A Physics Model Embedded Hybrid Deep Neural Network for DrillstringWashout Detection

摘要

One of the practical challenges in the oil and gas industry is the lack of quality data for applying machinelearning techniques. A way to tackle this problem is to build a hybrid system that combines physics modelswith machine learning workflows. To demonstrate the applicability, the proposed hybrid model has beenapplied to drillstring washout detection which is relatively a common but severe and very expensive failurein drilling. We propose a hybrid deep neural network (hybrid-DNN) composed of three components – ParameterNetwork (PNet) for estimating model parameters, Residue Network (RNet) for predicting regression orclassification results, and a physics model appropriate for the problem at hand. PNet learns the systembehavior based on the embedded physics model, which it controls through adjusting model parameters.RNet utilizes the outputs from the PNet and physics model as input and is being trained for predicting theresidual. Once trained, the hybrid system can control the parameters of the physics model and predict thedesired results in real-time.