A Prediction Method of Well Deviation Angle Based on Integrated Learning Algorithm

Abstract

The invention discloses a well deviation angle prediction method based on an integrated learning algorithm, which uses an SVR learner model, a neural network regression algorithm learner model, a random forest regression algorithm learner model and a Gaussian regression algorithm learner model in machine learning , learn the learning samples composed of the drilled wellbore trajectory data, drilling mode and bottom hole structural parameters of a certain well, train the above four learner models, and predict the inclination angle of the blind area at the bottom of the well respectively, and then Linear regression is performed between the training result and the target value to obtain the final prediction result. The verification results of the actual drilling data show that the method of the present invention has high prediction accuracy, effectively reduces the error of the traditional fixed curvature extrapolation method in predicting the bottom hole inclination, and improves the accuracy of predicting the inclination angle.

Description

technical field [0001] The invention belongs to the technical field of oil and natural gas drilling equipment, in particular to a well deviation angle prediction method based on an integrated learning algorithm applied in the technical field of rotary steering systems. Background technique [0002] In recent years, due to the difficulty and high cost of downhole measurement while drilling near the drill bit, it is basically monopolized by the three major oil companies. At present, MWD and LWD wireless while drilling instruments are mainly used in China, and the measurement point is 8-20m away from the drill bit. Therefore, the measurement point The data cannot correctly reflect the actual wellbore conditions at the position of the drill bit, which brings certain difficulties to the prediction and control of the wellbore trajectory. In the traditional wellbore trajectory prediction, there are a large number of influencing parameters related to drilling construction, and it is...

AI Technical Summary

Problems solved by technology

However, with the advancement of drilling technology and the rise of directional and horizontal drilling technologies, the traditional wellbore trajectory prediction method is increasingly unsuitable for modern drilling monitoring. Therefore, it is urgent to develop a new bottomhole inclination angle prediction method. to overcome the above shortcomings

Images