CN-122021228-A - Drilling speed prediction method and device, storage medium and electronic equipment

CN122021228ACN 122021228 ACN122021228 ACN 122021228ACN-122021228-A

Abstract

The disclosure belongs to the technical field of drilling engineering, and provides a drilling rate prediction method, a device, a storage medium and electronic equipment, wherein the method comprises the steps of obtaining first drilling data of a current stratum; based on the first drilling data, carrying out iterative training on a drilling prediction model to be trained by adopting a random forest algorithm until the current condition accords with a training iteration stopping condition, stopping the iterative training to obtain a drilling rate prediction model for predicting the drilling rate, acquiring second drilling data of the current stratum, inputting the second drilling data into the drilling rate prediction model for prediction, and outputting and obtaining a drilling rate prediction result. According to the drilling speed prediction method, corresponding drilling speed prediction models are trained for different strata, the drilling speed prediction model corresponding to each stratum is iteratively trained by adopting the first drilling data corresponding to each stratum, and therefore the accuracy of drilling speed prediction can be greatly improved.

Inventors

WU PENGCHENG
WANG XUEQIANG
WANG XUDONG
WANG YAO
CHEN YE
LI ZHENGTAO

Assignees

中国石油天然气股份有限公司

Dates

Publication Date: 20260512
Application Date: 20241111

Claims (13)

1. A method of predicting a rate of penetration, the method comprising: acquiring first drilling data of a current stratum, wherein the first drilling data are obtained when the current drilling depth of the current stratum is smaller than a preset depth threshold value; based on the first drilling data, performing iterative training on a drilling prediction model to be trained by adopting a random forest algorithm until the current condition accords with a training iteration stopping condition, stopping the iterative training, and obtaining a drilling rate prediction model for predicting the drilling rate; Acquiring second drilling data of the current stratum, wherein the second drilling data are obtained when the current drilling depth of the current stratum is greater than the preset depth threshold value; And inputting the second drilling data into the drilling speed prediction model to predict, and outputting and obtaining a drilling speed prediction result.
2. The method of predicting a drilling rate according to claim 1, wherein, The training iteration stop condition comprises that the current drilling depth reaches the preset depth threshold value.
3. The method for predicting the drilling rate according to claim 1, wherein the performing iterative training on the drilling prediction model to be trained by adopting a random forest algorithm based on the first drilling data until the current condition meets the training iteration stop condition, stopping the iterative training, and obtaining the drilling rate prediction model for predicting the drilling rate comprises: performing data cleaning treatment on the first drilling data in a normalization treatment mode to obtain cleaned first drilling data; Carrying out characteristic engineering processing on the cleaned first drilling data to obtain important characteristic data; And carrying out iterative training on the well drilling prediction model to be trained by adopting the random forest algorithm based on the important characteristic data until the current condition accords with the training iteration stop condition, and stopping iterative training to obtain the drilling rate prediction model for predicting the drilling rate.
4. The method of claim 3, wherein the performing feature engineering processing on the cleaned first drilling data to obtain important feature data comprises: Performing correlation analysis processing on the cleaned first drilling data to obtain a correlation analysis result; performing first dimension reduction processing on the first drilling data according to the correlation analysis result to obtain first dimension reduction data; performing feature importance analysis on the first dimension reduction data to obtain an importance analysis result; and performing second dimension reduction processing on the first dimension reduction data according to the importance analysis result to obtain second dimension reduction data, wherein the second dimension reduction data is the important characteristic data.
5. The method of predicting a drilling rate according to claim 4, The first drilling data includes a plurality of first drilling parameters, and the correlation analysis result includes a correlation coefficient between any two of the plurality of first drilling parameters.
6. The method of claim 5, wherein said performing a first dimension reduction process on said first drilling data based on said correlation analysis results comprises: Randomly selecting a correlation coefficient between any two first drilling parameters in the correlation analysis result as a current correlation coefficient; Acquiring a preset correlation coefficient threshold; And under the condition that the current correlation coefficient is larger than the preset correlation coefficient threshold value, removing one of the two first drilling parameters corresponding to the current correlation coefficient.
7. The method of predicting a drilling rate according to claim 4, The first drilling data comprises a plurality of second drilling parameters, the importance analysis result comprises importance values corresponding to each second drilling parameter, and each importance value is used for assigning importance degree corresponding to each drilling parameter.
8. The method of claim 7, wherein performing a second dimension reduction process on the first dimension reduction data according to the importance analysis result comprises: randomly selecting an importance value of any one second drilling parameter in the importance analysis result as a current importance value; Acquiring a preset importance threshold; and eliminating the second drilling parameter corresponding to the current importance value under the condition that the current importance value is lower than the preset importance threshold value.
9. The method of claim 1, wherein prior to the acquiring the first drilling data for the current formation, the method further comprises: Obtaining regional geological features; dividing the region to be drilled into at least two strata according to the geological features of the region; Determining a current formation from at least two of the formations according to the current drilling depth.
10. The method of predicting a drilling rate according to claim 1, wherein, The first drilling data includes engineering log data, logging while drilling data, mud data, and drilling tool data.
11. A drilling rate prediction apparatus, the apparatus comprising: the first acquisition module is used for acquiring first drilling data of the current stratum, wherein the first drilling data are obtained when the current drilling depth of the current stratum is smaller than a preset depth threshold value; The training module is used for carrying out iterative training on the drilling prediction model to be trained by adopting a random forest algorithm based on the first drilling data until the current condition accords with the training iteration stop condition, stopping the iterative training, and obtaining a drilling rate prediction model for predicting the drilling rate; The second acquisition module is used for acquiring second drilling data of the current stratum, wherein the second drilling data are obtained when the current drilling depth of the current stratum is greater than the preset depth threshold value; and the prediction module is used for inputting the second drilling data into the drilling speed prediction model to predict, and outputting and obtaining a drilling speed prediction result.
12. A computer readable storage medium, characterized in that a computer program is stored thereon, which, when executed in a computer, causes the computer to perform the method of any of claims 1 to 10.
13. An electronic device comprising a memory and a processor, the memory having executable code stored therein, the processor, when executing the executable code, implementing the method of any one of claims 1 to 10.

