• Structural Engineering and Mechanics
• /
• v.91 no.5
• /
• pp.443-457
• /
• 2024

At present, the traditional concrete surface inspection methods based on artificial vision have the problems of high cost and insecurity, while the computer vision methods rely on artificial selection features in the case of sensitive environmental changes and difficult promotion. In order to solve these problems, this paper introduces deep learning technology in the field of computer vision to achieve automatic feature extraction of structural damage, with excellent detection speed and strong generalization ability. The main contents of this study are as follows: (1) A method based on DeepLabV3+ convolutional neural network model is proposed for surface detection of post-earthquake structural damage, including surface damage such as concrete cracks, spaling and exposed steel bars. The key semantic information is extracted by different backbone networks, and the data sets containing various surface damage are trained, tested and evaluated. The intersection ratios of 54.4%, 44.2%, and 89.9% in the test set demonstrate the network's capability to accurately identify different types of structural surface damages in pixel-level segmentation, highlighting its effectiveness in varied testing scenarios. (2) A semantic segmentation model based on DeepLabV3+ convolutional neural network is proposed for the detection and evaluation of post-earthquake structural components. Using a dataset that includes building structural components and their damage degrees for training, testing, and evaluation, semantic segmentation detection accuracies were recorded at 98.5% and 56.9%. To provide a comprehensive assessment that considers both false positives and false negatives, the Mean Intersection over Union (Mean IoU) was employed as the primary evaluation metric. This choice ensures that the network's performance in detecting and evaluating pixel-level damage in post-earthquake structural components is evaluated uniformly across all experiments. By incorporating deep learning technology, this study not only offers an innovative solution for accurately identifying post-earthquake damage in civil engineering structures but also contributes significantly to empirical research in automated detection and evaluation within the field of structural health monitoring.

• Journal of Advanced Research in Ocean Engineering
• /
• v.1 no.4
• /
• pp.198-210
• /
• 2015

This paper presents the damage assessment of a free-fall dropped object on the seabed. The damage to a dropped object totally depends on the relationship between the impact energy and the soil strength at the mudline. In this study, unexpected dropping scenarios were first assumed by varying the relevant range of the impact velocity, structure geometry at the moment of impact, and soil strength profile along the penetration depth. Theoretical damage assessments were then undertaken for a free-fall dropping event with a fixed final embedment depth for the structure. This paper also describes the results of a three-dimensional large deformation finite element analysis undertaken for the purpose of validation. The analyses were carried out using the coupled Eulerian-Lagrangian approach, modifying the simple elastic-perfectly plastic Tresca soil model. The validation exercises for each dropping scenario showed good agreement, and the present numerical approach was capable of predicting the behavior of a free-fall dropped object.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.7 no.4
• /
• pp.121-128
• /
• 2003

In recent years, various artificial neural network algorithms are used in the damage assessment of civil infrastructures. So far, many researchers have used the artificial neural network as a pattern classifier for the structural damage assessment but, in this paper, the neural network is used as a structural reanalysis tool not as a pattern classifier. For the model updating using the optimization algorithm, the summation of the absolute differences in the structural vibration modes between undamaged structures and damaged ones is considered as an objective function. The stiffness of structural components are treated as unknown parameters to be determined. The structural damage detection is achieved using model updating based on the optimization techniques which determine the estimated stiffness of components minimizing the objective function. For the verification of the proposed damage identification algorithm, it is numerically applied to a simply supported bridge model.

• Computers and Concrete
• /
• v.34 no.4
• /
• pp.409-425
• /
• 2024

This paper utilizes LS-DYNA software to numerically investigate impact response and damage evaluation of fiber-reinforced polymer (FRP) bars-reinforced ultra-high-performance concrete (UHPC) composite beams (FRP-UHPC beams). Three-dimensional finite element (FE) models are established and calibrated by using literature-based static and impact tests, demonstrating high accuracy in simulating FRP-UHPC beams under impact loading. Parametric analyses explore the effects of impact mass, impactor height, FRP bar type and diameter, and clear span length on dynamic response and damage modes. Two failure modes emerge: tensile failure with bottom longitudinal reinforcement fracture and compression failure with local concrete compression near the impact region. Impact mass or height variation under the same impact energy significantly affects the first peak impact force, but minimally influences peak midspan displacement with a difference of no more than 5% and damage patterns. Increasing static flexural load-carrying capacity enhances FRP-UHPC beam impact resistance, reducing displacement deformation by up to 30%. Despite similar static load-carrying capacities, different FRP bars result in varied impact resistance. The paper proposes a damage assessment index based on impact energy, static load-carrying capacity, and clear span length, correlating well with beam end rotation. Their linearly-fitting coefficient was 1.285, 1.512, and 1.709 for the cases with CFRP, GFRP, and BFRP bars, respectively. This index establishes a foundation for an impact-resistant design method, including a simplified formula for peak midspan displacement assessment.

