An automated drilling system for marine exploration

Abstract

This invention relates to the field of marine drilling technology, and more particularly to an automated drilling system for marine exploration. The system includes a data acquisition module; a data analysis module for determining the drilling parameter correction type based on whether real-time drilling parameters show abnormal changes and whether real-time drilling platform stability parameters meet the drilling platform vibration conditions; a drilling parameter correction module, including a real-time correction unit for real-time correction of drilling parameters based on real-time drilling parameter deviation values ​​or predicting the stability of the drilling platform after a first preset time, and a delayed correction activation unit for delayed correction of drilling parameters or stopping drilling; and a delayed correction module for determining a prediction activation threshold for delayed correction of drilling parameters or correction of drilling platform stability using a trend prediction correction method based on whether the stability of the drilling platform changes regularly. This invention comprehensively analyzes drilling platform stability and drilling parameter correction to improve drilling accuracy.

Description

Technical Field

[0001] This invention relates to the field of marine drilling technology, and more particularly to an automated drilling system for marine exploration. Background Technology

[0002] Traditional offshore drilling technology has revealed many significant limitations when facing the complex and ever-changing marine environment. The complexity of the marine environment is reflected in many aspects. For example, factors such as waves, currents, and tides can cause drilling platforms to be in a state of continuous unstable vibration. This not only seriously affects the positioning accuracy of drilling equipment but also greatly increases the difficulty of parameter control during the drilling process. When the platform vibration amplitude is too large, the drilling equipment may experience displacement deviation, making it impossible for the drill bit to drill accurately along the predetermined trajectory. This, in turn, affects the quality of core sampling and the accuracy of the judgment of underground geological structure. At the same time, the strata properties of different sea areas vary greatly, and traditional drilling equipment may be unable to adjust drilling parameters in a timely manner, leading to excessive wear of the drill bit and low drilling efficiency.

[0003] However, existing technologies suffer from low drilling accuracy and efficiency due to a lack of analysis combining drilling geology and drilling platform stability on drilling parameters during offshore drilling. Summary of the Invention

[0004] Therefore, the present invention provides an automated drilling system for marine exploration, which overcomes the problems of low drilling accuracy and low efficiency caused by the lack of analysis of the impact of drilling geology and drilling platform stability on drilling parameters in the existing marine drilling process.

[0005] To achieve the above objectives, the present invention provides an automated drilling system for marine exploration, comprising:

[0006] The data acquisition module is used to collect real-time cuttings return data, real-time drilling parameter data, and real-time drilling platform stability parameter data.

[0007] The data analysis module, which is connected to the data acquisition module, is used to determine the drilling parameter correction type based on whether the real-time drilling parameters have undergone abnormal changes and whether the real-time drilling platform stability parameters meet the drilling platform vibration conditions.

[0008] A drilling parameter correction module, which is connected to the data analysis module, includes,

[0009] The real-time correction unit is used to determine the stability of the drilling platform after real-time correction of drilling parameters based on the real-time drilling parameter deviation value or to predict the stability of the drilling platform after a first preset time, based on the drilling parameter correction type.

[0010] The delay correction activation unit, which is connected to the real-time correction unit, is used to determine the delay correction drilling parameters or stop drilling based on the predicted stability of the drilling platform after a first preset time.

[0011] The delay correction module, which is connected to the drilling parameter correction module, is used to determine the prediction activation threshold for drilling parameter delay correction or drilling platform stability correction based on whether the stability of the drilling platform changes regularly.

[0012] The adjustment module, which is connected to the drilling parameter correction module and the delay correction module respectively, is used to determine whether to calibrate the detection sensor of the drilling platform stability parameters based on whether at least two delay corrections are performed within a second preset time period after the drilling parameters are delayed.

[0013] Furthermore, the data analysis module determines the drilling parameter correction type based on whether the real-time drilling parameters have undergone abnormal changes and whether the real-time drilling platform stability parameters meet the drilling platform vibration conditions.

[0014] If the real-time drilling parameters change abnormally and the real-time drilling platform stability parameters do not meet the drilling platform vibration conditions, the data analysis module determines the drilling parameter correction type to be the real-time correction type.

[0015] If the real-time drilling parameters change abnormally and the real-time drilling platform stability parameters meet the drilling platform vibration conditions, the data analysis module determines that the drilling parameter correction type is a delayed correction type.

[0016] Furthermore, the data analysis module determines that the real-time drilling parameters have undergone abnormal changes based on the comparison results of at least two real-time drilling parameters whose rate of change is greater than the corresponding preset rate of change. The data analysis module also determines that the real-time drilling platform stability parameters meet the drilling platform vibration conditions based on the fact that the stability of the real-time drilling platform is greater than the first preset stability and the difference between the stability of the real-time drilling platform and the stability of the previous drilling platform is a positive value.

[0017] Furthermore, the data analysis module calculates the stability of the drilling platform based on its vibration frequency and maximum tilt angle.

[0018] Furthermore, the real-time correction unit determines, based on the drilling parameter correction type, the stability of the drilling platform after real-time correction of the drilling parameters or prediction of the drilling parameter deviation value for a first preset time period, wherein...

[0019] If the drilling parameter correction type is a real-time correction type, the real-time correction unit determines to correct the drilling parameters in real time based on the real-time drilling parameter deviation value;

[0020] If the drilling parameter correction type is a time-delay correction type, the real-time correction unit determines the stability of the drilling platform after predicting the first preset time.

[0021] Furthermore, the delay correction activation unit determines the delay correction drilling parameters or stops drilling based on the predicted stability of the drilling platform after a first preset time period, wherein,

[0022] If the predicted stability of the drilling platform is less than or equal to the preset predicted stability after the first preset time, the delay correction activation unit determines the delay correction drilling parameters.

