Study of Machine Learning Methods in the Analysis of Production and Maintenance of Oil Wells

In this article, methods for determining significant features, as well as forecasting methods for optimizing oil drilling, were considered, their advantages and disadvantages were determined. To identify the most significant features that affect the amount of oil produced, feature selection was carried out based on correlation analysis, as well as comparison and identification of effective neural network training methods. As a result, the BRANN (artificial neural network with Bayesian regularization) method was chosen, as it showed the best results with a relatively short training time. Thus, the neural network was trained to predict the amount of oil produced from oil wells.