CN-121996936-A - Intelligent data processing method for petroleum drilling

CN121996936ACN 121996936 ACN121996936 ACN 121996936ACN-121996936-A

Abstract

The invention provides an intelligent data processing method for petroleum drilling, which comprises the steps of collecting underground data in real time, transmitting the collected data to a ground data processing system for storage and analysis, analyzing the real-time data transmitted to the ground data processing system, identifying a time period passing through a stratum containing high-concentration hydrogen sulfide gas, marking the data period as a risk section, analyzing the underground data marked as the risk section, predicting whether abnormal pressure fluctuation is caused by hydrogen sulfide gas permeation, analyzing all signals generated in the risk section when abnormal pressure fluctuation is caused by hydrogen sulfide gas permeation is predicted, identifying fluctuation signals caused by hydrogen sulfide gas permeation, classifying and marking, generating corresponding risk early warning and operation advice according to the signal classification and marking results in the risk section, continuously collecting and updating the underground data, and dynamically adjusting the risk marking and the operation advice based on the real-time analysis of the change of the data.

Inventors

DENG JUN
WEN YISEN
LI ZIHAN
CHANG ZHIPENG
LUO YUFAN
LIN YANGSHENG

Assignees

中国石油天然气集团有限公司
中国石油集团川庆钻探工程有限公司

Dates

Publication Date: 20260508
Application Date: 20241104

Claims (10)