Description

Drilling speed prediction method and device, storage medium and electronic equipment Technical Field The disclosure belongs to the technical field of drilling engineering, and particularly relates to a drilling rate prediction method, a drilling rate prediction device, a storage medium and electronic equipment. Background One of the key indicators of drilling efficiency is the rate of penetration, which is directly related to the overall progress and cost effectiveness of the drilling operation. Therefore, this field has been the focus of drilling engineering research, attracting attention of many scholars and engineers. In traditional prediction models, the complexity of geological factors is often excessively simplified, and the prediction is mainly performed by relying on standard theory and algorithm. However, this simplified process ignores the significant impact of geological conditions on the rate of penetration, such as formation hardness, rock type, porosity, etc., resulting in insufficient prediction accuracy. CN106909759B, a method and a device for predicting the mechanical drilling speed of a PDC drill bit of a shale stratum, and the method and the device for predicting the mechanical drilling speed of the PDC drill bit of the shale stratum are provided, and the method comprises the steps of measuring the mechanical drilling speed, the acoustic time difference, the uniaxial compressive strength and the triaxial compressive strength of the PDC drill bit of a shale sample in a preset drilling direction; the method comprises the steps of determining dynamic elastic modulus and dynamic poisson ratio according to measured acoustic wave time difference, determining cohesive force and internal friction angle in a preset drilling direction according to triaxial compressive strength, establishing a cohesive force-internal friction angle-acoustic wave time difference model according to acoustic wave time difference, cohesive force and internal friction angle data, and establishing a PDC drill bit mechanical drilling speed prediction model according to PDC drill bit mechanical drilling speeds, drilling weights, uniaxial compressive strength, dynamic elastic modulus, dynamic poisson ratio, cohesive force, internal friction angle and the established cohesive force-internal friction angle-acoustic wave time difference model to predict PDC drill bit mechanical drilling speeds of shale stratum in different drilling directions. CN111434886B is used for calculating the mechanical drilling speed in the drilling process, and provides a mechanical drilling speed calculating method for the drilling process, which comprises the steps of determining lithology information of a stratum corresponding to a well section to be measured of a target well, calculating the mechanical drilling speed of the well section to be measured according to a constructed mechanical drilling speed prediction model to obtain a predicted drilling speed value, carrying out rationality judgment on the predicted drilling speed value according to block drilling speed information of a block to which the target well belongs and adjacent well drilling speed information of an adjacent well of the target well, adjusting influence parameters influencing the mechanical drilling speed in the mechanical drilling speed prediction model based on a judgment result of rationality judgment, and obtaining the mechanical drilling speed value of the well section to be measured according to the adjusted mechanical drilling speed prediction model. The mechanical drilling speed is always a hot spot of drilling engineering research as a drilling efficiency index, and many students aim at establishing a proper model to predict the mechanical drilling speed. In the related art, geological factors are not considered in the process of well drilling prediction, but conventional theory and algorithm are directly adopted for prediction, so that the accuracy of a prediction result is low, and therefore, a more accurate mechanical drilling rate prediction method is needed at present. How to improve the accuracy of the mechanical drilling speed prediction method is a technical problem to be solved. Disclosure of Invention Based on this, it is necessary to provide a drilling rate prediction method, apparatus, storage medium and electronic device, aiming at the defect that the accuracy of the existing mechanical drilling rate prediction method is low. In a first aspect, an embodiment of the present invention provides a method for predicting a drilling rate, where the method includes: acquiring first drilling data of a current stratum, wherein the first drilling data are obtained when the current drilling depth of the current stratum is smaller than a preset depth threshold value; based on the first drilling data, performing iterative training on a drilling prediction model to be trained by adopting a random forest algorithm until the current condition accords with a training iter