising water loss and associated economic costs. The implications of this study are profound for the scientific community and stakeholders, as it provides a scalable and efficient solution for water pipeline monitoring. Implementing this approach can lead to more proactive maintenance strategies, ultimately contributing to the sustainability and resilience of urban water infrastructure systems. [ABSTRACT FROM AUTHOR]

The integration of visual algorithms with infrared imaging technology has become an effective tool for industrial gas leak detection. However, existing research has mostly focused on simple scenarios where a gas plume is clearly visible, with limited studies on detecting gas in complex scenes where target contours are blurred and contrast is low. This paper uses a cooled mid-wave infrared (MWIR) system to provide high sensitivity and fast response imaging and proposes the MWIRGas-YOLO network for detecting gas leaks in mid-wave infrared imaging. This network effectively detects low-contrast gas leakage and segments the gas plume within the scene. In MWIRGas-YOLO, it utilizes the global attention mechanism (GAM) to fully focus on gas plume targets during feature fusion, adds a small target detection layer to enhance information on small-sized targets, and employs transfer learning of similar features from visible light smoke to provide the model with prior knowledge of infrared gas features. Using a cooled mid-wave infrared imager to collect gas leak images, the experimental results show that the proposed algorithm significantly improves the performance over the original model. The segment mean average precision reached 96.1% (mAP50) and 47.6% (mAP50:95), respectively, outperforming the other mainstream algorithms. This can provide an effective reference for research on infrared imaging for gas leak detection. [ABSTRACT FROM AUTHOR]

There is an extensive need for surface sensors for applications such as tactile sensing for robotics, damage and strain detection for structural health monitoring and leak detection for buried structures. One type of surface sensor is electrical impedance tomography (EIT)-based sensing skins, which use electrically conductive coatings applied on the object's surface to monitor physical or chemical phenomena on the surface. In this article, we propose a sensing skin with two electrically coupled layers separated by an insulator. Based on electrical measurements, the spatial distribution of the electrical coupling between the layers is estimated. This coupling is sensitive to both the pressure distribution on the surface and water entering between the layers through a leak. We present simulations and experimental studies to evaluate the feasibility of the proposed method for pressure sensing and leak detection. The results support the feasibility of the proposed method for both of these applications. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this paper is to use the corresponding magnetic sensor and detection method to detect and image the defects of small diameter pipelines. Urban gas pipeline is an energy transportation tool for urban industrial production and social life, which is closely related to urban safety. Preventing the occurrence of urban gas pipeline transportation accidents and carrying out pipeline defect detection are of great significance for the urban economic and social stability. To perform pipeline defect detection, the magnetic flux leakage internal detection method is generally used in the detection of large-diameter long-distance oil and gas pipelines. However, in terms of the internal detection of small-diameter pipelines, due to the heavy weight, large structure of the detection device and small pipe diameter, the detection is more difficult. Design/methodology/approach: In order to solve the above matters, self-made three-dimensional magnetic sensor and three-dimensional magnetic flux leakage imaging direct method are proposed for studying the defect identification. Firstly, for adapting to the diameter range of small-diameter pipelines, and containing the complete information of the defect, a self-made three-dimensional magnetic sensor is made in this paper to improve the accuracy of magnetic flux leakage detection. And on the basis of it, a small diameter pipeline defect detection system is built. Secondly, as detection signal may be affected by background magnetic field interference and the jitter interference, the complete ensemble empirical mode decomposition with adaptive noise method is utilized to screen the detected signal. As a result, the useful signal is reconstructed and the interference signal is removed. Finally, the defect contour inversion imaging of detection is realized based on the direct method of three-dimensional magnetic flux leakage imaging, which includes three-dimensional magnetic flux leakage detection data and data segmentation recognition. Findings: The three-dimensional magnetic flux leakage imaging experimental results shown that, compared to the actual defects, the typical defects, irregular defects and crack groove defects can be analyzed by the magnetic flux leakage defect contour imaging method in qualitative and quantitative way respectively, which provides a new idea for the research of defect recognition. Originality/value: A three-dimensional magnetic sensor is made to adapt the diameter range of small diameter pipeline, and based on it, a small-diameter pipeline defect detection system is built to collect and display the magnetic flux leakage signal. [ABSTRACT FROM AUTHOR]