[0023] If the predicted stability of the drilling platform is greater than the preset predicted stability after the first preset time, the delay correction activation unit determines to stop drilling.

[0024] Furthermore, the delay correction module determines the threshold for predicting and initiating drilling parameter delay correction or drilling platform stability correction using a trend prediction correction method based on whether the stability of the drilling platform changes regularly.

[0025] If the stability of the drilling platform changes in a regular manner, the delay correction module determines to use the trend prediction correction method to perform the delay correction of drilling parameters;

[0026] If the stability of the drilling platform changes irregularly, the delay correction module determines the prediction activation threshold for correcting the stability of the drilling platform.

[0027] The prediction activation threshold is a first preset stability level.

[0028] Furthermore, the delay correction module determines that the stability of the drilling platform changes in a regular manner based on the comparison results of the variance of the stability of the drilling platform being less than a preset variance over several consecutive periods.

[0029] Furthermore, the correction amount for the first preset stability level is positively correlated with the variance of the drilling platform's stability level over several consecutive periods.

[0030] Furthermore, the adjustment module determines whether to calibrate the detection sensor for the drilling platform stability parameters based on whether at least two delay corrections are performed within a second preset time period after the drilling parameters have been corrected by delay.

[0031] If at least two delay corrections are performed within the second preset time period after the delay correction of the drilling parameters, the adjustment module determines the detection sensor for calibrating the stability parameters of the drilling platform.

[0032] Compared with existing technologies, the advantages of this invention are as follows: Based on the combined logic of real-time drilling parameter anomalies and platform vibration conditions, this invention clearly distinguishes between real-time correction and delayed correction scenarios, fundamentally avoiding safety hazards caused by improper correction. When parameters are abnormal but the platform does not vibrate, real-time correction is triggered, which can quickly respond to parameter problems such as sudden formation changes and prevent the anomaly from escalating. At the same time, because the platform is stable, the correction operation will not increase the burden on the equipment or structure. When the platform meets the vibration conditions, delayed correction is triggered. Considering the time delay between detecting platform vibration and adjusting drilling parameters, delayed correction can synchronize the actual vibration state of the platform with the adjustment of drilling parameters. Anomalies are only judged when the change rate of at least two real-time drilling parameters is greater than the corresponding preset change rate, avoiding the misjudgment of normal fluctuations of a single parameter as anomalies and reducing invalid corrections. Accurate anomaly judgment and reasonable correction type selection can maximize drilling efficiency while ensuring safety, indirectly reducing engineering costs and avoiding unnecessary downtime caused by misjudgment of anomalies or failure downtime caused by missed anomaly judgment.

[0033] Furthermore, the real-time correction unit of this invention dynamically selects correction operations based on the correction type, achieving precise matching between correction strategies and operational scenarios. For real-time correction types, it directly initiates parameter correction processes, quickly responding to issues such as sudden formation changes and avoiding anomaly accumulation. For delayed correction types, it prioritizes predicting future stability. If the stability is insufficient to support correcting drilling parameters, corrections are only made after the platform stabilizes, avoiding forced parameter adjustments during vibrations that could exacerbate structural risks. If the stability is sufficient to support correcting drilling parameters, delayed corrections can synchronize the actual vibration state of the drilling platform with drilling parameter corrections, improving drilling accuracy. This ensures rapid response in risk-free situations and eliminates blind operations during high-risk situations, ensuring that correction actions always match the operational safety state. By combining feedback control (PID / PI) with feedforward compensation, the lag problem of traditional single feedback control is avoided. With the combination of the two, the original correction command can eliminate current deviations and proactively mitigate future risks, significantly reducing the duration of parameter fluctuations.

[0034] Furthermore, this invention predicts the stability of the drilling platform after a first preset time period, rather than relying on the current state. When the predicted stability exceeds the preset predicted stability, drilling is directly stopped. This means the system anticipates that the platform will exceed the safe stability threshold within a future timeframe. Stopping the machine in advance cuts off the additional load on the platform from the drilling operation before the actual risk occurs. If forced correction is made when the platform is still unstable in the future, the parameter adjustment may cause additional stress on the platform, further amplifying tilting or vibration. By accurately selecting the correction timing through predicted stability, delayed correction of drilling parameters is triggered only when the predicted stability is less than or equal to the preset predicted stability. This not only resolves previously accumulated parameter anomalies through parameter adjustment but also synchronizes the actual vibration state of the drilling platform with the adjustment of drilling parameters, improving drilling accuracy. A balance between safety and efficiency is achieved through accurate prediction and graded decision-making. The machine only stops when the predicted stability exceeds the safe value, avoiding the overly conservative approach of stopping for minor vibrations. Operations are only stopped when there is a real structural risk. Delayed correction is performed when stability is predicted, avoiding the risk of immediate correction and preventing footage loss due to shutdown.

[0035] Furthermore, this invention categorizes the fluctuations in the stability of drilling platforms into two types: regular changes (such as periodic vibrations caused by ocean waves and tides, with fixed periods and amplitudes) and irregular changes (such as irregular vibrations caused by sudden gusts of wind or instantaneous equipment failures, with no fixed pattern). For regular changes, a trend prediction correction method is used. Utilizing the periodicity of the fluctuations, the dominant period is extracted, the trend term and periodic term are separated, and future fluctuation curves are predicted. Corrections are applied in advance, ensuring the correction effect is synchronized with the fluctuation peak, thus solving the lag problem of correcting only after fluctuations occur. For irregular changes, the prediction threshold is adjusted accordingly. Irregular fluctuations usually indicate a sudden change in the current environment or equipment state, and the original threshold... The original threshold may no longer be suitable (e.g., judging based on the original threshold would frequently trigger false corrections). In this case, adjusting the threshold allows the judgment standard to adapt to the new fluctuation characteristics. This enables precise control of predictable regular fluctuations while flexibly responding to unpredictable sudden fluctuations, significantly improving the adaptability of corrections in complex environments. The system can still maintain reliable judgments during sudden disturbances, greatly improving robustness. The trend prediction correction method can offset the impact of fluctuations by making corrections in advance without interrupting operations, avoiding footage loss due to waiting for fluctuations to subside. Dynamic correction thresholds can reduce false corrections and reduce the impact of frequent parameter adjustments on drilling efficiency. The collaboration of the two modes ensures platform stability and reduces ineffective operations, achieving dual optimization of safety and efficiency.