1. The intelligent data processing method for the petroleum drilling is characterized by comprising the following steps of: In the drilling process, underground data are collected in real time by deploying an underground sensor, and the collected data are transmitted to a ground data processing system in real time for storage and analysis; Analyzing the real-time data transmitted to the ground data processing system, identifying a time period passing through the stratum containing the high-concentration hydrogen sulfide gas, and marking the data period in the time period as a risk section; Analyzing the underground data marked as a risk section, and predicting whether abnormal pressure fluctuation is caused by hydrogen sulfide gas permeation; Under the condition that abnormal pressure fluctuation caused by hydrogen sulfide gas permeation is predicted, analyzing all signals generated in the risk section, identifying fluctuation signals caused by hydrogen sulfide gas permeation, and classifying and marking the signals; generating corresponding risk early warning and operation suggestions according to the signal classification and marking results in the risk section; Downhole data is continuously collected and updated, and risk markers and operation suggestions are dynamically adjusted based on real-time analysis of changes in the data.
2. The intelligent data processing method for petroleum drilling according to claim 1, wherein the analyzing of the real-time data transmitted to the surface data processing system identifies the time period passing through the stratum containing high concentration hydrogen sulfide gas and marks the data period as a risk section, and the method specifically comprises the steps of: Preprocessing underground data transmitted to a ground data processing system in real time; extracting hydrogen sulfide gas concentration data in the preprocessed underground data, and recording the actual concentration of the hydrogen sulfide gas at each moment and a corresponding time period; determining a preset hydrogen sulfide gas concentration threshold value; Comparing the actual concentration of the hydrogen sulfide gas at all times with a preset hydrogen sulfide gas concentration threshold value, and marking a time period corresponding to the actual concentration of the hydrogen sulfide gas which is larger than the preset hydrogen sulfide gas concentration threshold value as a risk section.
3. The intelligent data processing method for petroleum drilling according to claim 1, wherein said analyzing the downhole data marked as a risk section predicts whether the permeation of hydrogen sulfide gas will cause abnormal pressure fluctuations, specifically comprising the steps of: extracting pressure fluctuation information and hydrogen sulfide gas diffusion information in downhole data marked as risk sections; Analyzing the extracted pressure fluctuation information and the hydrogen sulfide gas diffusion information to respectively generate a permeation influence coefficient and a dissolution diffusion index; And constructing an anomaly prediction model by using the generated permeation influence coefficient and the dissolution diffusion index, generating an anomaly coefficient, comparing the generated anomaly coefficient with a preset anomaly coefficient threshold value, and predicting whether the permeation of the hydrogen sulfide gas can cause anomaly pressure fluctuation according to the comparison result.
4. The intelligent data processing method for petroleum drilling according to claim 3, wherein the obtaining logic of the osmotic influence coefficient and the dissolution and diffusion index is as follows: Extracting pressure fluctuation information in underground data marked as a risk section, wherein the information comprises actual pressures of the bottom of a well at different times in a period of time, the actual pressures of the bottom of the well at different times in the period of time are calibrated to be P i , i represents the number of the actual pressures of the bottom of the well at different times in the period of time, and i=1, 2, 3, and g are positive integers; Constructing a set G according to time sequences by using actual pressures P i at the bottom of a well at different moments in a period of time, and marking the set G as G= { P 1 ,P 2 ,P 3 ,P 4 ,…,P g-1 ,P g }; Carrying out frequency domain analysis on actual bottom hole pressure at different moments in a period of time in the set G by using fast Fourier transform, and calculating main frequency of bottom hole pressure fluctuation, wherein the main frequency represents the change frequency of bottom hole pressure in unit time, and the specific formula is as follows: f bd ＝FFT(G)＝FFT({P 1 ,P 2 ,P 3 ,P 4 ,…,P g-1 ,P g }) In the above formula, f bd is the main frequency of the bottom hole pressure fluctuation, and the FFT is a fast Fourier transform algorithm for converting the time domain signal into a frequency domain signal; The fluctuation amplitude of the actual pressure at the bottom of the well at different moments in a period of time is calculated, and a specific calculation formula is as follows: In the above formula, A bd is the fluctuation amplitude of the actual pressure at the bottom of the well at different moments in time; the osmotic influence coefficient is calculated, and a specific calculation formula is as follows: PIC=f bd *A bd in the above formula, PIC is the osmotic influence coefficient; Extracting hydrogen sulfide gas diffusion information in underground data marked as a risk section, wherein the information comprises hydrogen sulfide gas concentrations at different depth positions and depth values corresponding to the depth positions, the hydrogen sulfide gas concentrations at the different depth positions and the depth values corresponding to the depth positions are respectively calibrated to be C j and X j , j represents the hydrogen sulfide gas concentrations at the different depth positions and the numbers of the depth values corresponding to the depth positions, and j=1, 2, 3, and the values of the hydrogen sulfide gas concentrations at the different depth positions and the depth values corresponding to the depth positions are positive integers; calculating the concentration gradient of the hydrogen sulfide gas at different depth positions, wherein the concentration gradient represents the change rate of the hydrogen sulfide gas along with the depth, and the specific formula is as follows: In the above formula, ΔCT is the concentration gradient of the hydrogen sulfide gas at different depth positions, X j represents the depth value at the j-th depth position, and C j represents the concentration of the hydrogen sulfide gas at the j-th depth position; the diffusion coefficient of the hydrogen sulfide gas and the viscosity of the slurry are obtained, the diffusion coefficient of the hydrogen sulfide gas and the viscosity of the slurry are respectively calibrated to be D and theta, and the diffusion speed of the hydrogen sulfide gas in the slurry is calculated according to the specific calculation formula: In the above formula, V ks is the diffusion speed of hydrogen sulfide gas in slurry; The dissolution and diffusion index is calculated by the following specific calculation formula: in the above formula, DDI is the dissolution diffusion index.