Utility companies lose approximately 35 liters of water for every 100 produced due to incorrect, illegal connections and the poor condition of pipes. This study develops an intelligent model to detect leaks using the Kalman filter, BiLSTM neural networks, and the Monte Carlo Dropout algorithm. Using data from the Empresa de Acueductos y Alcantarillados de Bogotá (EAAB), Colombia, autocorrelation analysis, PCA, cluster analysis, ADF and Durbin-Watson tests, Hurst exponent, spectral analysis, and wavelet transform were performed. Then, Kalman filtering techniques were applied, and a BiLSTM architecture controlled with Monte Carlo dropout was implemented. The results showed an accuracy of 87.48% in training and 80.48% in validation. Temporal analysis revealed a stationary behavior in the flow series, and the decrease in spectral intensity around 0.25 Hz was related to pressure perturbations caused by leaks. A detailed evaluation of pressure and flow signals identified leak patterns with high precision, demonstrating the effectiveness of the wavelet spectrogram in detecting energy disturbances. The novelty of the study lies in the integration of advanced artificial intelligence and combinatorial optimization techniques to improve water resource management, allowing early and accurate detection of leaks, significantly improving compared to traditional methods. [ABSTRACT FROM AUTHOR]

Returning extraterrestrial samples to Earth for analysis is crucial for planetary geological research and resource utilization. However, during their return journey, these samples undergo significant environmental changes, such as varying pressure, temperature, impact, and vibration. Hence, a dependable sealing technique that preserves sample integrity without requiring high-powered tools is necessary. This article aims to develop an automatic sealing system for collecting extraterrestrial samples, conducting parametric design and mechanical analysis on two types of proposed sealing structures with minimal sealing force as the benchmark. Additionally, the system will undergo validation through sealing and leakage detection experiments. An automated sealing system, capable of storing samples in multiple sampling tubes, is assembled for the extraterrestrial sampling device. [ABSTRACT FROM AUTHOR]

Due to the presence of numerous branches in the pressurized water pipeline network (PWPN), leak-caused negative pressure wave (NPW) has non-unique propagation path, so leak detection and positioning for PWPN is highly challenging. This paper proposes a leak monitoring and positioning method for PWPN based on the time delay matching of NPW propagation. Via introducing the NPW velocity equation, the PWPN is divided into short segments with equal NPW propagation duration, and then the shortest NPW propagation path from each discrete point, namely hypothetical leak point, to each deployed pressure transmitter, is efficiently numerically searched and then used to calculate the NPW arrival time and establish a contrastive time delay library (CTDL) in advance. A set of signal processing algorithm are developed to capture the NPW arrival time and compare these measured values with the pre-established CTDL to accurately position leak point in the PWPN. The effectiveness of this method is verified via leak localization experiments on a field PWPN with an area of 12 × 12 km and 1478 branches. The proposed method can achieve leak positioning with errors less than 100 m throughout the entire pipeline network when only six pressure transmitters are effective. [ABSTRACT FROM AUTHOR]

Background and Objective: Anastomotic leak is one of the most dangerous complications of gastrointestinal surgery, leading to significant morbidity and mortality. Early detection of leaks is crucial for reducing mortality, hospital stay duration, and postoperative complications. This study was designed to identify a suitable prognostic factor. Materials and Methods: This cross-sectional study included fifteen patients selected for gastric and esophagogastric cancer surgery between April 2020 and April 2023. They underwent gastrectomy and primary anastomosis. Demographic information was collected. Complete Blood Count (CBC) and inflammatory markers were recorded preoperatively and on the third day postoperatively. Data analysis was divided into descriptive and inferential statistics. All data were analyzed using Statistical Package for Social Sciences (SPSS) version 24 statistical software, with a significance level of 0.05 for all tests considered. Results: Three cases of anastomotic leakage post-surgery were reported. Significant statistical differences were observed between the two groups with and without anastomotic leakage in terms of Neutrophil to Lymphocyte Ratio (NLR) (2.67 ± 1.14 vs. 1.94 ± 1.06) and Systemic Inflammation Response Index (SIRI) (1.36 ± 1.02 vs. 0.87 ± 0.52) preoperatively (p<0.05). Additionally, on the third day post-surgery, significant statistical differences were found between the two groups in terms of leukocyte count (13.54 ± 2.89 vs. 9.47 ± 2.17), lymphocyte count (1.88 ± 1.10 vs. 2.95 ± 1.22), neutrophil count (9.14 ± 1.77 vs. 7.25 ± 1.08), CRP levels (54.12 ± 4.57 vs. 28.44 ± 2.13), NLR (4.86 ± 1.74 vs. 1.45 ± 0.94), and SIRI (3.88 ± 1.50 vs. 1.36 ± 0.78) (p<0.05). There were no significant differences in Platelet to Lymphocyte Ratio (PLR), Lymphocyte to Monocyte Ratio (LMR), and platelet count between the two groups preoperatively and on the third day postoperatively. Conclusion: C-Reactive Protein (CRP) levels, leukocyte and neutrophil counts, as well as NLR and SIRI ratios on the third day post-surgery, have predictive value for anastomotic leakage. [ABSTRACT FROM AUTHOR]

