Real-time prediction method for drilling leakage layer position based on decision tree data mining

Abstract

The embodiment of the invention provides a real-time prediction method for a drilling leakage layer position based on decision tree data mining. The method comprises the steps that related historicaldrilling data, historical drilling leakage layer position true values and real-time drilling data of a target block are collected, and data preprocessing is conducted; dividing the preprocessed historical drilling data into a training set and a test set, selecting the historical drilling data in the training set as input and a historical drilling leakage layer position true value as a standard, and performing data mining by adopting a decision-making tree algorithm to form a leakage layer position prediction initial decision-making tree; pruning the initial decision tree through the data of the test set, selecting an optimal sub-tree through cross validation and performing precision evaluation, and regenerating the decision tree if the precision requirement is not met; and generating a real-time prediction model of the drilling leakage layer position according to the classification rule determined by the final decision tree, and connecting the model with a drilling data real-time collection platform, thereby realizing an effect of real-time prediction of the drilling leakage layer position.

Description

technical field [0001] The invention relates to a method for real-time prediction of the position of a drilling leaky layer based on decision tree data mining, and belongs to the field of decision tree data mining and the field of drilling fluid loss plugging. Background technique [0002] Lost circulation is one of the most serious complex situations that affect the safety of drilling operations. Lost circulation not only brings great difficulties to drilling engineering, but also seriously affects the development speed of oil and gas resources. Finding out the location of the leakage layer is the key to formulate technical measures for leakage prevention and plugging, and to minimize the loss caused by the leakage. [0003] At present, the methods for measuring the location of the leakage layer at home and abroad mainly include well temperature method, acoustic wave measurement method, turbine flowmeter method, noise method, sensor measurement method, electromagnetic measu...

AI Technical Summary

Problems solved by technology

Judging from the current application situation, these methods all have certain shortcomings: the measurement accuracy of the well temperature method is affected by the slow heat conduction velocity, and the temperature change is not obvious when the leakage is small, and the position of the leakage layer cannot be accurately determined, only the approximate range; The measurement method has multiple solutions, and it is easy to misjudge irregular well sections, fractures without leakage, and gas-bearing intervals as leaky layers; the turbine flowmeter method is greatly affected by drilling fluid parameters, and it is easy to produce measurement deviations; noise The method is easily affected by the friction noise between the instrument and the borehole wall; most of the instruments such as sensors, electromagnetic measuring instruments, and t

Images