CN-122020328-A - Lithology while drilling recognition method, device, equipment and medium

CN122020328ACN 122020328 ACN122020328 ACN 122020328ACN-122020328-A

Abstract

The invention discloses a lithology recognition method while drilling, a device, equipment and a medium, wherein the method comprises the steps of obtaining a triaxial vibration acceleration signal of a target well section in a drilling process, preprocessing and fusing the triaxial vibration acceleration signal to generate a fused vibration signal, carrying out modal decomposition on the fused vibration signal to generate a plurality of eigenvalued components, selecting effective modal components from the eigenvalued components, extracting a target feature vector from the effective modal components, carrying out parameter optimization on a support vector machine model based on an improved artificial bee colony algorithm, training the support vector machine model after parameter optimization based on the target feature vector to obtain a lithology recognition model, and inputting the feature vector of a real-time vibration signal of the well section to be recognized into the lithology recognition model to obtain a lithology recognition result of the well section to be recognized. The lithology identification precision while drilling is remarkably improved, and the engineering requirements of the drilling site on real-time lithology identification and quick decision-making can be fully met.

Inventors

ZHANG CHUANJIU
YANG JUN
CHEN JIE
SHI KAIWEN
ZHOU ZELIN
JIA YULIANG
LI XUANLIANG
LU ZHEN
Jiang Jiangsong
SONG SHUYI

Assignees

国能神东煤炭集团有限责任公司
中国神华能源股份有限公司神东煤炭分公司
重庆大学

Dates

Publication Date: 20260512
Application Date: 20251208

Claims (10)

1. A lithology while drilling recognition method, comprising: Acquiring a triaxial vibration acceleration signal of a target well section in the drilling process; preprocessing and signal fusion are carried out on the triaxial vibration acceleration signals to generate fusion vibration signals; Performing modal decomposition on the fusion vibration signal to generate a plurality of intrinsic modal components, and selecting an effective modal component from the plurality of intrinsic modal components; Extracting a target feature vector from the effective modal component; Performing parameter optimization on the support vector machine model based on an improved artificial bee colony algorithm, and training the support vector machine model subjected to parameter optimization based on the target feature vector to obtain a lithology recognition model; and inputting the characteristic vector of the real-time vibration signal of the well section to be identified into the lithology identification model to obtain the lithology identification result of the well section to be identified.
2. The lithology while drilling recognition method according to claim 1, wherein the step of preprocessing and signal fusion of the triaxial vibration acceleration signals to generate fusion vibration signals specifically comprises: Carrying out trend removal processing on each frame of vibration acceleration signal; performing band-pass filtering processing on the triaxial vibration acceleration signal subjected to trend removal based on a preset cut-off frequency interval; and calculating the composite vector amplitude of the three-axis vibration acceleration signals after filtering to obtain the fusion vibration signals.
3. The lithology while drilling recognition method according to claim 2, wherein the step of performing trending processing on each frame of vibration acceleration signal specifically comprises: Performing straight line fitting on each frame of vibration acceleration signal by a least square method to obtain a trend term of each frame of vibration acceleration signal; and subtracting the trend term from the vibration acceleration signal of each frame to obtain a vibration acceleration signal after trend removal.
4. The lithology while drilling recognition method according to claim 1, wherein the step of performing modal decomposition on the fused vibration signal to generate a plurality of eigenmode components and selecting an effective modal component from the plurality of eigenmode components specifically comprises: Decomposing the fusion vibration signal by an improved self-adaptive noise complete set empirical mode decomposition method to obtain a plurality of intrinsic mode components and a residual component; calculating the cross-correlation coefficient of each eigenmode component and the fusion vibration signal; Calculating an energy duty ratio of each eigenmode component based on the energy of each eigenmode component and the total energy of the fused vibration signal; and screening out the intrinsic mode components with the cross-correlation coefficient larger than a first preset threshold and the energy duty ratio larger than a second preset threshold as the effective mode components.
5. The lithology while drilling recognition method according to claim 1, wherein the step of extracting the target feature vector from the effective modal component specifically further comprises: Performing Hilbert transform on each effective modal component to obtain an envelope signal; performing Fourier transform on the envelope signal to obtain an envelope spectrum; Extracting amplitude characteristics related to drill bit movement frequency in an envelope spectrum based on the drill plate rotating speed and the drill bit tooth number of the target well section; calculating the arrangement entropy characteristic of each effective modal component; Combining the amplitude feature and the permutation entropy feature as a high-dimensional feature vector; performing dimension reduction processing on the high-dimensional feature vector by a principal component analysis method to obtain a low-dimensional feature vector; and selecting the first d principal components with the accumulated contribution rate exceeding a third preset threshold value from the low-dimensional feature vectors as target feature vectors.
6. The lithology while drilling recognition method of claim 5, wherein the step of extracting the amplitude feature related to the bit movement frequency in the envelope spectrum based on the bit rotational speed and bit tooth number of the target well section specifically comprises: Calculating revolution frequency, tooth engagement frequency and harmonic frequency of the drill bit based on the rotation speed of the drill plate and the tooth number of the drill bit; in an envelope spectrum, determining a target frequency band corresponding to the revolution frequency of the drill bit, the tooth engagement frequency and the harmonic frequency; And extracting the amplitude value in the target frequency band to be used as an amplitude value characteristic.
7. The lithology while drilling recognition method according to claim 1, wherein the step of optimizing parameters of the support vector machine model based on the improved artificial bee colony algorithm, and training the support vector machine model after optimizing the parameters based on the target feature vector to obtain the lithology recognition model specifically comprises the steps of: performing global optimization on penalty parameters and kernel function parameters of the support vector machine model through an improved artificial bee colony algorithm; And marking the lithology category label on the target feature vector, and inputting the marked target feature vector into a support vector machine model with optimized parameters for training to obtain the lithology recognition model.
8. A lithology while drilling recognition device, comprising: the acquisition module is used for acquiring a triaxial vibration acceleration signal of the target well section in the drilling process; The first generation module is used for preprocessing and fusing the triaxial vibration acceleration signals to generate fused vibration signals; The second generation module is used for carrying out modal decomposition on the fusion vibration signals to generate a plurality of intrinsic modal components, and selecting effective modal components from the plurality of intrinsic modal components; the feature extraction module is used for extracting a target feature vector from the effective modal components; The model training module is used for carrying out parameter optimization on the support vector machine model based on the improved artificial bee colony algorithm, training the support vector machine model after parameter optimization based on the target feature vector, and obtaining a lithology recognition model; and the third generation module is used for inputting the characteristic vector of the real-time vibration signal of the well section to be identified into the lithology identification model in the drilling process to obtain the lithology identification result of the well section to be identified.
9. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the lithology while drilling recognition method according to any one of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the lithology while drilling recognition method according to any one of claims 1 to 7.

