A Study on the Application of Machine and Deep Learning Using the Impact Response Test to Detect Defects on the Piston Rod and Steering Rack of Automobiles.

The main parts of automobiles are the piston rod of the shock absorber and the steering rack of the steering gear, and their quality control is critical in the product process. In the process line, these products are normally inspected through visual inspection, sampling, and simple tensile tests; however, if there is a problem or abnormality, it is difficult to identify the type and location of the defect. Usually, these defects are likely to cause surface cracks during processing, which in turn accelerate the deterioration of the shock absorber and steering, causing serious problems in automobiles. As a result, the purpose of this study was to present, among non-destructive methods, a shock response test method and an analysis method that can efficiently and accurately determine the defects of the piston rod and steering rack. A test method and excitation frequency range that can measure major changes according to the location and degree of defects were proposed. A defect discrimination model was constructed using machine and deep learning through feature derivation in the time and frequency domains for the collected data. The analysis revealed that it was possible to effectively distinguish the characteristics according to the location as well as the presence or absence of defects in the frequency domain rather than the time domain. The results indicate that it will be possible to quickly and accurately check the presence or absence of defects in the shock absorber and steering in the automobile manufacturing process line in the future. It is expected that this will play an important role as a key factor in building a smart factory.

Application of a Wireless and Contactless Ultrasonic System to Evaluate Optimal Sawcut Time for Concrete Pavements.

A recently developed contactless ultrasonic testing scheme is applied to define the optimal saw-cutting time for concrete pavement. The ultrasonic system is improved using wireless data transfer for field applications, and the signal processing and data analysis are proposed to evaluate the modulus of elasticity of early-age concrete. Numerical simulation of leaky Rayleigh wave in joint-half space including air and concrete is performed to demonstrate the proposed data analysis procedure. The hardware and algorithms developed for the ultrasonic system are experimentally validated with a comparison of saw-cutting procedures. In addition, conventional methods for the characterization of early-age concrete, including pin penetration and maturity methods, are applied. The results demonstrated that the developed wireless system presents identical results to the wired system, and the initiation time of leaky Rayleigh wave possibly represents 5% of raveling damage compared to the optimal saw cutting. Further data analysis implies that saw-cutting would be optimally performed at approximately 11.5 GPa elastic modulus of concrete obtained by the wireless and contactless ultrasonic system.

A Study on the Applicability of the Impact-Echo Test Using Semi-Supervised Learning Based on Dynamic Preconditions.

The Impact-Echo (IE) test is an effective method for determining the presence, depth, and area of cracks in concrete as well as the dimensions of the sound concrete without defects. In addition, shallow delamination can be measured by confirming a flexural mode in the low-frequency region. Owing to the advancement of non-contact sensors and automated measurement equipment, the IE test can be measured at multiple points in a short period. To analyze and distinguish a large volume of data, applying supervised learning (SL) associated with various contemporary algorithms is necessary. However, SL has limitations due to the difficulty in accurate labeling for increased volumes of test data, and reflection of new specimen characteristics, and it is necessary to apply semi-supervised learning (SSL) to overcome them. This study analyzes the accuracy and evaluates the applicability of a model trained with SSL rather than SL using the data from the air-coupled IE test based on dynamic preconditions. For the detection of delamination defects, the dynamic behavior-based flexural mode was identified, and 21 features were extracted in the time and frequency domains. Three principal components (PCs) such as the real moment, real RMS, and imaginary moment were derived through principal component analysis (PCA). PCs were identical in slab, pavement, and deck. In the case of SSL considering a dynamic behavior, the accuracy increased by 7-8% compared with SL, and it could categorize good, fair, and poor status to a higher level for actual structures. The applicability of SSL to the IE test was confirmed, and because the crack progress varies under field conditions, other parameters must be considered in the future to reflect this.

Effectiveness Analysis for Smart Construction Safety Technology (SCST) by Test Bed Operation on Small- and Medium-Sized Construction Sites.

