A Prediction Method of Drilling Loss Probability Based on Naive Bayesian Algorithm

Abstract

The invention belongs to the fields of drilling loss prediction and artificial intelligence machine learning, and in particular relates to a drilling loss probability prediction method based on a naive Bayesian algorithm. The method is as follows: extract historical reservoir development data from a reservoir well history database in a certain reservoir development zone and perform preprocessing, create a sample set and a test set, determine characteristic attributes and output category space from the sample data, and according to the sample Collect the data of drilling loss parameters and count the prior probability of each characteristic attribute value, and use the prior probability learning to calculate the conditional probability. The method calculates the corresponding conditional probability to improve the generalization performance of the model, calculates the posterior probability jointly by the prior probability, establishes the naive Bayesian model classifier, uses the verification set for verification, and inputs the real-time drilling parameter data into the model to obtain the corresponding loss probability size.

Description

technical field [0001] The invention belongs to the fields of drilling loss prediction and artificial intelligence machine learning, and in particular relates to a drilling loss probability prediction method based on a naive Bayesian algorithm. Background technique [0002] Lost circulation is a complex and common working condition in the drilling process, which will cause serious harm and economic loss to the drilling industry. During the drilling process, many drilling parameters will affect lost circulation, such as well depth, layer, lithology, weight on bit, torque, speed of penetration, vertical pressure, inlet flow, outlet flow, porosity, permeability, drilling fluid density, drilling Liquid-solid phase content, Young's modulus, bit type, bit size, hook load, etc. The changes of these drilling parameters during the drilling process may make the drilling fluid column pressure greater than the formation rock fracture pressure, thereby forming induced fractures, or reop...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the insufficiency and limitation of the existing lost circulation prediction methods, which cause the well site technicians to be unable to control the corresponding lost circulation parameters, making the lost circulation difficult to control and occur repeatedly, and enrich the methods of predicting lost circulation by using drilling parameters , the present invention proposes a drilling loss probability prediction method based on naive Bayesian algorithm

Images