Classification of Composite Piece Failure Types in PDC Drill Bits Based on Improved Resnet

There are failure types of PDC drill bit composite pieces such as tooth crumbling, Flat crown wear, spalling, delamination, tooth loss, etc. Failed PDC drill bit composite pieces will seriously affect the drilling efficiency, in order to deal with and repair the composite pieces of different failure types, so it is necessary to categorize the composite piece failure types. In this paper, we propose an improved Resnet-based method for classifying the failure types of PDC drill composite pieces, based on ResNet152, ECANet is introduced into the residual module, which reduces the number of parameters while maintaining the performance of the model, and at the same time, it can help the model better capture the correlation between different channels in the image; the $3 \times 3$ convolution in the residual module is replaced by DO-Conv, which improves the performance and accuracy of the model. In addition, to address the sample imbalance problem, Focal loss is chosen as the loss function during the training process, allowing the model to learn more fully from difficult to classify samples, thus improving the overall classification accuracy. The experimental results show that the algorithm in this paper achieves an accuracy of 93.1 % in the task of classifying the failure of the composite piece of a PDC drill bit. The proposed method for classifying the failure types of PDC drill composite pieces can realize the requirements for the industrial application of failure composite piece classification.