• Composites Research
• /
• v.26 no.2
• /
• pp.79-84
• /
• 2013

The prediction and analysis procedure of fatigue damage crack growth life for a stiffener bonded composite laminate panel including center hole and edge notch damage, was studied. It was performed on the basis of fatigue damage growth test results on a laminated skin panel specimens and the analysis results of stress intensity factor for the stiffener bonded composite panel. According to the comparison between experimental test and prediction results of fatigue damage growth life, it was concluded that the residual strength and damage tolerance assessment can be carried out along to the edge notch crack growth.

• Nuclear Engineering and Technology
• /
• v.50 no.8
• /
• pp.1217-1233
• /
• 2018

Following a surge of interest in multi-unit risk in the last few years, many recent studies have suggested methods for multi-unit probabilistic safety assessment (MUPSA) and addressed several related aspects. Most of the existing studies though focused on two-unit nuclear power plant (NPP) sites or used rather simplified probabilistic safety assessment (PSA) models to demonstrate the proposed approaches. When considering an NPP site with three or more units, some approaches are inapplicable or yield very conservative results. Since the number of such sites is increasing, there is a strong need to develop and validate practical approaches to the related MUPSA. This article provides several detailed approaches that are applicable to multi-unit Level 1 PSA for sites with up to six or more reactor units. To validate the approaches, a multi-unit Level 1 PSA model is developed and the site core damage frequency is estimated for each of four representative multi-unit initiators, as well as for the case of a simultaneous occurrence of independent single-unit initiators in multiple units. For this purpose, an NPP site with six identical OPR-1000 units is considered, with full-scale Level 1 PSA models for a specific OPR-1000 plant used as the base single-unit models.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.12 no.2
• /
• pp.139-148
• /
• 2008

The direct inspection of the outward aspects by field engineers is the important and critical part for structural safety assessment according to the related reports. This study presents an improved method of the state assessment for concrete floor slab by separating and evaluating the individual damage types. First, the various types of damage symptoms are separated, which have been included and dealt in a group. Secondly, they are weighted and scored independently based on the present guide and references. Overall procedures other than the above are retained as same as possible to avoid the confusion. The proposed method is applied and tested to a performed assessment project for a bridge for validation. The result shows that it is reasonable and applicable in respect that it is able to make up for the controversial points of the present guide revealed in practices. Careful check of excessively deteriorated parts in addition to the reasonable assessment of system by this method grants the structural repair and reinforcement propriety and economy, and assures of more safety. Twofold appraisal of this approach expands the applicable areas of value engineering to the structural maintenance.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2000.04b
• /
• pp.297-303
• /
• 2000

Damage estimation of a bridge structure is presented using ambient vibration data caused by the traffic loadings. The procedure consists of identification of the modal properties and assessment of the damage locations and severities. An experimental study is carried out on the bridge model subjected to vehicle loadings. Vertical accelerations of the bridge deck are measured at a limited number of locations. The modal parameters are identified from the free vibration signals extracted using the random decrement method. Then, the damage assessment is carried out based on the estimated modal parameters using the neural networks technique. The identified damage locations and severities agree reasonably well with the inflicted damages on the structure.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2005.04a
• /
• pp.367-374
• /
• 2005

The use of system identification approaches for damage detection has been expanded in recent years. Soft computing techniques such as neural networks have been utilized increasingly. Damage assessment using neural networks is presented in this study. Data set for training neural networks are acceleration response of simple beam under the various damage states ,which are the inputs. The outputs are the damage locations and extents. Not only the trained damages but also untrained damages are. detected accuratelyintheassessmentstage.

• Computers and Concrete
• /
• v.7 no.4
• /
• pp.285-301
• /
• 2010

In this work we first review the statistical data on large fires in urban areas, presenting a detailed list of causes of fires, the type of damage to concrete and reinforced concrete structures. We also present the modern experimental approach for studying the fire-resistance of different structural components, along with the role of numerical modeling to provide more detailed information on quantifying the temperature and heat flux fields. In the last part of this work we provide the refined models for assessment of fire-induced damage in structures built of concrete and/or reinforced-concrete. We show that the refined models of this kind are needed to provide a more thorough explanation of damage and to complete the damage assessment and post-fire evaluations.