A Identification Method of Logging Mud Ratio is Based on Deep Learning
Minghui Lu, Yuxue Wang · Journal of Physics Conference Series · 2021
During the logging, the sand ratio of rock debris is an important index to describe the oil content of the well at a certain depth, which can reflect the well depth of the bottom of the drill rod in different kinds of rock layers. The traditional calculation method of ratio sandstone to mudstone is artificial recognition, which takes a long time and has a low accuracy. This text applies the deep learning method to realize the automatic recognition calculation of ratio sandstone to mudstone, which has low cost, fast speed and high accuracy, and is suitable for practical field application. This text first uses the morphological watershed algorithm to divide the rock chips and obtain the rock particles, and then uses the convolutional neural network to calculate ratio sandstone to mudstone.