Surface Damage Detection Research Articles

The Significance of In-Service Factors for the Visual Detectability of Impact Damage in Composite Airframe

The inspection conditions may significantly affect the visual detectability of impact damage in aircraft composite skins. The goal of the present study was to identify the most affecting factors related to standard field control procedures and thus establish reasonable barely visible impact damage threshold in the frameworks of Irkut MC-21 aircraft certification. The probability of surface damage detection as a function of damage size was evaluated in relation to qualification of experts, paint color, viewing distance and surface contamination. The experiments with representative panels were carried out, the empirical detectability data was analyzed by methods of mathematical statistics. Using non-parametric tests the statistical significance of mentioned in-service factors was estimated. The bootstrap method was applied to make empirical data sufficiently representative. The minimum reliably detectable sizes of dents caused by 25 mm indenter for considered inspection conditions were identified. For detailed inspection in case of the clean surface the size of 0.3 mm can be accepted as a relevant criterion, in case of contamination this value can grow up to three times. For general inspection the values of 1 mm for clean and 1.7 mm for contaminated surface can be accepted conservatively. The obtained results were shown to be consistent with the literature and were recommended for use on the stages of design and maintenance.

Road Surface Damage Detection Based on Semi-supervised Learning Using Pseudo Labels

Road Surface Damage Detection Based on Semi-supervised Learning Using Pseudo Labels

Detection of surface damages in ground spur gears by non-destructive micromagnetic methods

The analysis of the Barkhausen noise signal is a non-destructive micromagnetic method to detect grinding damages, such as grinding burn. This method is increasingly used in industrial production for quality control. However, there is still no scientifically validated knowledge about the influences and interactions of different material parameters resulting from the heat treatment on the Barkhausen noise signal. Therefore, no standardized guideline for interpretation of inspection results exists up to now; each company uses own test procedures.

Performance Enhancement Algorithm using Supervised Learning based on Background Object Detection for Road Surface Damage Detection

Performance Enhancement Algorithm using Supervised Learning based on Background Object Detection for Road Surface Damage Detection

Road Surface Damage Detection based on Object Recognition using Fast R-CNN

도로 관리 주체는 도로 파손을 보수하기 위해 적지 않은 비용을 투입한다. 이러한 파손은 자연 요인과 노후화로 인하여 필연적으로 발생을 하는데, 효율적인 보수를 위한 유지보수 기 술이 필요하다. 이런 수요에 대응하기 위해 여러 가지 기술들이 개발되고 적용되고 있지만, 최 근 들어서는 차량용 블랙박스 형태로 수집한 영상 정보를 바탕으로 도로 노면 파손 유지 보수 기술이 개발되고 있다. 이 파손 영역을 추출하는 방법에는 여러 가지가 있지만, 본 논문에서는 최근 활발히 연구되고 있는 심층 신경망 구조의 영상인식 기술에 대해 논하고자 한다. 특히 영역 기반의 합성곱 알고리즘을 이용하여 영상 내에서 도로 파손 유무와 그 영역을 추정할 수 있는 새로운 심층 신경망을 소개한다. 이를 개발하기 위해 실제 주행을 통해서 600여장의 영상 데이터를 수집하였고, 이를 활용하여 학습을 수행하였다. 그 결과 기존 모델과 성능을 비교하 여 10.67% 향상된 신경망을 개발하였다.

Wind Turbine Surface Damage Detection by Deep Learning Aided Drone Inspection Analysis

Timely detection of surface damages on wind turbine blades is imperative for minimizing downtime and avoiding possible catastrophic structural failures. With recent advances in drone technology, a large number of high-resolution images of wind turbines are routinely acquired and subsequently analyzed by experts to identify imminent damages. Automated analysis of these inspection images with the help of machine learning algorithms can reduce the inspection cost. In this work, we develop a deep learning-based automated damage suggestion system for subsequent analysis of drone inspection images. Experimental results demonstrate that the proposed approach can achieve almost human-level precision in terms of suggested damage location and types on wind turbine blades. We further demonstrate that for relatively small training sets, advanced data augmentation during deep learning training can better generalize the trained model, providing a significant gain in precision.

