• Journal of the Korean Society for Nondestructive Testing
• /
• v.30 no.4
• /
• pp.302-310
• /
• 2010

In this paper, we studied Bagging neural network for predicting defect size of steam generator(SG) tube in nuclear power plant. Bagging is a method for creating an ensemble of estimator based on bootstrap sampling. For predicting defect size of SG tube, we first generated eddy current testing signals for 4 defect patterns of SG tube with various widths and depths. Then, we constructed single neural network(SNN) and Bagging neural network(BNN) to estimate width and depth of each defect. The estimation performance of SNN and BNN were measured by means of peak error. According to our experiment result, average peak error of SNN and BNN for estimating defect depth were 0.117 and 0.089mm, respectively. Also, in the case of estimating defect width, average peak error of SNN and BNN were 0.494 and 0.306mm, respectively. This shows that the estimation performance of BNN is superior to that of SNN.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.20 no.6
• /
• pp.555-561
• /
• 2000

The remote field eddy current (RFEC) inspection was performed on the ductile cast iron pipes with nominal outer diameter of 100mm, which were machined with various shapes and sizes of defects. Ductile cast iron pipes which are used as water supply pipe have the non-uniform thickness and asymmetric cross section due to relatively high degree of allowable errors during the manufacturing processes. These characteristics of ductile cast in pipes cause the long range background noises in RFEC signals along the pipe. In this study, tile machined defects in pipes were effectively classified by the moving window average (MWA) method which eliminated the long-range noise. The voltage plane polar plots (VPPP) method was used to quantitatively evaluate the depth and circumferential degree of defects. The VPPP signatures showed that the angle between defect signature and the normalized in-phase component on the VPPP is linear to the depth of defects. The nondestructive RFEC technique proved to be capable of quantitatively evaluating the machined defects of underground water supply pipe.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.29 no.2
• /
• pp.130-136
• /
• 2009

Turbine, which is one of major components in nuclear power plants, requires reliable nondestructive inspections. But, accessibility of transducers is limited and interpretation of acquired signals is not easy at all due to the complication. So, in this study, we have fabricated mock-up specimens of real size and shape. we applied pulse-echo method and time-of-flight diffraction(TOFD) method for precise inspection of turbine key and wheel bore. And phased array ultrasonic testing method was adopted for wheel dovetail of turbines by using mock-up. Furthermore, an automatic scanner system was developed for in-service inspection of the developed methods.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.1
• /
• pp.73-80
• /
• 1995

In this paper, Polarity Thresholding(PT) algorithm has been studied to enhance the received signal in ultrasonic inspection of the stainless-steel(SUS 304) which is the primary piping material of a nuclear power plant. The spectral decomposition components obtained by splitting the spectrum of received signals are composed of dispersive signal of the interference pattern produced by the grain boundaries and nondispersive signal by the flaw. PT algorithm enhance the SNR of received signal by using above properties. In experiment the stainless-steel has been chosen as the sample and heat-treated at 1125, 1150, 1175, and $1200^\circ{C}$, respectively. And the flat-bottom hole type defects have been made artificially in samples. The pulse-echo signals from the sample by using ultrasonic transducer of center frequency 5 MHz have been processed by PT algorithm. It has been shown that PT algorithm enhanced the SNR by average 14.2 dB.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.37 no.1
• /
• pp.29-36
• /
• 2017

Recently, KTX (Korean Train Express) train stoppage accidents were mainly caused by malfunctioning equipment, aging and cracking of railway vehicles, crack breakages of brake disks, and breakages of brake disks. Breakage of brake disk can cause large-scale casualties such as high-speed collision and concern about derailment by hitting lower axle and wheel. Therefore, in this study, a brake disk with solid and ventilation type, which is the brake disk of a KTX train was modeled, and a dynamometer system was constructed to operate the disk. A Rayleigh wave was used to inspect the surface of the brake disk. An ultrasonic inspection module was developed for the brake disk by using a local immersion method due to the difficulty involved in ultrasonic inspection using an existing immersion method. In addition, the surface defects of the brake disk were evaluated using a dynamometer mock-up system and an ultrasonic inspection module of the brake disk.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.24 no.1
• /
• pp.8-14
• /
• 2004

