A Novel Parameter-Optimized Recurrent Attention Network for Pipeline Leakage Detection

Accurate detection of pipeline leakage is essential to maintain the safety of pipeline transportation.Recently,deep learning(DL)has emerged as a promising tool for pipeline leakage detection(PLD).However,most existing DL methods have difficulty in achieving good performance in identifying leakage types due to the complex time dynamics of pipeline data.On the other hand,the initial parameter selection in the detection model is generally random,which may lead to unstable recognition performance.For this reason,a hybrid DL framework referred to as parameter-optimized recurrent attention network(PRAN)is presented in this paper to improve the accuracy of PLD.First,a parameter-optimized long short-term memory(LSTM)network is introduced to extract effective and robust features,which exploits a particle swarm optimization(PSO)algorithm with cross-entropy fitness function to search for globally optimal parameters.With this framework,the learning representation capability of the model is improved and the convergence rate is accelerated.Moreover,an anomaly-attention mechanism(AM)is proposed to discover class discriminative information by weighting the hidden states,which contributes to amplifying the normalabnormal distinguishable discrepancy,further improving the accuracy of PLD.After that,the proposed PRAN not only implements the adaptive optimization of network parameters,but also enlarges the contribution of normal-abnormal discrepancy,thereby overcoming the drawbacks of instability and poor generalization.Finally,the experimental results demonstrate the effectiveness and superiority of the proposed PRAN for PLD.

Automatic Sensing and Detection for Subway Tunnel Pathologies

Subway tunnels often suffer from surface pathologies such as cracks,corrosion,fractures,peeling,water and sand infiltration,and sudden hazards caused by foreign object intrusions.Installing a mobile visual pathology sensing system at the front end of operating trains is a critical measure to ensure subway safety.Taking leakage as the typical pathology,a tunnel pathology automatic visual detection method based on Deeplabv3+(ASTPDS)was proposed to achieve automatic and high-precision detection and pixel-level morphology extraction of pathologies.Compared with similar methods,this approach showed significant advantages and achieved a detection accuracy of 93.12%,surpassing FCN and U-Net.Moreover,it also exceeded the recall rates for detecting leaks of FCN and U-Net by 8.33%and 8.19%,respectively.

Automatic Visual Leakage Detection and Localization from Pipelines in Chemical Process Plants Using Machine Vision Techniques

Liquid leakage from pipelines is a critical issue in large-scale process plants.Damage in pipelines affects the normal operation of the plant and increases maintenance costs.Furthermore,it causes unsafe and hazardous situations for operators.Therefore,the detection and localization of leakages is a crucial task for maintenance and condition monitoring.Recently,the use of infrared(IR)cameras was found to be a promising approach for leakage detection in large-scale plants.IR cameras can capture leaking liquid if it has a higher(or lower)temperature than its surroundings.In this paper,a method based on IR video data and machine vision techniques is proposed to detect and localize liquid leakages in a chemical process plant.Since the proposed method is a vision-based method and does not consider the physical properties of the leaking liquid,it is applicable for any type of liquid leakage(i.e.,water,oil,etc.).In this method,subsequent frames are subtracted and divided into blocks.Then,principle component analysis is performed in each block to extract features from the blocks.All subtracted frames within the blocks are individually transferred to feature vectors,which are used as a basis for classifying the blocks.The k-nearest neighbor algorithm is used to classify the blocks as normal(without leakage)or anomalous(with leakage).Finally,the positions of the leakages are determined in each anomalous block.In order to evaluate the approach,two datasets with two different formats,consisting of video footage of a laboratory demonstrator plant captured by an IR camera,are considered.The results show that the proposed method is a promising approach to detect and localize leakages from pipelines using IR videos.The proposed method has high accuracy and a reasonable detection time for leakage detection.The possibility of extending the proposed method to a real industrial plant and the limitations of this method are discussed at the end.

Gas Leakage Detection and Pressure Difference Identification by Asymmetric Differential Pressure Method

Currently,the measurement methods for pneumatic system leakage include bubbling,ultrasonic,and pressure detection methods.These methods are sensitive to high-precision sensors,long detection times,and stable external environments.The traditional differential pressure method involves severe differential pressure fluctuations caused by environmental pressure fluctuations or electromagnetic noise interference of sensors,leading to inaccurate detection.In this paper,a differential pressure fitting method for an asymmetric differential pressure cylinder is proposed.It overcomes the limitation of the detection efficiency caused by the asynchronous temperature recovery of the two chambers in the asymmetric differential pressure method and uses the differential pressure substitution equation to replace the differential calculation of the differential pressure.The improved differential pressure method proposes an innovation based on the detection principle and calculation method.Additionally,the influence of the parameters in the differential pressure substitution equation on the leakage calculation results was simulated,and the specific physical significance of the parameters of the differential pressure substitution equation was explained.The experiments verified the fitting effect and proved the accuracy of this method.Compared with the traditional differential pressure method,the maximum leakage deviation of inhibition was 0.5 L/min.Therefore,this method can be used to detect leaks in air tanks.