[0036] Furthermore, the adjustment module of this invention triggers sensor calibration based on the correction frequency, promptly detecting correction failures caused by data distortion, and ensuring decision reliability from the source. The stability parameters of the drilling platform depend on sensor acquisition, but sensors in the marine environment are susceptible to salt spray corrosion, vibration interference, and electromagnetic interference, leading to data drift. Data distortion can cause the system to repeatedly correct but fail to solve the problem. Traditional systems lack real-time monitoring of sensor reliability. The adjustment module of this invention uses the correction frequency to infer data reliability, achieving early warning of sensor calibration. If at least two delayed corrections occur within the second preset time after the delayed correction, it indicates that the correction has not achieved the expected effect, and calibration is directly triggered. At this time, the adjustment module determines that the sensor may be inaccurate and starts the calibration process (such as comparison with a backup sensor, zero-point calibration) to eliminate the influence of data distortion. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the structure of an automated drilling system for marine exploration according to the present invention;

[0038] Figure 2 This is a schematic diagram of the drilling parameter correction module in an automated drilling system for marine exploration according to the present invention.

[0039] Figure 3 This is a flowchart illustrating the workflow of the data analysis module in an automated drilling system for marine exploration according to the present invention. Detailed Implementation

[0040] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0041] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0042] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0043] Please see Figures 1-3 As shown, Figure 1 This is a schematic diagram of the structure of an automated drilling system for marine exploration according to the present invention; Figure 2 This is a schematic diagram of the drilling parameter correction module in an automated drilling system for marine exploration according to the present invention. Figure 3 This is a flowchart illustrating the workflow of the data analysis module in an automated drilling system for marine exploration according to the present invention.

[0044] An embodiment of the present invention discloses an automated drilling system for marine exploration, comprising:

[0045] The data acquisition module is used to collect real-time cuttings return data, real-time drilling parameter data, and real-time drilling platform stability parameter data.

[0046] The data analysis module, which is connected to the data acquisition module, is used to determine the drilling parameter correction type based on whether the real-time drilling parameters have undergone abnormal changes and whether the real-time drilling platform stability parameters meet the drilling platform vibration conditions.

[0047] A drilling parameter correction module, which is connected to the data analysis module, includes,

[0048] The real-time correction unit is used to determine the stability of the drilling platform after real-time correction of drilling parameters based on the real-time drilling parameter deviation value or to predict the stability of the drilling platform after a first preset time, based on the drilling parameter correction type.

[0049] The delay correction activation unit, which is connected to the real-time correction unit, is used to determine the delay correction drilling parameters or stop drilling based on the predicted stability of the drilling platform after a first preset time.

[0050] The delay correction module, which is connected to the drilling parameter correction module, is used to determine the prediction activation threshold for drilling parameter delay correction or drilling platform stability correction based on whether the stability of the drilling platform changes regularly.

[0051] The adjustment module, which is connected to the drilling parameter correction module and the delay correction module respectively, is used to determine whether to calibrate the detection sensor of the drilling platform stability parameters based on whether at least two delay corrections are performed within a second preset time period after the drilling parameters are delayed.

[0052] The real-time cuttings output data in this embodiment of the invention includes, but is not limited to, "cuttings lithology data, cuttings particle size data, and cuttings weight data"; the real-time drilling parameter data includes, but is not limited to, "real-time drill pressure data, real-time drilling speed data, and real-time torque data"; and the real-time drilling platform stability parameter data includes, but is not limited to, "real-time drilling platform three-dimensional vibration data, real-time drilling platform tilt angle data, and platform displacement deviation data".

[0053] Specifically, the data analysis module determines the drilling parameter correction type based on whether real-time drilling parameters have undergone abnormal changes and whether real-time drilling platform stability parameters meet the drilling platform vibration conditions.

[0054] If the real-time drilling parameters change abnormally and the real-time drilling platform stability parameters do not meet the drilling platform vibration conditions, the data analysis module determines the drilling parameter correction type to be the real-time correction type.

[0055] If the real-time drilling parameters change abnormally and the real-time drilling platform stability parameters meet the drilling platform vibration conditions, the data analysis module determines that the drilling parameter correction type is a delayed correction type.

[0056] If the real-time drilling parameters do not change abnormally, the data analysis module determines that the drilling parameters do not need to be corrected.

[0057] Specifically, the data analysis module determines that the real-time drilling parameters have undergone abnormal changes based on the comparison results of at least two real-time drilling parameters whose rate of change is greater than the corresponding preset rate of change. The data analysis module determines that the real-time drilling platform stability parameters meet the drilling platform vibration conditions based on the fact that the stability of the real-time drilling platform is greater than the first preset stability and the difference between the stability of the real-time drilling platform and the stability of the previous drilling platform is a positive value.

[0058] Specifically, the data analysis module calculates the stability of the drilling platform based on its vibration frequency and maximum tilt angle.