Detection of surface damage using resistance tomography in thin graphite layer

The article presents the application of the resistive tomography method to the location of surface damage. A very thin layer of current-carrying graphite was applied to the tested surface. The use of a thin layer results in a relatively high resistance, which facilitates measurement and reduces possible errors. The damage was simulated by making holes in the layer. Measurement of resistance between electrode pairs on opposite edges allowed for the creation of maps representing surfaces of relative resistance in two perpendicular directions. The resultant surface formed on their basis shows a clear extreme in the place of the simulated damage. The process was provided for one or two independent failures with different diameters, showing the correctness of the location. The experimental study was supplemented with numerical calculations showing a very good correlation. The proposed method can be used in any case where it is possible to apply a thin layer of conductor to a non-conductive surface

Research on Post Fire Detection and Evaluation of Liu Jiang the Yellow River Bridge

After the bridge is damaged by fire, it is necessary to quickly probe the situation of the damaged structure, to take a response as soon as possible to ensure the safe passage of the bridge. A series of related work such as the component surface damage detection, carbonation depth, concrete strength testing and Axial compressive strength tested by core drilling combined with the analysis and calculation after fire had provided a basis for the safety assessment of structures. According to the results of the test and the conclusion of the finite element calculation, it was found that the bridge belongs to the dangerous situation, and the relevant strengthening measures should be taken.

Road Surface Damage Detection Algorithms and its Road Test using Asymmetric Auto-Encoder Deep Neural Network

Road Surface Damage Detection Algorithms and its Road Test using Asymmetric Auto-Encoder Deep Neural Network

Remote surface damage detection on rotor blades of operating wind turbines by means of infrared thermography

Abstract. Wind turbines are constantly exposed to wind gusts, dirt particles and precipitation. Depending on the site, surface defects on rotor blades emerge from the first day of operation on. While erosion increases quickly with time, even small surface defects can affect the performance of the wind turbine. Consequently, there is demand for an easily applicable remote monitoring method for rotor blades that is capable of detecting surface defects at an early stage. In this work it is investigated if infrared thermography (IRT) can meet these requirements by visualizing differences in the thermal transport and the corresponding surface temperature of the wall-bounded flow.Firstly, a validation of the IRT method compared to stereoscopic particle image velocimetry measurements is performed comparing both types of experimental results for the boundary layer of a flat plate. Then, the main characteristics of the flow in the wake of generic surface defects on different types of lifting surfaces are studied both experimentally and numerically: temperature gradients behind protruding surface defects on a flat plate and a DU 91-W2-250 profile are studied by means of IRT. The same is done with the wall shear stress from Reynolds-averaged Navier–Stokes simulations of a wind turbine blade. It is consistently observed, both in the experiments and the simulations, that turbulent wedges are formed on the flow downstream of generic surface defects. These wedges provide valuable information about the kind of defects that generate them. At last, experimental and numerical performance measures are taken into account for evaluating the aerodynamic impact of surface defects on rotor blades. We conclude that the IRT method is a suitable remote monitoring technique for detecting surface defects on wind turbines at an early stage.

Road pavement defect assessment through vibration analysis inside vehicles

Miniature accelerometer sensors were used for the evaluation of road surface roughness. The device designed for installation in the vehicles is composed of a GPS receiver and of multi-axis accelerometers. Smartphones with built-in accelerometers were also used. Measurement data were collected through the recording of road trips employing 3 car types on diversified surface roughness roads and with varied vehicle speed on each investigated road section. The first step of data processing was the sensor alignment made to achieve proper values of acceleration vector. Subsequently, the influence of the car suspension system to the measurement results was diminished employing a designed filter. The magnitude of coefficients of Gabor transform analysis of accelerometer signals was calculated to discover differences between good and damaged surface. Research results show that road sections quality can be assessed by the applied vibration analysis. Although the precision of low-cost devices may be lower than the application of expensive professional laser profilograph scanning on road pavements, they can help to increase the effectiveness and the coverage of road surface damage detection, through the monitoring of road surfaces with typical cars instead of special test vehicles only. [Research was subsidized by the Polish National Centre for Research and Development and the General Directorate of Public Roads and Motorways within the Grant No. OT4- 4B/AGH-PG-WSTKT.]

