Enhancing Weld Quality Assessment Using a Comprehensive Study on Combined Thermal and Residual Stress Through Lenet Pattern Recognition

This study aims to conduct a thorough examination of improving the evaluation of weld quality. This will be achieved by utilizing a mix of thermal and residual stress measurement, along with a pattern recognition approach employing the LeNet neural network architecture. The primary aim of this study is to provide a reliable and precise approach for assessing the quality of welds. In order to enhance the evaluation of weld quality, we propose the utilization of the LeNet convolutional neural network, which is a deep learning framework specifically designed for problems involving image identification. Photographs of weld cross-sections are analyzed using the network, which lets visual patterns of both thermal and residual stress distributions be captured. The findings of our research indicate a notable enhancement in the precision of weld quality evaluation when compared to conventional approaches. We use thermal and residual stress data along with deep learning techniques to make our method much better at finding and classifying defects. The proposed methodology exhibits significant potential for enhancing the overall safety and performance of welded structures across many sectors.