Intelligent Prediction of Stuck Pipe Using Combined Data-Driven and Knowledge-Driven Model
Shuo Zhu, Xianzhi Song, Zhaopeng Zhu, Xuezhe Yao, Muchen Liu
Abstract
Stuck pipe phenomena can have disastrous effects on drilling performance, with outcomes that can range from time delays to loss of expensive machinery. In this work, we provide three methods for the prediction of stuck pipe. The first method targets the detection of friction coefficient which can represent the trend of stuck pipe. The second method targets the prediction of probability for stuck pipe using ANN (artificial neural network). The last model establishes a comprehensive indicator based on the first and the second method using fuzzy mathematics which can give more accurate probability for stuck pipe. The results show that the best model is the last one which can predict stuck pipe events with a F1 of 0.98 and a FAR (false alarm rate) of 1%. Preliminary experimental results on the available dataset indicate that the use of the proposed model and can help mitigate the stuck pipe issue.