Damage Detection and Localization in Metallic Structures Based on Jointed Electromagnetic Waveguides: A Proof-of-Principle Study

This paper presents a novel approach for the detection of surface damage in metallic structures, where the electromagnetic waveguide forms a union with the structure to be inspected. The sensing concept in such jointed electromagnetic waveguides will be described here and measurement results will be presented using a hollow metallic waveguide on an aluminium sample. The experimental proof of principle is performed in the frequency band from 20 to 40 GHz using a network analyzer. It is possible to (i) detect holes as small as 1 mm, (ii) discriminate between different hole sizes, and, (iii) localize the axial position of the defect in the waveguide. This approach has great potential to be used for structural health monitoring of complex technical structures and components.

Cable surface damage detection in cable-stayed bridges using optical techniques and image mosaicking

A vision inspection system is developed for detecting surface damages on cables in long-span cable-stayed bridges. The system consists of a climbing robot, an image processing platform, and 4 fixed cameras. While the climbing robot loads the whole system and ascends along the cables, 4 cameras can continuously acquire cable surface images. Then, image processing techniques can be employed to achieve defect identification. In our optical system, a single image may only collect a part of the whole defect because of the field of view (FOV) of each camera. In order to obtain the morphology of the whole defect, we present an efficient scale-invariant feature transform (SIFT) algorithm to achieve the multi-image mosaic with partially overlapped regions in different defect images. Experimental results demonstrate that the proposed vision inspection technologies are suitable for application to the bridge cables for defect detection and condition assessment.

BRIDGE SURFACE DAMAGE DETECTION APPLICATION WITH A LASER-BASED SOFTWARE PROTOTYPE

Abstract. In order to determine the necessary actions such as rehabilitation, repair and etc. required for maintaining the operating status of aging bridges, the current conditions of these structures need to be assessed. Terrestrial laser scanners that are used to capture texture-mapped point clouds have been frequently used for assessing the current conditions of these structures. Authors’ have already developed surface damage detection algorithms in their previous work. The developed algorithms are capable of detecting several damage types from the collected texture-mapped point clouds. This short paper focuses on presenting the field application of a recently developed prototype software application that uses the previously developed surface damage detection algorithms. Capabilities of the prototype software application can be listed as importing point clouds, working with images, extracting damage features such as shape and size, determining condition ratings and producing damage reports for future investigations.

Experimental study on the vibration characteristic responses on the surface damage of wheel materials

The objective of the study was to explore the vibration signal responses on the surface damage of wheel materials using a JD-1 wheel–rail simulation facility. Vibration signals were extracted using the methods of local mean decomposition and Hilbert envelope spectrum. The surface damage of wheel rollers varies with different tangential forces. The results indicate that the surface damage of wheel materials has a corresponding characteristic frequency under different tangential forces conditions. When surface cracks appear on the surface of wheel rollers, the characteristic frequency of wheel roller is about 1830 Hz. However, the characteristic frequencies are about 800 Hz and 73 Hz for peeling and spalling damage on wheel roller. Multifractal dimensions of the vibration signals quantificationally identify and distinguish the surface damage types of wheel rollers, which can provide a meaningful guidance for the subsequent online detection of surface damage of wheel materials.

Laser-based surface damage detection and quantification using predicted surface properties

Abstract Damage due to age or accumulated effects from hazards on existing structures poses a worldwide problem. In order to evaluate the current status of aging, deteriorating and damaged structures, it is vital to accurately assess the present conditions. It is possible to capture the in situ condition of structures by using laser scanners that create dense three-dimensional point clouds. This paper investigates the use of high resolution three-dimensional terrestrial laser scans coupled with images to capture geometric range data of complex scenes for surface damage detection and quantification. Although using images with varying resolution to detect cracks is an extensively researched topic, damage detection using laser scanners with and without color images is a new research area that holds many opportunities for enhancing the current practice of visual inspections. Thus, this paper mainly focuses on combining the best features of laser scans and images to create an automatic and effective surface damage detection method, which will reduce the need for skilled labor during visual inspections and allow automatic documentation of related information. A novel surface normal-based damage detection and quantification method that uses the local surface properties extracted from laser scanner data along with color information is presented. Color data provides information in the fourth dimension that enables detecting damage types such as cracks, corrosion, and related surface defects that are generally difficult to identify using only laser scanner.