Description

Lithology while drilling recognition method, device, equipment and medium Technical Field The invention relates to the technical field of oil and gas well engineering, in particular to a lithology while drilling identification method, device, equipment and medium. Background Lithology identification is an important piece of geological basis work in the drilling process. Lithology recognition has important significance for the works of building stratum grids, analyzing deposition and spreading, optimizing drilling engineering parameters, comprehensively evaluating reservoirs and the like. Currently, lithology While Drilling (LWD) and Measurement While Drilling (MWD) technologies are mainly adopted at present, and the downhole lithology is inferred in an indirect manner by acquiring geophysical parameters such as gamma, resistivity, density, neutron porosity and the like of the stratum. However, LWD/MWD equipment itself is expensive and its subsequent use and maintenance costs are also significantly higher. And the underground measurement data is usually transmitted back to the ground by a mud pulse telemetry technology, and unavoidable time delay is introduced in the process, so that the decision feedback timeliness is poor. In addition, because the measured parameters belong to the category of geophysical response, the measured parameters and lithology do not have direct correspondence, so that the interpretation result often has multiple solutions, and the reliability and the accuracy of the lithology recognition result are influenced by the subjective judgment of an experience model and an interpreter to a great extent. Disclosure of Invention In view of the above, the invention provides a lithology while drilling identification method, a lithology while drilling identification device, an electronic device and a medium, so as to solve the technical problems of high cost, delayed decision and insufficient reliability of identification results of the traditional logging while drilling and measurement while drilling technology. In a first aspect, there is provided a lithology while drilling recognition method, the method comprising: Acquiring a triaxial vibration acceleration signal of a target well section in the drilling process; Preprocessing and signal fusion are carried out on the triaxial vibration acceleration signals to generate fusion vibration signals; carrying out modal decomposition on the fusion vibration signal to generate a plurality of intrinsic modal components, and selecting an effective modal component from the plurality of intrinsic modal components; Extracting a target feature vector from the effective modal components; Performing parameter optimization on the support vector machine model based on an improved artificial bee colony algorithm, and training the support vector machine model subjected to parameter optimization based on a target feature vector to obtain a lithology recognition model; and inputting the feature vector of the real-time vibration signal of the well section to be identified into a lithology identification model to obtain lithology identification results of the well section to be identified. In a second aspect, there is provided a lithology while drilling recognition apparatus, the apparatus comprising: the acquisition module is used for acquiring a triaxial vibration acceleration signal of the target well section in the drilling process; The first generation module is used for preprocessing and fusing the triaxial vibration acceleration signals to generate fused vibration signals; the second generation module is used for carrying out modal decomposition on the fusion vibration signals to generate a plurality of intrinsic modal components, and selecting effective modal components from the plurality of intrinsic modal components; the feature extraction module is used for extracting a target feature vector from the effective modal components; the model training module is used for carrying out parameter optimization on the support vector machine model based on the improved artificial bee colony algorithm, training the support vector machine model after parameter optimization based on the target feature vector, and obtaining a lithology recognition model; and the third generation module is used for inputting the characteristic vector of the real-time vibration signal of the well section to be identified into the lithology identification model in the drilling process to obtain the lithology identification result of the well section to be identified. In a third aspect, an electronic device is provided comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the lithology while drilling recognition method described above when the computer program is executed by the processor. In a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium st