Intelligent operation system of drilling machine and deviation correction control method

Abstract

The invention relates to the field of mechanical automation, in particular to a drilling machine intelligent operation system and a deviation correction control method.According to the system, through multi-dimensional signal fusion and dynamic risk assessment, drill jamming recognition and disposal response time is remarkably shortened, and efficiency loss caused by traditional shutdown protection is avoided; the imaging and obstacle positioning technology is adopted to replace manual drill lifting inspection, and the no-load operation time and mechanical loss of equipment are reduced. The multi-physical field signal fusion technology breaks through the sensing limitation of a single sensor, and is beneficial to accurately identifying underground obstacle space distribution and geological interface mutation features; acoustic spectrum and thermal field distribution analysis can enhance the ability to analyze drill bit-stratum interaction dynamics, and multi-dimensional input is provided for intelligent decision making. And dynamic optimization of a control strategy is realized in cooperation with a virtual-real interaction mechanism to adapt to demand changes of different stratum conditions and construction stages.

Description

Technical Field

[0001] This invention relates to the field of mechanical automation, and in particular to a drilling rig intelligent operation system and a correction control method. Background Technology

[0002] As a core piece of equipment in foundation engineering construction, auger drilling equipment is widely used in building pile foundation construction, geological exploration, and other fields. Traditional drilling operations rely heavily on the operator's experience and judgment, and generally face several technical bottlenecks when encountering complex strata (such as hard rock layers and highly cohesive soil layers). Existing equipment mostly uses a single torque threshold alarm mechanism, which can only trigger shutdown protection after a stuck drill occurs, lacking the ability to predict progressive stuck drill in advance. This results in a delayed response to stuck drill, requiring repeated drill lifting and inspection for manual handling, leading to decreased construction efficiency and increased wear and tear on drill bits. Conventional laser guidance systems are limited by the accuracy of straightness measurement and are easily affected by drill rod deflection and borehole wall collapse during deep hole operations, making it difficult to achieve millimeter-level trajectory correction. Correction operations often rely on trial and error to adjust drilling parameters, which can easily lead to secondary deviations.

[0003] Current monitoring systems typically collect single-dimensional signals such as vibration and pressure independently, lacking spatiotemporal correlation analysis of multi-physics field data. This makes it difficult to accurately identify complex working conditions such as abrupt changes in soil interfaces and underground obstacles, resulting in a one-sided perception of the working conditions. Existing intelligent drilling rig control systems mostly adopt fixed control strategies and have not established a dynamic interaction mechanism between physical equipment and virtual models. This leads to a lack of adaptability to the geological formation in the correction strategies, making it impossible to achieve continuous optimization of the construction process. Summary of the Invention

[0004] The purpose of this invention is to address the problem that the single-dimensional signals used in the prior art are difficult to identify complex working conditions, resulting in poor effectiveness of stuck drill fault handling and correction strategies. This invention provides a smart drilling rig operation system and correction control method.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a smart drilling rig operation system, comprising: The acquisition module is used to acquire mechanical vibration signals, torque signals, spatial displacement signals, thermal field signals, acoustic signals, and image signals during the drilling process; The analysis module processes the signals acquired by the acquisition module to obtain the temporal evolution law of torque-displacement, displacement stagnation characteristics, vibration spectrum, thermal field distribution map, and acoustic time-frequency map. It then fuses these with a dynamic filtering algorithm and a machine learning model to generate a borehole condition feature vector. A stuck drill risk classification mechanism is established based on a historical working condition model library, and a stuck drill risk level is generated. Based on the abrupt change law of the vibration spectrum, the attenuation law of the thermal field distribution map, and the acoustic time-frequency map, the formation type and interface location are generated. A three-dimensional model of the borehole interior is constructed based on the image signals. The spatial coordinates of hard obstacles in the borehole path are located based on the energy reflection characteristics of the acoustic signals and the local temperature anomalies of the thermal field signals. A time-series prediction model of the borehole trajectory is constructed based on the nonlinear relationship between the torque growth rate and the displacement stagnation duration. Finally, borehole trajectory deviation characteristics are identified. The virtual simulation module establishes a coupled dynamic model of the drill pipe and the geological environment, optimizes the handling strategy through simulation and feeds it back to the decision-making module; it also simulates the equipment response under extreme working conditions, generates safety protection strategies and feeds them back to the decision-making module. The decision-making module determines whether to correct the deviation based on the risk level of the stuck drill; it generates a handling strategy and forms an adjustment instruction.

[0006] By employing multi-dimensional signal fusion and dynamic risk assessment, the response time for stuck drill bit identification and handling is significantly shortened, avoiding efficiency losses caused by traditional downtime protection. Imaging and obstacle location technologies replace manual drill bit lifting for inspection, reducing equipment idle time and mechanical wear. Multi-physics field signal fusion technology overcomes the limitations of single-sensor perception, facilitating accurate identification of the spatial distribution of underground obstacles and abrupt geological interface changes. Acoustic spectrum and thermal field distribution analysis enhance the analytical capability of drill bit-formation interaction dynamics, providing multi-dimensional input for intelligent decision-making. Furthermore, a virtual-real interaction mechanism enables dynamic optimization of control strategies, adapting to changing needs under different geological conditions and construction stages.

[0007] Furthermore, the data acquisition module includes: Vibration detection sensors collect mechanical vibration signals, including axial vibration, radial vibration, and circumferential vibration signals of the drill pipe. A laser displacement sensor and an inertial measurement unit are used to acquire spatial displacement signals of the drill bit; The thermal field detection unit is used to collect data on the temperature field distribution inside the borehole wall; The acoustic signature acquisition unit is used to acquire acoustic signals during the drilling process; Imaging components are used to acquire image signals.

