Defect Detection in CK45 Steel Structures through C-scan Images Using Deep Learning Method

In the present paper, a method for reliable estimation of defect profile in CK45 steel structures is presented using an eddy current testing based measurement system and post-processing system based on deep learning technique. So a deep learning method is used to determine the defect characteristics in metallic structures by magnetic field C-scan images obtained by an anisotropic magneto-resistive sensor. Having designed and adjusting the deep convolution neural network and applied it to C-scan images obtained from the measurement system, the performance of deep learning method proposed is compared with conventional artificial neural network methods such as multilayer perceptron and radial basis function on a number of metallic specimens with different defects. The results confirm the superiority of the proposed method for characterizing defects compared to other classical training-oriented methods.

Development for wind friction velocity retrieval algorithm based on the SFMR and NOAA dropwindsondes measurements in hurricane conditions

<p>The work is concerned with the development of a method for the retrieval of tropical cyclones boundary atmospheric layer parameters, namely the wind friction velocity and wind speed at meteorological height. For the analysis, we used the results of field measurements of wind speed profiles from dropwindsondes launched from National Oceanic and Atmospheric Administration (NOAA) aircraft and collocated data from the Stepped-Frequency Microwave Radiometer (SFMR) located onboard of the same aircraft.</p><p>The results of radiometric measurements were used to obtain the emissivity values, which were compared with the field data obtained from the falling dropwindsondes. Using the algorithm taking into account the self-similarity of the velocity defect profile (Ermakova et al., 2019), the parameters of the atmospheric boundary layer were determined from the data measured by dropwindsondes. This algorithm gives an opportunity to obtain the wind speed value at meteorological height and wind friction velocity from the averaged data in the wake part of the profiles of the marine atmospheric boundary layer.</p><p>A comparison of the wind speed U10 dependencies, retrieved from the SFMR data and measurements from dropwindsondes, with the similar dependencies obtained in (Uhlhorn et al., 2007), was made, and their satisfactory agreement was demonstrated. This work was supported by the RFBR projects No. 19-05-00249, 19-05-00366.</p>

Word Line Defect Localization Methods in 2D NAND Flash

In this work, two analysis methods for word line (WL) defect localization in NAND flash memory array are presented. One is to use the Emission Microscope (EMMI) and Optical Beam Induced Resistance Change (OBIRCH) to analyze the device through backside, which has no risk of damage during sample preparation. Depending on the I-V characteristics of defects, different analysis tools can be applied. The second method is to analyze a device defect location that is hard to detect through backside analysis. The precise defect site can be localized by Electron Beam Induce Resistance Change (EBIRCH) [1,2], and the defect profile can be observed. The large memory array in NAND flash structure leads to the wide sample movement during EBIRCH analysis. The sub-stage movement function used successfully solves this problem.

A nonlinear dynamic vibration model of a defective bearing: the importance of modelling the angle of the leading and trailing edges of a defect

Rolling element bearings eventually become worn and fail by developing surface defects, such as spalls, dents and pits. Previous researchers have tested bearings with defects that have sharp [Formula: see text] rectangular edges that were used to develop analytical models of a defective bearing. These models have limitations that require smooth surfaces and constant curvature of the bearing components; as well as assuming the defect profile. A method has been created to capture the surface topography of a bearing defect. A numerical model has been developed for a rolling element bearing that uses the measured defect profile and removes the limitations of models by previous researchers that use analytical expressions for contact area and force. The predicted vibration response of a bearing with a defect that has sloped leading and trailing edges on the outer and inner raceway was compared with experimental results. It was found that the new numerical model was able to predict the vibration response of a defective bearing. The defect topographies and the developed model have been made publicly available.

Quantitative research of defects for pipelines based on metal magnetic memory testing

Pipelines are widely used in the oil & gas industry but defects seriously affect their safe operation. Therefore, it is necessary to perform non-destructive testing (NDT) to quantify the defects. In this study, a magnetic dipole model was established to characterise the defects and magnetic flux leakage (MFL) of defects was simulated using the finite element method (FEM) to reveal the spatial distribution of the magnetic vector. Magnetic signals were measured using a tunnel magnetoresistance (TMR) sensor array and the results showed that defects with different sizes could be quantified using the metal magnetic memory (MMM) method. Three-dimensional distribution of the magnetic signal and its gradient reflected the defect profile well and the gradient of the magnetic signal was found to reduce the effect of non-uniform magnetisation. Furthermore, experimental results were verified by comparison with the simulation results and the comparison results showed a consistent variation trend. Quantitative analysis was conducted and the characteristic parameters of the gradients could be used to quantify the defects.

Vibration Analysis of a Roller Bearing With a Bump Defect

Vibration performances of roller bearings are greatly affected by various localized defects. Thus, it is very important to analyze vibration characteristics of the roller bearings with the localized defects. In this paper, a nonlinear dynamic model is developed to formulate the effect of a localized defect on the vibration characteristics of a cylindrical roller bearing (CRB). A bump defect is formulated in this model. The defect profile is defined as a spherical one. The time-varying displacement excitation caused by the defect is modelled. Effects of the defect sizes on the vibration characteristics of the roller bearing are discussed. The simulation results show that the developed method can provide some guidance for understanding the vibration characteristics of the CRB with a bump defect.