Detection and Prediction of Crane Arms Based on Deep Learning

Welding technology has been widely applied in aerospace, national defense science and industry, machinery manufacturing, petrochemical industries, and other important fields. Due to the influence of subjective factors on artificial identification, there are misdetection, omission, and other issues. This paper focuses primarily on welding deformation detection and prediction technology of crane boom barrels using deep learning in intelligent manufacturing. Deep learning is used in this paper to directly extract classification features from input images. A multi-layer convolutional network is used to extract high-level abstract features to find the common features of welding defect images. For various types of cranes, it is possible to achieve accurate and rapid detection of welding flaws. The dropout drop rate is optimized to further enhance the model’s recognition capability. Experimental results demonstrate that the welding defect detection method of crane boom barrels based on deep learning proposed in this paper can effectively detect welding defects in crane boom barrels.