In the global construction industry, government policies have recently focused on smart construction technologies, such as those concerning the "smartization" of construction, improvements of productivity, and automation technologies. In addition, smart construction safety technologies (SCSTs) have been developed to ensure workers' safety, under the initiative of the private sector. In regards to overseas occupational safety, wearable technologies have been developed for various types of industries, and the integrated platform developments needed to link them have become mainstream. In South Korea, individual companies are focusing on developing basic SCSTs and platforms for integrated control, aiming to prevent accidents in the construction field. The goal of this study was to identify the pros and cons of SCSTs through test bed operation and to derive improvement directions. Therefore, a test bed embedded with SCSTs was built and operated to provide effective safety management for small- and medium-sized sites exposed to fatal accidents. From analyzing the data from the test bed, it was found that it is difficult to change the tendencies of workers' behaviors based solely on the introduction of SCSTs. This indicates that the effects of SCSTs are insignificant without the cooperation of workers. In addition, technical problems in field application were identified for each sensor and equipment, and the necessity, problems, and effectiveness of SCSTs were analyzed. As a result, both the installation and attachment types were found to be effective; however, workers avoided wearing certain attachment types. Based on the results derived through analysis of the pros and cons of SCSTs, the directions and guidelines were suggested for future use. This result can be used for future technology development directions, and policy establishment. Additionally, for the activation of SCSTs in the field, the cooperation of workers and the interest of managers remain essential factors, and improvements to the equipment are required.

A Study on the Detection of Internal Defect Types for Duct Depth of Prestressed Concrete Structures Using Electromagnetic and Elastic Waves.

Prestressed concrete (PSC) is widely used for the construction of bridges. The collapse of several bridges with PSC has been reported, and insufficient grout and tendon corrosion were found inside the ducts of these bridges. Therefore, non-destructive testing (NDT) technology is important for identifying defects inside ducts in PSC structures. Electromagnetic (EM) waves have limited detection of internal defects in ducts due to strong reflections from the surface of the steel ducts. Spectral analysis of the existing impact echo (IE) method is limited to specific conditions. Moreover, the flexural mode in upper defects of ducts located at a shallow depth and delamination defects inside ducts are not considered. In this study, the applicability of the elastic wave of IE was analyzed, and multichannel analysis of surface, EM, and shear waves was employed to evaluate six types of PSC structures. A procedure using EM waves, IE, and principal component analysis (PCA) was proposed for a more accurate classification of defect types inside ducts. The proposed procedure was effective in classifying upper, internal, and delamination defects of ducts under 100 mm in thickness, and it could be utilized up to 200 mm in the case of duct defect limitations.

Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques.

The static elastic modulus (Ec) and compressive strength (fc) are critical properties of concrete. When determining Ec and fc, concrete cores are collected and subjected to destructive tests. However, destructive tests require certain test permissions and large sample sizes. Hence, it is preferable to predict Ec using the dynamic elastic modulus (Ed), through nondestructive evaluations. A resonance frequency test performed according to ASTM C215-14 and a pressure wave (P-wave) measurement conducted according to ASTM C597M-16 are typically used to determine Ed. Recently, developments in transducers have enabled the measurement of a shear wave (S-wave) velocities in concrete. Although various equations have been proposed for estimating Ec and fc from Ed, their results deviate from experimental values. Thus, it is necessary to obtain a reliable Ed value for accurately predicting Ec and fc. In this study, Ed values were experimentally obtained from P-wave and S-wave velocities in the longitudinal and transverse modes; Ec and fc values were predicted using these Ed values through four machine learning (ML) methods: support vector machine, artificial neural networks, ensembles, and linear regression. Using ML, the prediction accuracy of Ec and fc was improved by 2.5-5% and 7-9%, respectively, compared with the accuracy obtained using classical or normal-regression equations. By combining ML methods, the accuracy of the predicted Ec and fc was improved by 0.5% and 1.5%, respectively, compared with the optimal single variable results.