Pipe leakage in water distribution networks (WDNs) has been an emerging concern for water utilities worldwide due to its public health and economic significance. Not only does it cause significant water losses, but it also deteriorates the quality of the treated water in WDNs. Hence, a prompt response is required to avoid or minimize the eventual consequences. This raises the necessity of exploring the possible approaches for detecting and locating leaks in WDNs promptly. Currently, various leak detection methods exist, but they are not accurate and reliable in detecting leaks. This paper presents a novel GIS-based spatial machine learning technique that utilizes currently installed pressure, flow, and water quality monitoring sensors in WDNs, specifically employing the Geographically Weighted Regression (GWR) and Local Outlier Factor (LOF) models, based on a WDN dataset provided by our partner utility authority. In addition to its ability as a regression model for predicting a dependent variable based on input variables, GWR was selected to help identify locations on the WDN where coefficients deviate the most from the overall coefficients. To corroborate the GWR results, the Local Outlier Factor (LOF) is used as an unsupervised machine learning model to predict leak locations based on spatial local density, where locality is given by k-nearest neighbours. The sample WDN dataset provided by our utility partner was split into 70:30 for training and testing of the GWR model. The GWR model was able to predict leaks (detection and location) with a coefficient of determination (R2) of 0.909. The LOF model was able to predict the leaks with a matching of 80% with the GWR results. Then, a customized GIS interface was developed to automate the detection process in real-time as the sensor's readings were recorded and spatial machine learning was used to process the readings. The results obtained demonstrate the ability of the proposed method to robustly detect and locate leaks in WDNs. [ABSTRACT FROM AUTHOR]

As a key system in nuclear power plants, nuclear power systems contain high-temperature, high-pressure water media. A steam leak, if it occurs, can at minimum cause system functional loss and at worst lead to casualties. Therefore, it is urgent to carry out steam leakage detection work for high-temperature, high-pressure loop systems. Currently, steam leaks are primarily detected through visual monitoring and pressure gauges. However, if there is a minor leak under high system pressure, the slight decrease in pressure may not be enough to alert the operators, leading to a delay in detecting the steam leak. Thus, this detection method has certain drawbacks. In view of these issues, this paper introduces computer vision technology to monitor the high-temperature, high-pressure loop system and proposes the use of an improved background subtraction method to detect steam leaks in the loop system. The results show the following advantages of this method: (1) It can effectively identify steam leaks at an early stage; (2) it overcomes the difficulty of determining the threshold value for the binarization of grayscale images in traditional background subtraction methods; (3) it eliminates the noise impact brought by the binarization of grayscale images in existing improved background subtraction methods. The introduction of this method provides a new approach for detecting steam leaks in high-temperature, high-pressure loop systems and can be effectively applied in engineering fields. It also offers reference value for the detection of high-temperature, high-pressure media leaks in other fields. [ABSTRACT FROM AUTHOR]

The recent transition to H2-based energy storage demands reliable H2 sensors that allow for easy, fast, and reliable detection of leaks. Conventional H2 detectors are based on the changes of physical properties of H2 probes induced by subsurface H-atoms to a material such as electrical conductivity. Herein, we report on highly reactive gasochromic H2 detectors based on the adsorption of H2 on the material surface. We prepared supraparticles (SPs) containing different types of noble metal nanoparticles (NPs), silica NPs, and the dye resazurin by spray-drying and tested their performance for H2 detection. The material undergoes a distinct color change due to the hydrogenation of the purple resazurin to pink resorufin and, finally, colorless hydroresorufin. The stepwise transition is fast and visible to the naked eye. To further improve the performance of the sensor, we tested the reactivity of SPs with different catalytically active NPs by means of in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS). We show that the choice of the NP catalyst has a pronounced effect on the response of the H2 indicator. In addition, we demonstrate that the performance depends on the size of the NPs. These effects are attributed to the availability of reactive H-atoms on the NP surface. Among the materials studied, Pt-containing SPs gave the best results for H2 detection. [ABSTRACT FROM AUTHOR]

