Multivariate Nonlinear Statistical Analysis Of Shallow Cluster Horizontal Well Drill String Friction Prediction Method

Drilling string friction has traditionally been an important research topic in the field of engineering technology.It's very important for us to drill a well in a high quality,highly efficient and fast way.A reasonable drill string friction prediction method can provide a strong guarantee for drilling construction safety prediction.Based on the research of former experts,firstly,this paper studies the characteristics and applications and the bottom hole assembly's characteristics of the shallow cluster horizontal wells.Systematically analyzes the drilling construction of drill string friction factors.And then analyzes the drilling parameters,the tubing string,borehole conditions,drilling fluid performance influence on drill string friction.This paper also studies the current model of drill string mechanics,understand its characteristics and its application limitations.Secondly,applied statistics theory,according to the multivariate nonlinear statistical analysis,put forward forecast cluster horizontal friction method research train of thought,and determine the segmented statistical analysis principle.According to the dividing principle,segmented statistical analysis of shallow cluster horizontal friction problem,laid the foundation for improving the prediction accuracy.Then SPSS mathematical software partial correlation analysis was carried out for the friction factors,Finally,apply the multivariate nonlinear regression analysis of mathematical software for SPSS,conducted a multivariate nonlinear regression analysis for the related factors,obtained the subsection regression model,and conducted the relevant inspection.Through regression analysis concluded index additive model and the power function multiplication model.From relevant inspection,we can get to know that the index additive model is not in accordance with the requirements,while the power function multiplication model through the inspection,illustrated the rationality of the model.Expectations can be applied to cluster horizontal well drill string friction prediction,applied to drilling engineering design and construction guide.