CN-121997692-A - RBF-based drilling leakage speed prediction method, system, equipment and medium

CN121997692ACN 121997692 ACN121997692 ACN 121997692ACN-121997692-A

Abstract

The invention belongs to the technical field of petroleum exploration and development, and provides a drilling leakage speed prediction method, a drilling leakage speed prediction system, electronic equipment and a storage medium based on RBF. The method comprises the steps of collecting historical drilling data and real-time drilling data of a target block, carrying out data preprocessing, obtaining a parameter vector and a hyperplane formula according to the preprocessed historical drilling data, carrying out RTF dual conversion on the parameter vector segmented according to the hyperplane formula to obtain a nuclear matrix of an RBF support vector machine, generating a sensor model according to the parameter vector, the hyperplane formula and the nuclear matrix to obtain a drilling loss speed real-time prediction model based on the RBF support vector machine, and predicting the real-time drilling data by utilizing the real-time prediction model to obtain a drilling loss speed real-time prediction result. The invention achieves the effect of predicting the leakage speed in real time and provides more accurate and effective decision basis for drilling plugging technicians and constructors.

Inventors

WANG XUDONG
WU PENGCHENG
CHEN YE
GAO DEWEI
LI ZHENGTAO
MA CHENGYU

Assignees

中国石油天然气股份有限公司

Dates

Publication Date: 20260508
Application Date: 20241107

Claims (10)

1. An RBF-based drilling leak rate prediction method, comprising: collecting historical drilling data and real-time drilling data of a target block, and performing data preprocessing; Obtaining a parameter vector and a hyperplane formula according to the preprocessed historical drilling data, and performing RTF dual conversion according to the parameter vector segmented by the hyperplane formula to obtain a kernel matrix of an RBF support vector machine; and generating a sensor model according to the parameter vector, the hyperplane formula and the nuclear matrix to obtain a drilling leakage speed real-time prediction model based on an RBF support vector machine, and predicting real-time drilling data by using the real-time prediction model to obtain a drilling leakage speed real-time prediction result.
2. The prediction method according to claim 1, wherein the data preprocessing includes: The data noise reduction is carried out, historical drilling data is judged through a box graph method, irrelevant data, repeated data and smooth noise data are obtained, and the irrelevant data, the repeated data and the smooth noise data are deleted; Data is subjected to data complementation, irrelevant data obtained by deleting in the data noise reduction process is subjected to k nearest neighbor algorithm, repeated data are obtained, and the smooth noise data is complemented; converting data, namely converting stratum lithology and character parameters of drill bit types in data obtained by data noise reduction by a single-heat coding method into digital parameters; Data integration, merging and processing historical drilling data in a multi-file or multi-database operating environment.
3. The method of predicting as set forth in claim 1, wherein said deriving a parameter vector and a hyperplane formulation from the preprocessed historical drilling data comprises: Dividing the preprocessed historical drilling data into a training set and a testing set; normalizing the training set and the testing set according to different parameters, and mapping the training set and the testing set into parameter vectors; And carrying out Gaussian distribution mapping on the parameter vector to obtain a hyperplane formula.
4. A prediction method according to claim 3, wherein the generating a perceptron model comprises: testing the sensor model by using the test set; and if the test result does not meet the precision requirement, regenerating the sensor model.
5. The prediction method according to claim 4, wherein the obtaining the real-time prediction model of the drilling loss rate based on the RBF support vector machine comprises: and connecting the sensor model meeting the precision requirement with an instant acquisition platform for obtaining real-time drilling data to obtain a drilling leakage speed real-time prediction model based on the RBF support vector machine.
6. The drilling leakage speed prediction system based on RBF is characterized by comprising a data acquisition module, a data processing module and a model generation module; the data acquisition module is used for acquiring historical drilling data and real-time drilling data of the target block and carrying out data preprocessing; The data processing module is used for obtaining a parameter vector and a hyperplane formula according to the preprocessed historical drilling data, and performing RTF dual conversion on the parameter vector segmented according to the hyperplane formula to obtain a kernel matrix of an RBF support vector machine; And the model generation module is used for generating a perceptron model according to the parameter vector, the hyperplane formula and the nuclear matrix to obtain a drilling leakage speed real-time prediction model based on an RBF support vector machine, and predicting real-time drilling data by using the real-time prediction model to obtain a drilling leakage speed real-time prediction result.
7. The prediction system of claim 6, comprising: the data acquisition module is also used for data noise reduction, historical drilling data is judged through a box graph method, irrelevant data, repeated data and smooth noise data are obtained, and the irrelevant data, the repeated data and the smooth noise data are deleted; The data acquisition module is also used for data complementation, and the k nearest neighbor algorithm is used for completing the obtained irrelevant data deleted in the data denoising process, the obtained repeated data and the smooth noise data; The data acquisition module is also used for data conversion, and characters parameters of formation lithology and drill bit type in the data obtained by data noise reduction are converted into digital parameters through a single-heat coding method; The data acquisition module is also used for data integration, merging and processing historical drilling data in a multi-file or multi-database operating environment.
8. The prediction system of claim 6, comprising: The data processing module is also used for dividing the preprocessed historical drilling data into a training set and a testing set; The data processing module is also used for carrying out normalization processing on the training set and the testing set according to different parameters and mapping the training set and the testing set into parameter vectors; The data processing module is also used for carrying out Gaussian distribution mapping on the parameter vector to obtain a hyperplane formula.
9. An electronic device, comprising: A processor; and a memory storing computer-executable instructions that, when executed, cause the processor to perform the method of any of claims 1-5.
10. A computer storage medium, wherein the computer storage medium stores one or more programs which, when executed by a processor, implement the method of any of claims 1-5.