As the main support part of the working platform of a high-rise building machine, the bearing pin support (BPS) plays a crucial role in the safety and stability of the platform, the conventional method has the problems of low detection efficiency, low accuracy, and high cost. To improve the accuracy and robustness of the detection algorithm under weak light, this paper proposes an intelligent detection algorithm for the BPS-piece states of the BS-YOLOV8, to improve the feature map utilization and reduce the model leakage detection error detection rate, Swin transformer is used to improve the YOLOV8 backbone network. In addition, the BiFormer attention mechanism is used to weigh the feature map to solve the problem of feature information loss in different feature layers and weak lighting conditions, and then the Scylla-IOU loss function is used instead of the original localization loss function to guide the model to learn to generate a predicted bounding box closer to the real target bounding box. Finally, the BS-YOLOV8 algorithm is used to compare with its classical algorithm on the self-constructed dataset of this study, The results show that the mAP0.5, mAP0.5:0.95, and FPS values of the BS-YOLOV8 algorithm reach 97.9%, 96.3% and 40 under normal lighting. The mAP0.5 value reaches 87.6% under low light conditions, which effectively solves the problems of low detection efficiency and poor detection under low light conditions, and is superior compared to other algorithms. [ABSTRACT FROM AUTHOR]

AbstractIntroduction: Conventional leak detection methods are costly and time-consuming, so attention has recently been drawn to methods that detect leakage by network modeling comprehensively.Methods: In these methods, the difference between the measured pressure and flow rate and those obtained by the model is minimized as a fitness function by calibrating the network and adjusting the nodal demands. It has been established by analysis that the fitness function in itself cannot be a good guide to achieving the minimum point. This drawback was solved by using the ACO method. In addition to the fitness, this method employs a parameter called heuristic guidance to improve its capabilities.Results: This research compared three ACO-based optimization methods. The first was ACO with fixed heuristic guidance already introduced in the literature. In the second method, the heuristic guidance varies by the values of the observed pressures and pressure decline versus the no-leak state. The third is studied and compared by integrating SSEM and ACO models. These methods were investigated on a network derived from the literature and the Birjand network.Conclusion: The results revealed that the methods proposed here achieved more precise results in a shorter time and with fewer iterations.

To investigate the leakage characteristics of pure hydrogen and hydrogen-blended natural gas in medium- and low-pressure buried pipelines, this study establishes a three-dimensional leakage model based on Computational Fluid Dynamics (CFD). The leakage characteristics in terms of pressure, velocity, and concentration distribution are obtained, and the effects of operational parameters, ground hardening degree, and leakage parameters on hydrogen diffusion characteristics are analyzed. The results show that the first dangerous time (FDT) for hydrogen leakage is substantially shorter than for natural gas, emphasizing the need for timely leak detection and response. Increasing the hydrogen blending ratio accelerates the diffusion process and decreases the FDT, posing greater risks for pipeline safety. The influence of soil hardening on gas diffusion is also examined, revealing that harder soils can restrict gas dispersion, thereby increasing localized concentrations. Additionally, the relationship between gas leakage time and distance is determined, aiding in the optimal placement of gas sensors and prediction of leakage timing. To ensure the safe operation of hydrogen-blended natural gas pipelines, practical recommendations include optimizing pipeline operating conditions, improving leak detection systems, increasing pipeline burial depth, and selecting materials with higher resistance to hydrogen embrittlement. These measures can mitigate risks associated with hydrogen leakage and enhance the overall safety of the pipeline infrastructure. [ABSTRACT FROM AUTHOR]

Detecting pipeline leaks is an essential factor in maintaining the integrity of fluid transport systems. This paper introduces an advanced deep learning framework that uses continuous wavelet transform (CWT) images for precise detection of such leaks. Transforming acoustic signals from pipelines under various conditions into CWT scalograms, followed by signal processing by non-local means and adaptive histogram equalization, results in new enhanced leak-induced scalograms (ELIS) that capture detailed energy fluctuations across time-frequency scales. The fundamental approach takes advantage of a deep belief network (DBN) fine-tuned with a genetic algorithm (GA) and unified with a least squares support vector machine (LSSVM) to improve feature extraction and classification accuracy. The DBN-GA framework precisely extracts informative features, while the LSSVM classifier precisely distinguishes between leaky and non-leak conditions. By concentrating solely on the advanced capabilities of ELIS processed through an optimized DBN-GA-LSSVM model, this research achieves high detection accuracy and reliability, making a significant contribution to pipeline monitoring and maintenance. This innovative approach to capturing complex signal patterns can be applied to real-time leak detection and critical infrastructure safety in several industrial applications. [ABSTRACT FROM AUTHOR]