[0059] In this embodiment of the invention, the corresponding preset rate of change is the average of the real-time drilling parameter change rates when several formation abrupt changes occur under the same drilling conditions (including but not limited to "same drilling tools, same geological parameters, and same drilling tool working mode"); the first preset stability level is the average of the stability levels when the drilling platform is unstable and drilling parameters need to be adjusted under the same drilling conditions (not exceeding the safe stability level of the drilling platform; if it exceeds, drilling is stopped directly). For example, from the drilling log of the project over the past year, 12 historical cases meeting the conditions are selected, the vibration frequency and maximum tilt angle of each case are recorded, and the corresponding stability level is calculated. The average value is taken as the first preset stability level. The stability level is the sum of the ratio of vibration frequency to preset vibration frequency and the ratio of maximum tilt angle to preset tilt angle. The preset vibration frequency is the maximum vibration frequency allowed by the drilling platform (through modal analysis (ModalAna)). The drilling platform's first few natural frequencies (such as overall sway frequency, torsional frequency, and local structural vibration frequency) are obtained through finite element analysis (FEA). The maximum allowable vibration frequency (f_max) must be far away from any of the platform's natural frequencies to prevent resonance amplification effects that could lead to structural damage. It should not be within ±10%-15% of the natural frequency, and f_max should ultimately be less than the lowest natural frequency or greater than the highest natural frequency. The preset tilt angle is the maximum allowable tilt angle of the drilling platform. (A precise model of the drilling platform is established through FEA. The stress in key supporting structures (such as legs, locking devices, and main deck beams) gradually increases as the tilt angle increases under maximum design load. When the platform is tilted to a certain angle, the maximum equivalent stress (such as Von Mises stress) in key parts reaches 2 / 3 of the yield strength of the platform's structural steel (i.e., a safety factor of 1.5). When the angle is the maximum allowable tilt angle based on structural integrity, the maximum tilt angle is the maximum angular deviation value generated in three-dimensional space when the drilling platform (such as the main structure that carries the drilling equipment, such as the base of a land drilling rig or the deck of an offshore drilling platform) tilts relative to the horizontal plane during real-time drilling operations. However, the above value is not limited to this and can be adjusted by those skilled in the art based on the actual situation.

[0060] This invention, based on a combined logic of real-time drilling parameter anomalies and platform vibration conditions, clearly distinguishes between real-time correction and delayed correction scenarios. This fundamentally avoids safety hazards caused by improper correction. When parameters are abnormal but the platform is not vibrating, real-time correction is triggered, allowing for rapid response to parameter issues such as sudden formation changes and preventing the anomaly from escalating. Simultaneously, because the platform is stable, the correction operation will not increase the burden on equipment or structure. When the platform meets vibration conditions, delayed correction is triggered. Considering the time delay between detecting platform vibration and adjusting drilling parameters, delayed correction can synchronize the actual platform vibration state with drilling parameter adjustments. Anomalies are only determined when the change rate of at least two real-time drilling parameters exceeds a corresponding preset change rate, avoiding misjudging normal fluctuations of a single parameter as anomalies and reducing ineffective corrections. Precise anomaly detection and reasonable correction type selection can maximize drilling efficiency while ensuring safety, indirectly reducing engineering costs and avoiding unnecessary downtime due to misjudged anomalies or malfunctions due to missed anomaly detection.

[0061] Specifically, the real-time correction unit determines, based on the drilling parameter correction type, whether to correct the drilling parameters in real time using the real-time drilling parameter deviation value or to predict the stability of the drilling platform after a first preset time period.

[0062] If the drilling parameter correction type is a real-time correction type, the real-time correction unit determines to correct the drilling parameters in real time based on the real-time drilling parameter deviation value;

[0063] If the drilling parameter correction type is a time-delay correction type, the real-time correction unit determines the stability of the drilling platform after predicting the first preset time.

[0064] In this embodiment of the invention, the real-time correction unit determines the real-time correction of drilling parameters based on the real-time drilling parameter deviation value. This includes calculating the real-time drilling parameter deviation value, which is a weighted sum of the deviation values ​​of the abnormally changing real-time drilling parameter change rate and its corresponding preset change rate after standardization. Based on the calculated comprehensive real-time drilling parameter deviation value, a basic correction amount (such as drilling pressure adjustment amount, rotation speed adjustment amount) is generated using a proportional-integral (PI) or proportional-integral-derivative (PID) algorithm. Based on the comprehensive deviation value and the cuttings return data prediction, an original correction command containing feedback amount and feedforward compensation amount is generated. After rate limitation and amplitude limitation on the original correction command, the final safety correction command is output to the drilling actuator to complete the real-time correction. The first preset duration (T_pre) refers to the time span for the system to predict the future platform state. Its value should be greater than the total time (T_response) required for the drilling parameter adjustment command to fully manifest its effect at the downhole drill bit. T_response mainly includes the actuator action time and the downhole drill string response time.