Detection of a milling-induced surface damage by the magnetic Barkhausen noise

The potential of the magnetic Barkhausen noise method for a non-destructive evaluation of the steel surface damage cased by milling was comprehensively investigated. A typical bearing steel was heat treated to three different hardnesses and then machined using the cutting tools with different degrees of the flank wear. The magnetic low-frequency measurements with a high reading depth were performed using a unique laboratory system providing a full control of the magnetization process. The high-frequency measurements were performed using a commercial Rollscan device. To study the induced magnetic anisotropy, the measurements were performed in two magnetization directions. In the feeding direction, the Barkhausen noise profiles showed a second high-field peak ascribed to an induced hardened surface layer, a so-called white layer. The most reliable results were obtained with the controlled waveform of the surface magnetic field measured directly by Hall sensors. In the perpendicular rotation direction, formation of the preferentially oriented matrix resulted in an enormously high Barkhausen noise activity. Based on these results, new magnetic parameters were proposed for the non-destructive evaluation of the white layer formation.

Analysis of Freight Wagon Wheel Failure Detection in Lithuanian Railways

The authors of this paper analyze the freight wagon wheel rolling surface damage detection process in Lithuanian Railways. The wagons were selected for this study in consideration of the Wheel Impact Load Detector (WILD) database. The results gained by experimental measurements of wheel rolling surface defects were compared with the data recorded by WILD. Distribution of failures of wheel types and dependence of the influence of the wheel-set location in a wagon on the occurrence of different defects of wheels was defined. The authors examined the reliability of WILD read-outs and wheel defect types influence on the dynamic coefficient value. Finally, basic conclusions and recommendations are given.

Identification of wall-thinning and cracks in pipes utilizing vibration modes and wavelets

Internal surface damage such as wall-thinning and cracks in piping systems are mainly caused by chemical reactions, heat, erosion, solid carrying fluids or a combination of such influences. Early detection of such surface damage is crucial in order to avoid sudden failure and the subsequent downtime losses and safety hazards. In this paper we develop methods for inner damage identification in pipes utilizing the vibration mode shapes. Wall-thinning and cracks are modeled by smooth gradual change and abrupt change in the internal surface of the pipe, respectively. Since wavelets are very efficient in detecting fast changes in signals and their derivatives, they are primarily used to detect crack development. While wavelets cannot be used to detect smooth wall-thinning changes, they can, nevertheless, be used to detect the beginning and end of such damage. We develop a new method for identifying the profile of the inner surface of the pipe, once the wavelet transform indicates that damage is taking place. The identification process requires the shape of one vibration mode of the pipe and does not require the elaborate monitoring and tracking of changes in the modal characteristics. No baseline data are needed aside from the nominal values of the pipe parameters. In addition to damage localization, the developed identification method is capable of quantizing damage severity from the identification of a more accurate profile of the inner surface of the pipe. The motivation and methods used in this work are introduced first in a general setting and then demonstrated by several pipe examples. The numerical simulations carried out here indicate the effectiveness of the developed approach.

A Noncontact Method for the Detection and Diagnosis of Surface Damage in Immersed Structures

Detection and diagnosis method is proposed for surface damage in immersed structures. It is based on noncontact ultrasonic echography measurements, signal processing tools, and artificial intelligence methods. Significant features are extracted from the measured signals and a classification method is developed to detect the echoes resulting from surface damage in an immersed structure. The identification of the damage is also provided. Gaussian neural networks trained with a specific learning algorithm are developed for this purpose. The performance of the method is validated by laboratory experiments which indicate that this method could be suitable for the monitoring of inaccessible systems like marine turbines whose unavailability causes severe economic losses.