Research on Pore Pressure Prediction Technology of HTHP Wells in South China Sea Based on Machine Learning

Abstract

The Yingqiong Basin in the South China Sea is located at the intersection of the Eurasian and Indo-Chinese plates, with complex geology and often accompanied by abnormally high pressure. In this paper, we analyze the causes of anomalous high pressure in the South China Sea and analyze the commonly used machine learning methods, support vector machine and BP neural network, and use both methods to predict a block in Yingqiong Basin. The field application was carried out using this method, and the application showed that the prediction accuracy exceeded 95%, the complexity was reduced by 42%, and the drilling efficiency was improved by more than 53%, which played a good guiding effect to the field.