Real-time recognition method for overflow and leakage based on random forest

Abstract

The invention provides a real-time recognition method for overflow and leakage based on the random forest, which comprises the steps of (1) automatically generating an overflow and leakage original data set; (2) preprocessing the overflow and leakage data set so as to enable the overflow and leakage data set to become a training data set capable of being applied to a random forest algorithm; (3) generating overflow and leakage sub-datasets based on a Bootstrap sampling technology; (4) creating a classification tree for each sub-dataset by adopting a CART algorithm; and (5) collecting prediction results of n trees for a new sample, deciding the category by adopting a voting mode, and recognizing the occurrence of overflow and leakage by using a voting result of the classification trees. Themethod provided by the invention automatically judges the drilling state according to real-time drilling data, calculates the expected wellhead flow and compares the expected wellhead flow with the measured wellhead flow, thereby being capable of performing real-time judgment on overflow and leakage occurred at different drilling states. The real-time judgment method for overflow and leakage based on the random forest is simpler in modeling method and more accurate in prediction for accidents.

Description

technical field [0001] The invention relates to the technical field of drilling engineering, in particular to a real-time identification method for overflow and leakage accidents during drilling. Background technique [0002] Drilling is an important means of exploration and development of oil and gas. In drilling construction operations, drilling complications and accidents threaten the entire drilling process from beginning to end, seriously affecting drilling speed, well construction quality and exploration and development benefits. Among them, lost circulation and overflow are the two most common downhole complex accidents that affect drilling safety. Lost circulation and overflow will not only cause serious reservoir damage, increase the investment cost of exploration and development, and cause low oil and gas development efficiency, but also cause major vicious accidents such as pipe sticking, well collapse, and blowout if the control is not effective, causing losses a...

AI Technical Summary

Problems solved by technology

This application adopts traditional data interpretation and calculation methods, and presents the calculation results in the form of a comprehensive result map. The monitoring results are lagging, and cannot well solve the problem of downhole overflow and leakage monitoring in special environments such as high-pressure gas wells and deep-water drilling.

Images