5. The intelligent data processing method for petroleum drilling according to claim 4, wherein an anomaly prediction model is constructed from the generated permeation influence coefficient PIC and the dissolution diffusion index DDI, an anomaly coefficient YC is generated by weighted summation, the generated anomaly coefficient YC is compared with a preset anomaly coefficient threshold YC yuzhi , and whether abnormal pressure fluctuation is caused by permeation of hydrogen sulfide gas is predicted according to the comparison result, and the specific analysis is as follows: if YC is less than or equal to YC yuzhi , predicting that abnormal pressure fluctuation cannot be caused by hydrogen sulfide gas permeation; If YC > YC yuzhi , abnormal pressure fluctuation is predicted to be caused by hydrogen sulfide gas permeation.
6. The intelligent data processing method for petroleum drilling according to claim 1, wherein in case that abnormal pressure fluctuation is predicted to be caused by permeation of hydrogen sulfide gas, all signals generated in the risk section are analyzed, fluctuation signals caused by permeation of hydrogen sulfide gas are identified, and the signals are classified and marked, specifically comprising the steps of: Extracting all signal data marked as risk sections under the condition that abnormal pressure fluctuation is predicted to be caused by hydrogen sulfide gas permeation, wherein the signal data comprise time sequence characteristic information of each signal; preprocessing the extracted time sequence characteristic information of each signal; analyzing the time sequence characteristic information of each preprocessed signal to respectively generate a signal consistency coefficient and a fluctuation amplitude index of each signal; And constructing a comprehensive analysis model by using the signal consistency coefficient and the fluctuation amplitude index of each generated signal, generating a fluctuation identification coefficient of each signal, comparing the generated fluctuation identification coefficient of each signal with a preset fluctuation identification coefficient threshold value, identifying fluctuation signals caused by hydrogen sulfide gas permeation according to the comparison result, and classifying and marking the signals.
7. The intelligent data processing method for petroleum drilling according to claim 6, wherein the logic for obtaining the signal consistency coefficient and the fluctuation amplitude index of each signal is as follows: Extracting time sequence characteristic information of each preprocessed signal, wherein the time sequence characteristic information comprises amplitude values of each signal at different moments in a period of time and corresponding time points, the time sequence is represented by a function S m (t), t is the time point, a time period is defined as [ t 1 ,t 2 ],S m (t) represents the amplitude value of an mth signal at the moment t in a period of time, and m=1, 2, 3, & gt, k and k are positive integers; Calculating the amplitude change rate of each signal according to the formula: in the above formula, V m (t) is the amplitude change rate of the mth signal; The average value of the amplitude change rate of each signal in a period of time is calculated, and a specific calculation formula is as follows: In the above-mentioned method, the step of, Is the average value of the amplitude change rate of the mth signal in a period of time; The standard deviation of the amplitude change rate of each signal in a period of time is calculated, and a specific calculation formula is as follows: In the above-mentioned method, the step of, Is the standard deviation of the amplitude change rate of the mth signal in a period of time; the signal consistency coefficient of each signal is calculated, and a specific calculation formula is as follows: In the above formula, CCON m is the signal consistency coefficient of the mth signal; The square sum of the amplitude values of each signal in the time period [ t 1 ,t 2 ] is calculated, and a specific calculation formula is as follows: In the above formula, EM m is the sum of squares of the magnitudes of the mth signal over the time period [ t 1 ,t 2 ]; the square sum of the amplitude of each signal in the whole time domain is calculated, and a specific calculation formula is as follows: In the above formula, ET m is the sum of squares of the magnitudes of the mth signal in the whole time domain; The fluctuation amplitude index of each signal is calculated, and a specific calculation formula is as follows: In the above formula, AAM m is the fluctuation amplitude index of the mth signal.
8. The intelligent data processing method for petroleum drilling according to claim 7, wherein a comprehensive analysis model is constructed by the generated signal consistency coefficient CCON m and fluctuation amplitude index AAM m of each signal, the fluctuation identification coefficient FIC m of each signal is generated by weighted summation, the generated fluctuation identification coefficient FIC m of each signal is compared with a preset fluctuation identification coefficient threshold value FIC yuzhi , the fluctuation signals caused by hydrogen sulfide gas permeation are identified according to the comparison result, and the signals are classified and marked, and the specific analysis is as follows: if the FIC m <FIC yuzhi is a normal fluctuation signal, dividing the signal into normal signals and marking; if the FIC m ≥FIC yuzhi is a fluctuation signal due to permeation of hydrogen sulfide gas, the signal is classified as an abnormal signal and labeled.
9. The intelligent data processing method for petroleum drilling according to claim 1, wherein the corresponding risk early warning and operation advice is generated according to the signal classification and marking result in the risk section, and the method specifically comprises the following steps: Generating risk-free prompt information for the signals marked as normal signals according to the results of the signals classified as normal signals, and keeping the current drilling operation parameters unchanged; and generating high-risk prompt information for the signal marked as the abnormal signal according to the result of classifying the signal as the abnormal signal, and generating operation advice according to the characteristic of the abnormal signal.
10. The intelligent data processing method for the petroleum drilling is characterized by comprising the following steps of: acquiring drilling real-time data, wherein the drilling real-time data at least comprises hydrogen sulfide gas real-time concentration data and a time period corresponding to the hydrogen sulfide gas real-time concentration data; confirming a first time period exceeding a preset first hydrogen sulfide gas concentration according to the drilling real-time data, and taking the drilling real-time data acquired in the first time period as first data; confirming a first abnormal coefficient according to the first data, wherein the first abnormal coefficient is used for representing the fluctuation degree of the first data; Confirming a first fluctuation identification coefficient of a first signal acquired in the first time period when the first abnormality coefficient exceeds a preset first threshold value, wherein the first fluctuation coefficient is used for representing the fluctuation degree of the first signal, and And when the first fluctuation identification coefficient exceeds a preset second threshold value, the first signal is used as an abnormal pressure fluctuation signal caused by hydrogen sulfide gas permeation.