Model-based inversion tools for eddy current signals have been developed by combining neural networks and finite element modeling, for quantitative flaw characterization in steam generator tubes. In the present work, interpretation of experimental eddy current signals was carried out in order to validate the developed inversion tools. A database was constructed using the synthetic flaw signals generated by the finite element model. The hybrid neural networks composed of a PNN classifier and BPNN size estimators were trained using the synthetic signals. Experimental eddy current signals were obtained from axisymmetric artificial flaws. Interpretation of flaw signals was conducted by feeding the experimental signals into the neural networks. The interpretation was excellent, which shows that the developed inversion tools would be applicable to the Interpretation of real eddy current signals.

• Journal of Welding and Joining
• /
• v.13 no.2
• /
• pp.42-52
• /
• 1995

In the evaluation of aging degradation on the structural materials based on the fracture mechanics, the detection and size prediction of defect are very important. Aiming at nondestructive detection and size prediction ol defect with high accuracy and resolution, therefore, an lnduced Current Focusing Potential Drop(ICFPD) technique has been developed. The principle of this technique is to induce a focusing current at an exploratory region by an induction wire flowing an alternating current(AC) that is a constant ampere and frequency. Defects are assessed with the potential drops that are measured the induced current on the surface of metallic material by the potential pick-up pins. In this study, the lCFPD technique was applied for evaluating the location and size of the surface crack and blind crack made in plate specimens, and also for detecting the defects existing in valve, a field component, that were developed by SCC etc. during the service. The results of this present study show that surface crack and blind crack are able to defect with potential drop. these cracks are distinguished with the distribution of potential drop, and the crack depths can be estimated with each normalized potential drop that are parameters estimating the depth of each type crack. In the field component, the defects estimated by experiment result correspond with those in the cutting face of the measuring point within a higher sensitivity.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.17 no.4
• /
• pp.248-261
• /
• 1997

Ultrasonic pattern recognition is the most effective approach to the problem of discriminating types of flaws in weldments based on ultrasonic flaw signals. In spite of significant progress in the research on this methodology, it has not been widely used in many practical ultrasonic inspections of weldments in industry. Hence, for the convenient application of this approach in many practical situations, we develop an intelligent ultrasonic signature classification software which can discriminate types of flaws in weldments based on their ultrasonic signals using various tools in artificial intelligence such as neural networks. This software shows the excellent performance in an experimental problem where flaws in weldments are classified into two categories of cracks and non-cracks. This performance demonstrates the high possibility of this software as a practical tool for ultrasonic flaw classification in weldments.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.22 no.2
• /
• pp.187-192
• /
• 2002

The open-connect type eddy current transducer for the flaw detection in continuously connected pipelines was developed. This eddy current transducer is for the on-line inspection of the tubes in industries, to which commercial encircling probes are not applicable. The excitation coil that consists of a ribbon type cable and a flat connector can be opened and closed on purpose. The sensing coils of this transducer are circumferentially arrayed near the outside of the tube wall but axially displaced from the exciter by about one and half tube diameter. In application to steel tubes, and the performance of this transducer was evaluated as a little behind those of magnetic saturation type in signal to noise ratio and flaw size decision, but usable to detect or to locate large size flaws in steel tubes. Surface cracks deeper than 19% of the tube thickness could be detected with good signal to noise ratio.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.13 no.3
• /
• pp.24-30
• /
• 1993

The necessity of ultrasonic inspection to detect the cracks in turbine blade is being increased as the forced outage of nuclear power plants have been occurred due to blade failure in turbine components. However, the complex blade root geometry causes the ultrasonic inspection technique not to be established yet and much effort is required to set up a more reliable inspection. In this paper, the ultrasonic inspection technique for flaw detectability, skew angle effect, identification of flaw and geometric signal have been investigated with a test block and discussed the interpretation of ultrasonic signal through the acquisition and analysis of RF waveform. The experimental results show that the proper examination procedure can be established. It is required that the skew angle is essential to decrease the effect of signals from the complex blade geometry. The present results of this study can be applied to the site inspection without blade disassembly.