A system for predicting the sticking risk of drill string

Abstract

The invention provides a system for predicating a drilling pipe sticking risk of a drill string. The system comprises a data acquisition and calculation module and a drilling pipe sticking risk predicating module. The data acquisition and calculation module is used for acquiring data of a well track to obtain the coordinates of each position of the well track. The geological parameter and the engineering parameter are determined according to the coordination of each position. The fuzzy quantity, contributing to generation of the drilling pipe sticking risk, of the influence factor of each position is determined by the drilling pipe sticking risk predicating module according to the geological parameter and the engineering parameter. The drilling pipe sticking risk factor of each position is obtained according to the fuzzy quantity. The drilling pipe sticking risk of any position of a full hole during the while-drilling process can be predicated, the drilling pipe sticking risk is recognized and predicated in a real-time manner, and constructive technicians are assisted in controlling the drilling risk.

Description

technical field [0001] The invention belongs to the technical field of petroleum drilling, and in particular relates to a system for predicting the sticking and sticking risk of a drill string. Background technique [0002] Since the wellbore cannot be completely vertical during drilling (especially with the large-scale application of directional wells and horizontal wells, this situation is more common), when the downhole drilling tool is stationary, under the action of the downhole pressure difference, the drilling Some parts of the column will stick to the well wall and stick together with the mud cake on the well wall. The longer the static time is, the larger the contact area between the drilling tool and the mud cake will be. For sticky stuck or differential pressure stuck. [0003] Sticking and sticking will bring great harm to oil drilling, loss of a lot of drilling time and even vicious events such as drilling tool breakage. The current prediction methods for stic...

AI Technical Summary

Problems solved by technology

[0003] Sticking and sticking will bring great harm to oil drilling, loss of a lot of drilling time and even vicious events such as drilling tool breakage

The current prediction methods for sticking pipe sticking include neural network method, decision tree method, support vector machine method and Bayesian network method. The premise of these methods is to use the historical case samples of sticking pipe sticking in adjacent wells for prediction Model training, this method is difficult to apply to the situation that a new block has not been drilled or there are few historical wells or similar failures have not occurred in historical drilling

Even if there are case samples of adjacent wells, the size of the sample size also determines the prediction accuracy of the method, that is to say, when the historical sample size is small, the prediction accuracy is very low

In addition, this method only relies on engineering parameters, which reflect the symptoms of "already produced" stuck pipe, so it is difficult to predict in advance

Images