For the automation of a laser beam welding (LBW) process, the weld quality must be monitored without destructive testing, and the quality must be assessed. A deep neural network (DNN)-based quality assessment method in spectrometry-based LBW is presented in this study. A spectrometer with a response range of 225–975 nm is designed and fabricated to measure and analyze the light reflected from the welding area in the LBW process. The weld quality is classified through welding experiments, and the spectral data are thus analyzed using the spectrometer, according to the welding conditions and weld quality classes. The measured data are converted to RGB (red, green, blue) values to obtain standardized and simplified spectral data. The weld quality prediction model is designed based on DNN, and the DNN model is trained using the experimental data. It is seen that the developed model has a weld-quality prediction accuracy of approximately 90%.

Nondestructive test (NDT) technology is required in the gas metal arc (GMA) welding process to secure weld robustness and to monitor the welding quality in real-time. In this study, a laser vision sensor (LVS) is designed and fabricated, and an image processing algorithm is developed and implemented to extract precise laser lines on tested welds. A camera calibration method based on a gyro sensor is used to cope with the complex motion of the welding robot. Data are obtained based on GMA welding experiments at various welding conditions for the estimation of quality prediction models. Deep neural network (DNN) models are developed based on external bead shapes and welding conditions to predict the internal bead shapes and the tensile strengths of welded joints.

No abstract

No abstract

Atomic force microscopy (AFM) has been used as a tool, not only for imaging surfaces, but also for measuring surface forces and mechanical properties at the nano-scale. Force calibration is crucial for quantitatively measuring the forces that act between the AFM probe of a force sensing cantilever and a sample. In this work, the lateral force calibrations of a V-shaped cantilever were performed using the finite element method, multiple pivot loading, and thermal noise methods. As a result, it was shown that the multiple pivot loading method was appropriate for the lateral force calibration of a V-shaped cantilever. Further, through crosschecking of the abovementioned methods, it was concluded that the thermal noise method could be used for determining the lateral spring constants as long as the lateral deflection sensitivity was accurately determined. To obtain the lateral deflection sensitivity from the sticking portion of the friction loop, the contact stiffness should be taken into account.