Recent evaluations of counter-based periodic testing strategies for fault detection in Microprocessor (μP) have shown that only a small set of counters is needed to provide complete coverage of severe faults. Severe faults are defined as faults that leak sensitive information, e.g., an encryption key on the output of a serial port. Alternatively, fault detection can be accomplished by executing instructions that periodically test the control and functional units of the μP. In this paper, we propose a fault detection method that utilizes an 'engineered' executable program combined with a small set of strategically placed counters in pursuit of a hardware Periodic Built-In-Self-Test (PBIST). We analyze two distinct methods for generating such a binary; the first uses an Automatic Test Generation Pattern (ATPG)-based methodology, and the second uses a process whereby existing counter-based node-monitoring infrastructure is utilized. We show that complete fault coverage of all leakage faults is possible using relatively small binaries with low latency to fault detection and by utilizing only a few strategically placed counters in the μP. [ABSTRACT FROM AUTHOR]

The instance segmentation model based on deep learning has addressed the challenges in intelligently detecting water leakage in shield tunneling. Due to the limited generalization ability of the baseline model, occurrences of missed detections, false detections, and repeated detections are encountered during the actual detection of tunnel water leakage. This paper adopts Mask R-CNN as the baseline model and introduces a mask cascade strategy to enhance the quality of positive samples. Additionally, the backbone network in the model is replaced with RegNetX to enlarge the model's receptive field, and MDConv is introduced to enhance the model's feature extraction capability in the edge receptive field region. Building upon these improvements, the proposed model is named Cascade-MRegNetX. The backbone network MRegNetX features a symmetrical block structure, which, when combined with deformable convolutions, greatly assists in extracting edge features from corresponding regions. During the dataset preprocessing stage, we augment the dataset through image rotation and classification, thereby improving both the quality and quantity of samples. Finally, by leveraging pre-trained models through transfer learning, we enhance the robustness of the target model. This model can effectively extract features from water leakage areas of different scales or deformations. Through instance segmentation experiments conducted on a dataset comprising 766 images of tunnel water leakage, the experimental results demonstrate that the improved model achieves higher precision in tunnel water leakage mask detection. Through these enhancements, the detection effectiveness, feature extraction capability, and generalization ability of the baseline model are improved. The improved Cascade-MRegNetX model achieves respective improvements of 7.7%, 2.8%, and 10.4% in terms of AP, AP0.5, and AP0.75 compared to the existing Cascade Mask R-CNN model. [ABSTRACT FROM AUTHOR]

In the context of the Internet of Things, there is a growing demand for real-time monitoring of water distribution systems (WDS). Among the various leak detection methods, acoustic leak detection is considered to be a suitable method. However, existing methods are not very effective in environments with high daytime ambient noise. To address this issue, this paper conducted on-site data collection experiments and designed a monitoring system that combines traditional nighttime monitoring with daytime monitoring, combining water company pipeline inspections and repair work. A large number of daytime audio samples were collected. In this paper, the logarithmic spectrogram (log spectrogram) was used to represent the features of the leak signal. By comparing the features of the signal during day and night, noisy and quiet environments, and leak and normal signals, we identified the interfering frames that required noise reduction, and applied frame-level noise reduction processing to the signal. Based on this, a log PS-ResNet18 model was developed to identify leaks, and its performance was compared with other classification models [including traditional nighttime detection methods, random forests, XGBoost, and convolutional neural network (CNN)]. The results showed that the log PS-ResNet18 model had the best performance, with an all-day accuracy rate of 99.4% and a daytime accuracy rate of 99.3%. In addition, by conducting ablation experiments to explore the role and contribution of the log PS-ResNet18 and noise reduction methods in the model, the results showed that the log spectrogram and noise reduction methods increased the all-day accuracy rate by 18.8% and 23.2%, respectively, and by 24.7% when used together. In another practical application, the log PS-ResNet18 model achieved an all-day detection accuracy rate of 99.6%. This study demonstrated the applicability of the log spectrogram and CNN combination in daytime leak detection, overcoming research limitations in the field. This research presents the log PS-ResNet18 framework, which combines deep learning models and denoised logarithmic spectrograms to improve leak detection in water supply pipelines under daytime environmental noise. The research focuses on field data collection and analysis of cast iron pipes with different diameters in Hangzhou (HZ). The model was tested on cast iron pipes in Lishui (LS) and proved to be effective. The proposed method is highly versatile and can be applied to different regions and pipe materials after sufficient sample collection and model training validation. The research recommends a comprehensive leak monitoring solution that involves initial intelligent detection using front-end noise meters and secondary identification of suspicious audio signals using the log PS-ResNet18 model in the cloud. This enables water utility operators to respond quickly to pipeline leaks, leading to more efficient water resource conservation and improved water supply service quality. [ABSTRACT FROM AUTHOR]