Description

Intelligent data processing method for petroleum drilling Technical Field The invention relates to the field of petroleum drilling, in particular to an intelligent data processing method for petroleum drilling. Background Petroleum drilling is the process of drilling a drill bit deep into the ground by drilling equipment to recover oil and gas resources. During drilling, a large amount of complex real-time data is involved to process, including downhole pressure, temperature, rate of penetration, bit torque, mud flow, and borehole trajectory. The real-time monitoring and processing of the data is of great importance, because the drilling operation environment is complex and changeable, the underground situation is changeable instantaneously, and any abnormal parameters can cause serious accidents such as lost circulation, well collapse or blowout. Thus, this data must be processed and analyzed in real time to ensure safe, stable and efficient operation of the drilling process. Existing petroleum drilling data processing techniques rely primarily on the collaborative work of downhole sensors, data Acquisition Systems (DAS) and data analysis software. First, downhole sensors (e.g., pressure, temperature, rate of drilling, torque sensors, etc.) collect various drilling parameters in real time and transmit data to the surface via a cable or wireless transmission system. The ground data acquisition system receives and primarily processes the raw data, and performs filtering, denoising and calibration operations to ensure the basic accuracy of the data. And then, the data analysis software inputs the primarily processed data into a preset mathematical model or algorithm (such as a drilling hydrodynamic model, a downhole pressure calculation model and the like), and predicts and analyzes risks possibly occurring in the drilling process by combining historical data and real-time monitoring results. Specific processes typically include steps of data cleaning, normalization, parameter fitting, feature extraction, etc., to generate well parameter optimization suggestions and alarm cues. And finally, the analysis result is presented to an operator through a monitoring system, and the operator adjusts the operating parameters such as the drilling pressure, the pumping pressure, the drilling speed and the like according to the system suggestion, so that the safe and efficient operation of the drilling process is ensured. In the deep well drilling process, especially when crossing the stratum containing high concentration hydrogen sulfide gas, the hydrogen sulfide gas is easy to permeate into the slurry and gradually reacts with the slurry in a dissolution way, so that the underground pressure data have tiny and irregular fluctuation. These fluctuations are often manifested as intermittent low amplitude pressure changes, particularly at low flow rates, where the fluctuations caused by hydrogen sulfide gas permeation are very close to the normal formation pressure change signal and are indistinguishable. The prior art cannot accurately identify whether these minor fluctuations are due to chemical effects caused by the infiltration of hydrogen sulfide gas into the mud, or normal formation pressure fluctuations. Therefore, the system cannot judge the gas permeation risk in time, and early warning cannot be sent out in the early stage of hydrogen sulfide dissipation, so that operators miss the best opportunity for slurry density adjustment or exhaust treatment, and finally, well control failure can be possibly caused, and the risk of serious accidents such as blowout or welling is increased. Disclosure of Invention The present invention aims to address at least one of the above-mentioned deficiencies of the prior art. For example, it is an object of the present invention to provide a method for intelligent data processing of petroleum drilling caused by hydrogen sulfide gas permeation. In order to achieve the above purpose, the invention provides an intelligent data processing method for petroleum drilling. The intelligent data processing method for petroleum drilling comprises the following steps: in the drilling process, underground data are collected in real time through deployment of underground sensors, and the collected data are transmitted to a ground data processing system in real time for storage and analysis. Real-time data transmitted to a surface data processing system is analyzed, a time period is identified that passes through a formation containing high concentrations of hydrogen sulfide gas, and the data period within the time period is marked as a risk segment. Downhole data, labeled as risk zones, are analyzed to predict whether hydrogen sulfide gas permeation will cause abnormal pressure fluctuations. In the case where abnormal pressure fluctuations are predicted to be caused by hydrogen sulfide gas permeation, all signals generated in the risk section are analyzed, fluctuation signals caused by hydrogen su