Bubble counter based on photoelectric technique for leakage detection of cryogenic valves

In order to overcome the inconvenience of manual bubble counting, a bubble counter based on photoelectric technique aiming for automatically detecting and measuring minute gas leakage of cryogenic valves is proposed. Experiments have been conducted on a self-built apparatus, testing the performance with different gas inlet strategies (bottom gas-inlet strategy and side gas-inlet strategy) and the influence of gas pipe length (0, 1, 2, 4, 6, 8, 10 m) and leakage rate (around 10, 20, 30, 40 bubbles/min) on first bubble time and bubble rate. A buffer of 110 cm3 is inserted between leakage source and gas pipe to simulate the down- stream cavum adjacent to the valve clack. Based on analyzing the experimental data, experiential parameters have also been summarized to guide leakage detection and measurement for engineering applications. A practical system has already been suc- cessfully applied in a cryogenic testing apparatus for cryogenic valves.

A Hardware Trojan Detection Method Based on the Electromagnetic Leakage

Hardware Trojan(HT) refers to a special module intentionally implanted into a chip or an electronic system. The module can be exploited by the attacker to achieve destructive functions. Unfortunately the HT is difficult to detecte due to its minimal resource occupation. In order to achieve an accurate detection with high efficiency, a HT detection method based on the electromagnetic leakage of the chip is proposed in this paper. At first, the dimensionality reduction and the feature extraction of the electromagnetic leakage signals in each group(template chip, Trojan-free chip and target chip) were realized by principal component analysis(PCA). Then, the Mahalanobis distances between the template group and the other groups were calculated. Finally, the differences between the Mahalanobis distances and the threshold were compared to determine whether the HT had been implanted into the target chip. In addition, the concept of the HT Detection Quality(HTDQ) was proposed to analyze and compare the performance of different detection methods. Our experiment results indicate that the accuracy of this detection method is 91.93%, and the time consumption is 0.042s in average, which shows a high HTDQ compared with three other methods.

Online Magnetic Flux Leakage Detection System for Sucker Rod Defects Based on LabVIEW Programming

Aiming at the detection of the sucker rod defects,a real-time detection system is designed using the non-destructive testing technology of magnetic flux leakage(MFL).An MFL measurement system consists of many parts,and this study focuses on the signal acquisition and processing system.First of all,this paper introduces the hardware part of the acquisition system in detail,including the selection of the Hall-effect sensor,the design of the signal conditioning circuit,and the working process of the single chip computer(SCM)control serial port.Based on LabVIEW,a graphical programming software,the software part of the acquisition system is written,including serial port parameter configuration,detection signal recognition,original signal filtering,real-time display,data storage and playback.Finally,an experimental platform for the MFL detection is set up,and the MFL measurement is carried out on the transverse and longitudinal defects of the sucker rod surface.The experimental result shows that the designed acquisition and processing system has good detection performance,simple design and high flexibility.

Composite excitation multi-extension direction defect magnetic flux leakage detection technology

In the traditional pipeline magnetic flux leakage(MFL)detection technology,circumferential or axial excitation is mainly used to excite the magnetic field of defects.However,the domestic and foreign pipeline detection devices currently in operation are mainly axial excitation MFL detection tools,in which circumferential cracks can be clearly identified,but the detection sensitivity of axial cracks is not high,thus forming a detection blind zone.Therefore,a composite excitation multi-extension direction defect MFL detection method is proposed,which can realize the simultaneous detection of axial and circumferential defects.On the basis of the electromagnetic theory Maxwell equation and Biot Savart law,a mathematical model of circumferential and axial magnetization is firstly established.Then finite element simulation software is used to establish a model of a new type of magnetic flux leakage detection device,and a simulation analysis of crack detection in multiple extension directions is carried out.Finally,under the conditions of the relationship model between the change rate of leakage magnetic field and external excitation intensity under unsaturated magnetization and the multi-stage coil magnetization model,the sample vehicle towing experiment is carried out.The paper aims to analyze the feasibility and effectiveness of the new magnetic flux leakage detection device for detecting defects in different extension directions.Based on the final experimental results,the new composite excitation multi extension direction leakage magnetic field detector has a good detection effect for defects in the axial and circumferential extension directions.

Empirical Evaluation of Information Leakage Detection Using Net-flow Analysis

Because of the widespread of Trojans,organizations and Internet users become more vulnerable to the threat of information leakage.This paper describes an information leakage detection system( ILDS) to detect sensitive information leakage caused by Trojan.In particular,the principles of the system are based on the analysis of net-flows in four perspectives: heartbeat behavior analysis,DNS abnormal analysis,uploaddownload ratio and content analysis.Heartbeat behavior analysis and DNS abnormal analysis are used to detect the existence of Trojans while upload-download ratio and content analysis can quickly detect when the information leakage happens.Experiments indicate that the system is reliable and efficient in detecting information leakage.The system can also help to collect and preserve digital evidence when information leakage incident occurs.

Enhanced Side-Channel Leakage Detection Method by Considering Combinational Logic

In this paper,we propose a hybrid power model that includes the power consumption of not only the registers but also part of the combinational logic.By doing knownkey analysis with this hybrid model,power side-channel leakage caused by correct keys can be detected.In experiment,PRINTcipher and DES algorithms were chosen as analysis targets and combinational logic s-box unit was selected to build power template.The analysis results showed the signal-to-noise ratio(SNR) power consumption increase of more than 20%after considering s-box's power consumption so that the information of keys can be obtained with just half number of power traces.In addition,the side channel-leakage detection capability of our method also shows better effectiveness that can identify the correct keys.

High voltage DC leakage detection for double-lined hazardous waste landfill based on finite element method

According to the structural characteristics of hazardous waste landfill, a new model based on the finite element method （FEM） is developed. The detection layer is considered as a sealed space and it is assumed that total current flows through the leak for the high resistivity of geomembrane liner. The leak current is regarded as a positive point current ＋I and the other current source is -I. Electrical potential of an arbitrary point in detection layer satisfies Poisson equation. Experiments for detecting leaks in liner were carried out. Excellent agreement between experimental data and simulated model data validates the new model. Parametric curves for a single leak show that with optimum selection of field survey parameters leaks can be detected effectively. For multiple leaks, the simulated results indicate that they are detectable when leak separation is larger than measurement spacing.

Design of hydrogen leakage detection and alarm processing system for the fuel cell power train lab

The hydrogen leakage detection and alarm processing system is established for the fuel cell （FC） power train lab to meet the hydrogen safety demand of the FC performance test and examination for the project named ＂Research and Development of the Vehicular Technology for the Fuel Cell City Bus＂ by Tsinghua University. The established hydrogen safety system includes the hydrogen supply system, hydrogen leakage detection system, alarm processing system, ventilation system, measures against electrostatic, thunder-arresting and explosion-protection, and the strict hydrogen operation rules. In this safety system, the explosion proof catalytic combustion sensors are used to detect the hydrogen leakage and the electrical control system is designed to process the alarm automatically. The hydrogen safety system plays an important role in the performance, examination of the FC and the assuring the personnel＇ s safety of the fuel cell power train lab.

Deep learning based water leakage detection for shield tunnel lining

Shield tunnel lining is prone to water leakage,which may further bring about corrosion and structural damage to the walls,potentially leading to dangerous accidents.To avoid tedious and inefficient manual inspection,many projects use artificial intelligence(Al)to detect cracks and water leakage.A novel method for water leakage inspection in shield tunnel lining that utilizes deep learning is introduced in this paper.Our proposal includes a ConvNeXt-S backbone,deconvolutional-feature pyramid network(D-FPN),spatial attention module(SPAM).and a detection head.It can extract representative features of leaking areas to aid inspection processes.To further improve the model's robustness,we innovatively use an inversed low-light enhancement method to convert normally illuminated images to low light ones and introduce them into the training samples.Validation experiments are performed,achieving the average precision(AP)score of 56.8%,which outperforms previous work by a margin of 5.7%.Visualization illustrations also support our method's practical effectiveness.

Broken and fine powder detection method of fuses

Since powder and cotton wool may cause mechanical test fail,and the accuracy of ray detection method is not very ideal,this paper introduces the tiny capacitance detection method for broken gunpowder,in which the medium type capacitive sensor’s capacitance changes when the gunpowder core diameter changes.The detection is based on the relationship between non-electric quantity and electric quantity.Simulation results show that the designed system works well with easy installation and low power consumption.Therefore,it is suitable for broken gunpowder detection.

Key technique of a detection sensor for coal mine wire ropes

Wire ropes,employed extensively in coal mine hoists and transportation systems are subject to damage due to wear,corrosion and fatigue.The extent of damage and the carrying capacity of ropes are closely related to the sense of safety by staff and equipments.Magnetic flux leakage detection method(MFL),as an effective method,is these days widely used in detection of broken strands of wire ropes.In order to improve the accuracy of detection of flaws in wire ropes by magnetic flux leakage(MFL),the effect of the distance between a sensor and the surface of a wire rope(i.e.,lift-off) on detection by magnetic flux leakage was in-vestigated.An analysis of the main principles for the choice of lift-off is described by us and a new method that improves the structure of the detector is proposed from the point of view of the design of a magnetic circuit,to restrain the impact of fluctuations of sensor lift-off.The effect of this kind of method is validated by simulation and computation.The results show that the detection sensitivity is markedly increased by this method.Furthermore,the signal-to-noise ratio(SNR) can be increased by over 28%.This method will lend itself to offer reliable scientific information to optimize the structure of excitation devices and improve the accuracy of MFL detection.

Instrument for insulation conditions detection of roof insulator on high-speed-train

With the rapid development of high-speed-railway,environment around high voltage device on train roof becomes very complicated. Most train accidents happened due to occurrence of flashover on roof insulator,but the insulation condition estimation of insulator in such environment is much difficult. To ensure the insulation property of electric equipment,and guarantee the operation safety of high-speed-train,here established an instrument with high reliability which can on-line monitor insulation condition of roof insulator and give out advanced alarm before the incipient insulator flashover. The instrument consists of three parts,Data Acquisition & Sensor,Data Processing and Back Processing. Anti-interference and protection methods are processed to Rogowski coil sensor for better leakage current signal. To avoid the fluctuation from railway power supply,four modules are set to filter the power supply waveform. Through laboratory measurement,it is shown that the leakage current and the impedance angle can be detected by the instrument accurately. From the comparison of leakage current and impedance angle results under different moisture condition and the alarm operation when leakage current value reached threshold,this instrument can give out enough information for staff to understand the insulation condition of insulator.

Research on Oil Pipeline Leak Detection and Location system

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.

改进YOLOv7-tiny的无人机目标检测算法

针对无人机视角下小目标难以检测、目标密集和环境复杂导致漏检概率增加的问题,提出一种改进YOLOv7-tiny的无人机目标检测算法。在原主干网络的基础上增加一个并行网络,加强主干网络对特征图信息的提取能力;增加细小目标采样尺度并改进FPN结构,使主干网络输出的特征图可以用于后续上采样和下采样当中,提高网络精度;加入CA注意力机制,优化主干网络输出特征图,减少特征信息损失;使用WIoU损失函数计算定位损失,增强网络对小目标的检测能力。实验结果表明,相较于原算法,改进YOLOv7-tiny算法的准确率和召回率分别提升了2.8和2.7个百分点,mAP@0.5和mAP@0.5:0.95分别提升了3.8和3.2个百分点,有效提高了算法的检测精度。

基于改进YOLOv7-tiny的高空作业人员安防装备检测算法

针对现有高空作业人员安防装备检测算法参数量较大且检测目标相对单一,难以适应复杂的高空作业场景等问题,提出一种基于改进YOLOv7-tiny的高空作业人员安防装备检测算法。首先,将主干网络重新设计为更轻量的YOLOv7-FasterNet,并调整空间金字塔池化结构,实现模型参数量的压缩;其次,在ELAN-L模块中扩展梯度传输路径的分支,解决了模型压缩造成的通道信息缺失问题,提升了特征信息的提取能力;最后,将网络中下采样部分替换为Involution模块,降低参数冗余,增强网络对全局的捕获能力。实验结果表明,改进的YOLOv7-tiny算法能够更好地适应复杂高空作业场景,在开源数据集上具备良好的性能。该算法的平均检测精度达到94.7%,较原模型提升1.5%,参数量较原模型下降11.6%,实验结果验证了算法改进措施的有效性。

TO–YOLOX: a pure CNN tiny object detection model for remotesensing images

Remote sensing and deep learning are being widely combined in tasks such as urban planning and disaster prevention.However,due to interference occasioned by density,overlap,and coverage,the tiny object detection in remote sensing images has always been a difficult problem.Therefore,we propose a novel TO–YOLOX(Tiny Object–You Only Look Once)model.TO–YOLOX possesses a MiSo(Multiple-in-Singleout)feature fusion structure,which exhibits a spatial-shift structure,and the model balances positive and negative samples and enhances the information interaction pertaining to the local patch of remote sensing images.TO–YOLOX utilizes an adaptive IOU-T(Intersection Over Uni-Tiny)loss to enhance the localization accuracy of tiny objects,and it applies attention mechanism Group-CBAM(group-convolutional block attention module)to enhance the perception of tiny objects in remote sensing images.To verify the effectiveness and efficiency of TO–YOLOX,we utilized three aerial-photography tiny object detection datasets,namely VisDrone2021,Tiny Person,and DOTA–HBB,and the following mean average precision(mAP)values were recorded,respectively:45.31%(+10.03%),28.9%(+9.36%),and 63.02%(+9.62%).With respect to recognizing tiny objects,TO–YOLOX exhibits a stronger ability compared with Faster R-CNN,RetinaNet,YOLOv5,YOLOv6,YOLOv7,and YOLOX,and the proposed model exhibits fast computation.