[0065] In this embodiment of the invention, the real-time correction unit extracts parameters that have undergone abnormal changes, namely drill pressure and torque; calculates the deviation of the rate of change of each parameter: drill pressure change rate deviation ΔV_wob = |ΔWOB / Δt| - V_wob_preset = 1.2 - 0.5 = 0.7 kN / s; torque change rate deviation ΔV_torque = |ΔTorque / Δt| - V_torque_preset = 0.17 - 0.1 = 0.07 kN·m / s; and standardizes the deviations (e.g., by dividing by their respective preset change rates): standardized drill pressure deviation ΔV_wob_norm = 0.7 / 0.5 = 1.4; standardized torque deviation ΔV_torque_norm = 0.07 / 0.1 = 0.7. Dynamic weight allocation is used because the cuttings data simultaneously indicate formation hardening (increased sandstone content), and the system automatically assigns a higher weight (e.g., 0.7, which can be determined through a regression model trained on historical data) to the drill pressure deviation, since drill pressure is a response to changes in formation hardness. The most direct parameter; the torque weight is reduced accordingly (e.g., 0.3); the weighted sum is calculated to obtain the comprehensive real-time drilling parameter deviation value (B), B=(0.7*1.4)+(0.3*0.7)=0.98+0.21=1.19; the original correction command is generated, and feedback control (PID algorithm) is used, with the comprehensive deviation value B=1.19 as the input of the PID controller; after algorithm calculation (assuming the integral and derivative terms are calculated within this period), a basic drill pressure increase command is output, for example: ΔWOB_feedback=+25kN; feedforward control (based on cuttings prediction), cuttings data (increased grain size, harder lithology) is a strong precursor to the formation becoming harder; the feedforward control unit predicts, based on historical data models, that without intervention, the torque will continue to soar and the drilling rate will decrease; therefore, a feedforward compensation amount, a stronger drill pressure increase command and a pre-emptive speed fine-tuning command are generated, for example: ΔWOB_feedforward=+15kN, ΔRPM_feedforward = +5RPM (slightly increase the rotational speed to help break harder rocks); synthesize the original correction command, the final drill pressure adjustment ΔWOB_total = ΔWOB_feedback + ΔWOB_feedforward = +40kN; the final rotational speed adjustment ΔRPM_total = ΔRPM_feedforward = +5RPM; (torque is not directly controlled, but indirectly managed by adjusting WOB and RPM); the maximum allowable single adjustment of drill pressure is 50kN, +40kN is within the allowable range, and the rotational speed adjustment +5RPM is also within the safe range; the rate of change of drill pressure must be less than 5kN / s. Assuming the control cycle is 2 seconds, then 40kN / 2s = 20kN / s, which far exceeds the allowable value; to prevent impact on the equipment, this command is split into 4 consecutive sub-command outputs, each cycle (0.The drill bit pressure adjustment is increased by +10kN over 5 seconds, reaching the target value smoothly within 2 seconds; the rotational speed adjustment is output in one step; a smoothed safety correction command (ΔWOB=+10kN / 0.5s, ΔRPM=+5RPM) is sent to the drilling actuators (drill bit pressure servo system and top drive frequency converter system).

[0066] This invention's real-time correction unit dynamically selects correction operations based on correction type, achieving precise matching between correction strategies and operational scenarios. For real-time correction types, it directly initiates parameter correction processes, quickly responding to issues such as sudden formation changes and avoiding anomaly accumulation. For delayed correction types, it prioritizes predicting future stability. If the stability is insufficient to support correcting drilling parameters, correction is only performed after the platform stabilizes, avoiding forced parameter adjustments during vibration that could exacerbate structural risks. If the stability is sufficient to support correcting drilling parameters, delayed correction can synchronize the actual vibration state of the drilling platform with drilling parameter correction, improving drilling accuracy. This ensures rapid response in risk-free situations and eliminates blind operations in high-risk situations, ensuring that correction actions always match the operational safety state. By combining feedback control (PID / PI) with feedforward compensation, it avoids the lag problem of traditional single feedback control. With the combination of the two, the original correction command can both eliminate current deviations and proactively avoid future risks, significantly reducing the duration of parameter fluctuations.

[0067] Specifically, the delay correction activation unit determines the delay correction drilling parameters or stops drilling based on the predicted stability of the drilling platform after a first preset time period.

[0068] If the predicted stability of the drilling platform is less than or equal to the preset predicted stability after the first preset time, the delay correction activation unit determines the delay correction drilling parameters.

[0069] If the predicted stability of the drilling platform is greater than the preset predicted stability after the first preset time, the delay correction activation unit determines to stop drilling.

[0070] In this embodiment of the invention, the preset predicted stability level is the safety and stability level of the drilling platform, and its value is 2. However, the above value is not limited to this, and those skilled in the art can adjust it according to the actual situation.

[0071] In this embodiment of the invention, predicting the stability of the drilling platform after a first preset time period includes: pre-calculating offline using professional hydrodynamic software (such as AQWA, WAMIT) to generate a database of motion response amplitude operators (RAO) and hydrodynamic coefficients covering various draft and load states of the platform; collecting a large amount of historical data (environmental data, platform motion data, vibration data), training LSTM or other machine learning prediction models offline, and after training, lightweighting the model and converting it into a format suitable for high-speed inference (such as TensorRT, ONNX); collecting environmental data and platform state data at the current moment; and the system does not perform complex CFD or full-time domain simulations. Instead, it directly interpolates and queries the RAO under the current working condition from the pre-generated RAO database. It combines the simplified predicted wave spectrum (considered as a linear superposition of multiple regular waves) with the RAO and quickly calculates the rigid body motion trend of the platform at the future time T_pre (such as the motion amplitude at the center of gravity) by linear superposition. The rigid body motion trend predicted by the simplified model, the current platform state, and real-time environmental data are used as inputs and fed into the pre-trained lightweight LSTM model. The LSTM model does not predict from scratch, but outputs a refined correction based on the results of the simplified model. It is mainly used to predict high-frequency vibrations and nonlinear response parts that the mechanistic model is not good at. The prediction results of the simplified mechanistic model are fused with the correction output of the data-driven model to finally obtain a high-precision stability prediction value at the future time T_pre that meets the real-time requirements.