Description

RBF-based drilling leakage speed prediction method, system, equipment and medium Technical Field The invention belongs to the technical field of petroleum exploration and development, and particularly relates to a drilling leakage speed prediction method, a drilling leakage speed prediction system, electronic equipment and a storage medium based on RBF. Background Lost circulation is a common complex condition in a well in a drilling engineering, most drilling processes have different degrees of lost circulation, severe lost circulation can lead to the reduction of pressure in the well, influence normal drilling, cause instability of a well wall, induce formation fluid to flow into the well bore and cause blowout, and finally cause disastrous accidents. The current method for measuring the leakage speed mainly comprises the steps of (1) judging the approximate leakage speed through repeated circulation after the leakage is found, then periodically replenishing liquid, periodically using a liquid level monitor to measure the liquid level, and judging the leakage speed according to the change condition of the liquid level. The method is simple to operate, low in accuracy, not only can cause a large amount of drilling fluid waste and stratum pollution, but also has limited effects, and is not suitable for the leakage of most large cracks, and (2) the imaging logging technology is utilized to judge the properties of the cracks, and then the leakage speed is calculated according to parameters such as drilling fluid density, cutting force and the like. CN103015996a discloses a method for predicting the leakage rate of a steep formation before drilling, which can predict the leakage rate of drilling fluid in a drilling well of a steep and complex formation before drilling, has a certain guiding effect on preventing malignant leakage of the steep formation, and also has a certain help on the selection of the size of plugging materials. The method has the advantages that firstly, the stratum leakage rate is obtained by carrying out mechanical simulation on stratum stress, but due to the complexity of geological composition, accurate results are difficult to obtain by using a simulation mode, the prediction result of the stratum leakage rate can be greatly influenced as long as one initial condition has small deviation, so that the prediction result is wrong, and secondly, the stratum leakage rate is related to the stratum, parameters such as drilling fluid density, drilling weight, drilling rate and the like can influence the stratum leakage rate, so that the accurate stratum leakage rate prediction result cannot be obtained only by researching the geological structure. Disclosure of Invention In order to solve the problems, the invention provides a method, a system, electronic equipment and a storage medium for predicting the leakage speed of drilling based on RBF, which are convenient and accurate and can realize the real-time prediction of the leakage speed of drilling by carrying out data analysis and data mining on data materials such as historical drilling data, real-time drilling data and the like related to a target block. In order to solve the above technical problems, a first aspect of the present invention provides a method for predicting a drilling loss speed based on RBF, the method comprising: collecting historical drilling data and real-time drilling data of a target block, and performing data preprocessing; Obtaining a parameter vector and a hyperplane formula according to the preprocessed historical drilling data, and performing RTF dual conversion according to the parameter vector segmented by the hyperplane formula to obtain a kernel matrix of an RBF support vector machine; and generating a sensor model according to the parameter vector, the hyperplane formula and the nuclear matrix to obtain a drilling leakage speed real-time prediction model based on an RBF support vector machine, and predicting real-time drilling data by using the real-time prediction model to obtain a drilling leakage speed real-time prediction result. According to a preferred embodiment of the present invention, the data preprocessing includes: The data noise reduction is carried out, historical drilling data is judged through a box graph method, irrelevant data, repeated data and smooth noise data are obtained, and the irrelevant data, the repeated data and the smooth noise data are deleted; Data is subjected to data complementation, irrelevant data obtained by deleting in the data noise reduction process is subjected to k nearest neighbor algorithm, repeated data are obtained, and the smooth noise data is complemented; converting data, namely converting stratum lithology and character parameters of drill bit types in data obtained by data noise reduction by a single-heat coding method into digital parameters; Data integration, merging and processing historical drilling data in a multi-file or multi-database operating enviro