Well-Log Acoustic Velocity Prediction Based on Relevance Vector Machine

By analyzing the relation between well-log data and seismic data, a novel method for well-log acoustic velocity prediction based on relevance vector machine (RVM) is proposed. The proposed method is applied to the well in Junggar Basin and the experimental results show it has higher prediction accuracy, faster convergence speed and better generalization. Through this algorithm, we can obtain high resolution well-log acoustic velocity profile and improve the drilling simulation quality and drilling engineering design level.