[0072] This invention predicts the stability of the drilling platform after a first preset time period, rather than relying on the current state. When the predicted stability exceeds the preset predicted stability, drilling is directly stopped. This means the system anticipates that the platform will exceed the safe stability threshold within a future timeframe. Stopping the machine in advance cuts off the additional load on the platform from the drilling operation before the actual risk occurs. If forced correction is made when the platform is still unstable in the future, the parameter adjustment may cause additional stress on the platform, further amplifying tilting or vibration. By accurately selecting the correction timing through predicted stability, delayed correction of drilling parameters is triggered only when the predicted stability is less than or equal to the preset predicted stability. This not only resolves previously accumulated parameter anomalies through parameter adjustment but also synchronizes the actual vibration state of the drilling platform with the adjustment of drilling parameters, improving drilling accuracy. A balance between safety and efficiency is achieved through accurate prediction and hierarchical decision-making. The machine only stops when the predicted stability exceeds the safe value, avoiding the overly conservative approach of stopping for minor vibrations. Operations are only stopped when there is a real structural risk. Delayed correction is performed when stability is predicted, avoiding the risk of immediate correction and preventing footage loss due to shutdown.

[0073] Specifically, the delay correction module determines the threshold for predicting and opening the drilling platform stability prediction based on whether the stability of the drilling platform changes regularly. This threshold is achieved using a trend prediction correction method.

[0074] If the stability of the drilling platform changes in a regular manner, the delay correction module determines to use the trend prediction correction method to perform the delay correction of drilling parameters;

[0075] If the stability of the drilling platform changes irregularly, the delay correction module determines the prediction activation threshold for correcting the stability of the drilling platform.

[0076] The prediction activation threshold is a first preset stability level.

[0077] Specifically, the delay correction module determines that the stability of the drilling platform changes in a regular manner based on the comparison results of the variance of the stability of the drilling platform being less than a preset variance over several consecutive periods.

[0078] In this embodiment of the invention, the drilling parameter delay correction method using trend prediction includes extracting the dominant cycle (denoted as T_cycle) of stability fluctuations using autocorrelation analysis or Fast Fourier Transform (FFT); calculating the autocorrelation function for real-time data of M consecutive cycles (M≥5, such as ocean wave cycles typically being 8-15 seconds, collecting 40-75 stability data points over 40-75 seconds), and the time difference corresponding to the peak value of the function is the dominant cycle (e.g., if the peak value appears at 10 seconds, it indicates that the fluctuation cycle is 10 seconds); converting the time-domain stability data into a frequency-domain signal, the reciprocal of the frequency corresponding to the frequency domain peak value is the dominant cycle, which can eliminate high-frequency noise interference (such as instantaneous equipment vibration); using the moving average method or polynomial fitting, separating the long-term trend term and the periodic fluctuation term from the original stability data; for example: the stability data for a certain period is S(t)=0.2t+1.5sin(2πt / T_cycle)+ε (t is time, 0.2t is the time of slow rise with the tide). The trend term is upward, 1.5sin(2πt / T_cycle) is a periodic fluctuation term, and ε is a small noise. The periodic fluctuation is filtered by a 5-point moving average to obtain a smooth trend term S_trend(t). The trend term is then subtracted from the original data to obtain the pure periodic fluctuation term S_cycle(t). For the separated periodic fluctuation term S_cycle(t), key characteristic parameters are calculated and stored as a regular fluctuation feature library: amplitude A is the difference between the maximum and minimum values ​​of the fluctuation term (e.g., A=3, indicating that the stability fluctuates by 3 within the period); phase offset φ is the time difference between the fluctuation start point and the standard sine wave (e.g., φ=π / 4, indicating that the fluctuation peak appears T_cycle / 8 earlier than the standard period); attenuation coefficient k, if the amplitude decreases slowly with time (e.g., the energy of ocean waves weakens), then k is a positive number (e.g., k=0).05 indicates a 5% decrease in amplitude per cycle; combining the quantified fluctuation characteristics, a time series prediction model is used to predict the stability within the first preset duration + parameter response time (ensuring the prediction covers the effective period of the correction command); based on the linear / nonlinear characteristics of the fluctuation pattern, a suitable prediction model is selected. If the fluctuation exhibits a linear trend + fixed period (such as stability changes caused by uniform tides), a seasonal autoregressive integral moving average (SARIMA) model is used, with the dominant period T_cycle as the seasonal parameter. The model is trained by inputting stability data (including trend and periodic terms) from the first 3 complete cycles, and outputting the predicted value S_pred(t) for the next 1-2 cycles; if the fluctuation exhibits a nonlinear trend + periodic change (such as amplitude slowly increasing over time), the prediction model is more suitable. For large wave fluctuations, a Long Short-Term Memory (LSTM) network is used. Historical stability values, real-time wave height, and wind speed are used as input features to train the model to learn a nonlinear mapping relationship and predict the continuous change curve of stability within future periods. To reduce accumulated error, the prediction is updated every 5 seconds (1 / 5 of the sampling frequency). The latest 5 seconds of actual stability data are collected and compared with the predicted values ​​for the same period to calculate the error ΔS_err = |actual value - predicted value|. If ΔS_err > 10% × target stability (the target stability is usually 80% of the first preset stability; for example, if the first preset stability is 5, then the target value is 4), then S_pred(t) for the remaining prediction period is corrected based on the error (if the error is positive, subsequent prediction values ​​are uniformly increased by 0).8×ΔS_err (to avoid error amplification); based on the predicted stability trend, calculate the amount of correction to be applied in advance, and precisely control the timing of the correction to ensure that the correction effect is synchronized with the peak of the fluctuation; set the target stability level S_target, and calculate the deviation between the stability level and the target value in the future prediction period, ΔS(t)=S_pred(t)-S_target; if ΔS(t)>0, it means that the predicted stability level is higher than the target value (platform vibration is aggravated, and the drilling pressure / speed needs to be reduced to reduce vibration feedback); if ΔS(t)<0, it means that the predicted stability level is lower than the target value (the platform is too stable, and the drilling pressure / speed can be appropriately increased to improve drilling efficiency); based on historical drilling data, establish a quantitative mapping relationship between stability deviation and drilling parameter correction (through multiple linear regression or neural network fitting), drilling pressure correction ΔP=k_P×ΔS(t), k_P is the drilling pressure sensitivity coefficient (experimentally measured, such as k_P=-6kN / unit ΔS, which means that for every increase of 1 in ΔS, the drilling pressure needs to be reduced by 6kN). The following parameters are used for correction: Rotational speed correction ΔR = k_R × ΔS(t): k_R is the rotational speed sensitivity coefficient (e.g., k_R = -3r / min / unit ΔS, meaning that for every 1 increase in ΔS, the rotational speed needs to be reduced by 3r / min); Torque correction ΔT: indirectly controlled through adjustments to drilling pressure and rotational speed (torque is positively correlated with drilling pressure and rotational speed, so direct calculation is unnecessary); Considering the physical response time t_response of drilling parameter adjustments (e.g., drilling pressure adjustments take 10-15 seconds to take effect, and rotational speed adjustments take 5-10 seconds to take effect), calculate the lead correction time t_lead, t_lead = 1.2 × t_response (1.2 is a safety factor to ensure early correction). For example, if t_response = 10 seconds, then t_lead = 12 seconds; if the predicted stability reaches its peak ΔS = 2 at 30 seconds, then a correction command is issued at 18 seconds (30-12). The corrected parameters at 30 seconds exactly offset the peak fluctuation, keeping the actual stability near S_target.

