Machine-learning-based predictive classifier for sand body types and reservoir properties using geological data

The classification and identification of reservoir sand body types has important research significance in oil-gas oilfield development plans. The traditional recognition method mainly relies on logging curves and manual recognition, which has the disadvantages of relying on expert experience and longtime consumption. This article is driven by geological data and compares various machine learning algorithms to select the intelligent algorithm suitable for solving this problem, achieving automatic recognition of reservoir sand body types and providing reference for reservoir engineering. The results show that the accuracy of the neural network algorithm reaches 97.4%, which can be used as an effective intelligent algorithm for reservoir sand body type classification and recognition.