[0079] The preset variance mentioned in this embodiment of the invention can be determined by the following method: From historical drilling logs under the same drilling conditions, at least 10 stable data points that have been verified as regular changes (such as periodic vibrations caused by ocean waves, which were later confirmed to have no sudden interference) are selected, and the variance of their stability is calculated to determine the upper limit of the maximum variance of the regular fluctuations; For each case of regular fluctuations, the stability values ​​of M consecutive cycles (M≥5, such as an 8-second ocean wave cycle, collecting 40 stability data points for 40 seconds) are collected at a sampling frequency of 1 time / second; The stability variance of a single case is calculated according to the statistical variance formula, and the variances of all 10 regular cases are statistically analyzed. The maximum value among them is taken as the preset variance, but the above value is not limited to this, and those skilled in the art can adjust it according to the actual situation.

[0080] This invention categorizes the fluctuations in the stability of drilling platforms into two types: regular changes (such as periodic vibrations caused by ocean waves and tides, with fixed periods and amplitudes) and irregular changes (such as irregular vibrations caused by sudden gusts of wind or momentary equipment failures, with no fixed pattern). For regular changes, a trend prediction and correction method is used. Utilizing the periodicity of the fluctuations, the dominant period is extracted, the trend term and periodic term are separated, and future fluctuation curves are predicted. Corrections are applied in advance, ensuring the correction effect is synchronized with the fluctuation peak, thus solving the lag problem of correcting only after fluctuations occur. For irregular changes, a correction prediction threshold is used instead. Irregular fluctuations usually indicate a sudden change in the current environment or equipment state, and the original threshold may change. If the original threshold is no longer suitable (e.g., judging based on the original threshold would frequently trigger false corrections), then adjusting the threshold allows the judgment standard to adapt to the new fluctuation characteristics. This enables precise control of predictable, regular fluctuations while flexibly responding to unpredictable, sudden fluctuations, significantly improving the system's adaptability to complex environments. The system can still maintain reliable judgments during sudden disturbances, greatly improving robustness. The trend prediction correction method can offset the impact of fluctuations by making corrections in advance without interrupting operations, avoiding footage loss due to waiting for fluctuations to subside. Dynamically correcting the threshold can reduce false corrections and reduce the impact of frequent parameter adjustments on drilling efficiency. The collaboration of these two modes ensures platform stability and reduces ineffective operations, achieving a dual optimization of safety and efficiency.

[0081] Specifically, the correction amount of the first preset stability level is positively correlated with the variance of the stability level of the drilling platform over several consecutive periods.

[0082] In this embodiment of the invention, the correction amount for the first preset stability level is the product of a proportionality coefficient and the current first preset stability level. The proportionality coefficient can be determined by calculating the relative variance offset rate (R). The relative variance offset rate is used to quantify the multiple relationship between the severity of the current irregular fluctuations and the baseline level of historical regular fluctuations. R = (σ_current² - σ_preset²) / σ_preset²; σ_current² represents the variance of the drilling platform stability level calculated over several consecutive periods; σ_preset² represents the preset variance (the upper limit of the variance of historical regular fluctuations). To prevent over-correction under extreme abnormal conditions, the relative offset rate R is input into a saturation function (such as the hyperbolic tangent function tanh) for compression processing, mapping it to the range of (-1, 1) to ensure the stability of the system. The proportionality coefficient λ is composed of the saturated relative offset rate R_saturated and a maximum allowable adjustment range (λ_max). The values ​​are determined jointly; λ = R_saturated * λ_max; λ_max is a pre-set constant representing the maximum allowable adjustment ratio in a single correction (e.g., λ_max = 0.2, meaning a maximum adjustment of ±20% of the current threshold in a single correction). Its specific value is determined through simulation and debugging using historical data.

[0083] Specifically, the adjustment module determines whether to calibrate the detection sensor for the drilling platform stability parameters based on whether at least two delay corrections are performed within a second preset time period after the drilling parameters are corrected by delay.

[0084] If at least two delay corrections are performed within the second preset time period after the delay correction of the drilling parameters, the adjustment module determines the detection sensor for calibrating the stability parameters of the drilling platform.

[0085] In this embodiment of the invention, the second preset duration is the average interval between two delay corrections under several identical drilling conditions, but the above value is not limited to this, and those skilled in the art can adjust it according to the actual situation.

[0086] The adjustment module of this invention triggers sensor calibration based on correction frequency, promptly detecting correction failures caused by data distortion, and ensuring decision reliability from the source. The stability parameters of drilling platforms rely on sensor data acquisition. However, in marine environments, sensors are susceptible to salt spray corrosion, vibration interference, and electromagnetic interference, leading to data drift. Data distortion can cause the system to repeatedly correct itself without solving the problem. Traditional systems lack real-time monitoring of sensor reliability. The adjustment module of this invention uses correction frequency to infer data reliability, achieving early warning of sensor calibration. If at least two delayed corrections occur within a second preset time after a delayed correction, it indicates that the correction has not achieved the expected effect, and calibration is directly triggered. At this time, the adjustment module determines that the sensor may be inaccurate and initiates the calibration process (such as comparison with a backup sensor, zero-point calibration) to eliminate the influence of data distortion.

[0087] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An automated drilling system for marine exploration, characterized in that, include: The data acquisition module is used to collect real-time cuttings return data, real-time drilling parameter data, and real-time drilling platform stability parameter data. The data analysis module, which is connected to the data acquisition module, is used to determine the drilling parameter correction type based on whether the real-time drilling parameters have undergone abnormal changes and whether the real-time drilling platform stability parameters meet the drilling platform vibration conditions. A drilling parameter correction module, which is connected to the data analysis module, includes, The real-time correction unit is used to determine the stability of the drilling platform after real-time correction of drilling parameters based on the real-time drilling parameter deviation value or to predict the stability of the drilling platform after a first preset time, based on the drilling parameter correction type. The delay correction activation unit, which is connected to the real-time correction unit, is used to determine the delay correction drilling parameters or stop drilling based on the predicted stability of the drilling platform after a first preset time. The delay correction module, which is connected to the drilling parameter correction module, is used to determine the prediction activation threshold for drilling parameter delay correction or drilling platform stability correction based on whether the stability of the drilling platform changes regularly. The adjustment module, which is connected to the drilling parameter correction module and the delay correction module respectively, is used to determine whether to calibrate the detection sensor of the drilling platform stability parameters based on whether at least two delay corrections are performed within a second preset time period after the drilling parameters are delayed.

2. The automated drilling system for marine exploration according to claim 1, characterized in that, The data analysis module determines the drilling parameter correction type based on whether real-time drilling parameters show abnormal changes and whether real-time drilling platform stability parameters meet the drilling platform vibration conditions. If the real-time drilling parameters change abnormally and the real-time drilling platform stability parameters do not meet the drilling platform vibration conditions, the data analysis module determines the drilling parameter correction type to be the real-time correction type. If the real-time drilling parameters change abnormally and the real-time drilling platform stability parameters meet the drilling platform vibration conditions, the data analysis module determines that the drilling parameter correction type is a delayed correction type.

3. The automated drilling system for marine exploration according to claim 2, characterized in that, The data analysis module determines that the real-time drilling parameters have undergone abnormal changes based on the comparison results of at least two real-time drilling parameters whose rate of change is greater than the corresponding preset rate of change. The data analysis module determines that the real-time drilling platform stability parameters meet the drilling platform vibration conditions based on the fact that the stability of the real-time drilling platform is greater than the first preset stability level and the difference between the stability of the real-time drilling platform and the stability of the previous drilling platform is a positive value.

4. An automated drilling system for marine exploration according to claim 3, characterized in that, The data analysis module calculates the stability of the drilling platform based on its vibration frequency and maximum tilt angle.

5. An automated drilling system for marine exploration according to claim 4, characterized in that, The real-time correction unit determines, based on the drilling parameter correction type, whether to correct the drilling parameters in real time using the real-time drilling parameter deviation value or to predict the stability of the drilling platform after a first preset time period. If the drilling parameter correction type is a real-time correction type, the real-time correction unit determines to correct the drilling parameters in real time based on the real-time drilling parameter deviation value; If the drilling parameter correction type is a time-delay correction type, the real-time correction unit determines the stability of the drilling platform after predicting the first preset time.

6. An automated drilling system for marine exploration according to claim 5, characterized in that, The delay correction activation unit determines the delay correction drilling parameters or stops drilling based on the predicted stability of the drilling platform after a first preset time period. If the predicted stability of the drilling platform is less than or equal to the preset predicted stability after the first preset time, the delay correction activation unit determines the delay correction drilling parameters. If the predicted stability of the drilling platform is greater than the preset predicted stability after the first preset time, the delay correction activation unit determines to stop drilling.

7. An automated drilling system for marine exploration according to claim 6, characterized in that, The delay correction module determines the threshold for predicting and correcting drilling platform stability using a trend prediction correction method based on whether the stability of the drilling platform changes regularly. If the stability of the drilling platform changes in a regular manner, the delay correction module determines to use the trend prediction correction method to perform the delay correction of drilling parameters; If the stability of the drilling platform changes irregularly, the delay correction module determines the prediction activation threshold for correcting the stability of the drilling platform. The prediction activation threshold is a first preset stability level.

8. An automated drilling system for marine exploration according to claim 7, characterized in that, The delay correction module determines that the stability of the drilling platform changes regularly based on the comparison results of the variance of the drilling platform's stability within several consecutive periods being less than a preset variance.

9. An automated drilling system for marine exploration according to claim 8, characterized in that, The correction amount for the first preset stability level is positively correlated with the variance of the drilling platform's stability level over several consecutive periods.

10. An automated drilling system for marine exploration according to claim 9, characterized in that, The adjustment module determines whether to calibrate the detection sensor for the drilling platform stability parameters based on whether at least two delay corrections are performed within a second preset time period after the drilling parameters are corrected by delay. If at least two delay corrections are performed within the second preset time period after the delay correction of the drilling parameters, the adjustment module determines the detection sensor for calibrating the stability parameters of the drilling platform.

Images