[0008] Furthermore, by performing wavelet packet decomposition on the axial, radial, and circumferential vibration signals of the drill pipe to obtain the vibration spectrum, the energy ratio of the low-frequency band (0-200Hz) and the high-frequency band (greater than 500Hz) is extracted to calculate the spectral kurtosis and obtain the vibration spectrum abrupt change law. The temperature field inside the borehole wall is divided into regions, the radial temperature gradient is calculated, and a thermal field distribution map is obtained. By extracting the formation resonance peak from the acoustic time-frequency diagram using the Mel spectrum cepstral coefficients, the acoustic energy attenuation rate is calculated, and the variation law of the attenuation rate is obtained.

[0009] Furthermore, a feature weight allocation model based on an attention mechanism is constructed to perform cross-dimensional signal fusion of vibration spectrum energy, thermal field distribution, and acoustic energy attenuation rate, generating a multi-dimensional working condition feature vector.

[0010] Furthermore, the dynamic weighting range for vibration spectrum energy is 0.4-0.6, the dynamic weighting range for thermal field distribution is 0.3-0.5, and the dynamic weighting range for acoustic energy attenuation rate is 0.2-0.4. Furthermore, based on the intensity of vibration spectrum mutations, the area of ​​abnormal thermal field regions, and the acoustic signal energy attenuation rate, combined with a historical operating condition model database, a three-level stuck drill risk classification mechanism (low, medium, and high) is established.

[0011] Furthermore, a convolutional neural network is used to perform joint feature learning on the vibration spectrum, thermal field distribution map, and acoustic time-frequency map to obtain the stratigraphic type and its interface location.

[0012] Secondly, the present invention also provides a method for correcting deviations during drilling operations, employing a drilling rig intelligent operation system as described above, comprising: Real-time acquisition of mechanical vibration signals, torque signals, spatial displacement signals, thermal field signals, and acoustic signals during drilling operations; The above signals are subjected to spatiotemporal fusion processing to extract the operating condition feature vector; Determine the risk level of a stuck drill bit; When the preset risk level is reached, image signals are acquired to construct a three-dimensional model of the borehole interior and determine the offset of the borehole trajectory. The optimal solution is obtained by simulating different handling strategies in a virtual environment, and adjustment instructions are generated.

[0013] Furthermore, spatiotemporal fusion processing includes signal denoising and feature enhancement operations; virtual environment simulation includes dynamic response prediction of geological conditions.

[0014] Furthermore, the execution effect data of the adjustment instructions is sent back to the virtual simulation module to form learning samples, and the decision weight parameters of the machine learning model are updated regularly.

[0015] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: Breaking through the limitations of traditional single torque threshold alarms, this system enables dynamic identification and prevention of progressive stuck drill risks, reducing construction interruptions and equipment damage caused by manual intervention. It improves the accuracy of dynamic borehole trajectory correction and enhances the ability to track straight lines in complex geological formations. By integrating multi-physical field signals such as mechanical vibration, thermal field distribution, and acoustic spectrum, it facilitates accurate identification of abrupt soil interface changes, underground obstacles, and geological anomalies. Furthermore, it constructs a dynamic interaction mechanism between physical equipment and virtual models, forming an intelligent decision-making optimization system that links the virtual and real worlds. This enables adaptive optimization of control strategies and continuous improvement of the construction process, successfully building a closed-loop collaborative control system of "multi-source perception - intelligent analysis - autonomous correction and prevention - virtual simulation". Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for corrective control in drilling operations. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0018] Unless otherwise specified, the use of terms such as "upper," "lower," "left," "right," "center," "inner," and "outer" to indicate orientation or positional relationships in the description of specific embodiments of the present invention is based on the orientation or positional relationships shown in the accompanying drawings, or the orientation or positional relationship in which the product / equipment / device is typically placed during use. These terms are merely for the purpose of facilitating the description of the present invention or simplifying the description in specific embodiments, enabling those skilled in the art to quickly understand the solution, and do not indicate or imply that a particular device / component / element must have a specific orientation, or be constructed and operated in a specific positional relationship. Therefore, they should not be construed as limitations on the present invention.

[0019] Furthermore, the use of terms such as "horizontal," "vertical," "suspended," and "parallel" does not imply that the corresponding device / component / element must be absolutely horizontal, vertical, suspended, or parallel, but rather that it can be slightly tilted or have a deviation. For example, "horizontal" merely means that its direction is more horizontal relative to "vertical," not that the structure must be completely horizontal, but that it can be slightly tilted. Alternatively, it can be simplified to mean that the corresponding device / component / element, when set in a "horizontal," "vertical," "suspended," or "parallel" direction, can have an error / deviation of ±10% relative to the corresponding direction, more preferably within ±8%, more preferably within ±6%, more preferably within ±5%, and more preferably within ±4%. As long as the corresponding device / component / element is within the error / deviation range, it can still achieve its function in the present invention.

[0020] Furthermore, the use of terms such as "first," "second," and "third" in terminology is merely for distinguishing descriptions of identical or similar components and should not be interpreted as emphasizing or implying the relative importance of a particular component.

[0021] Furthermore, in the description of the embodiments of the present invention, "several", "more than", and "a number of" represent at least two. The number can be any number, such as two, three, four, five, six, seven, eight, or nine, and can even exceed nine.

[0022] Furthermore, in the description of the technical solution of this invention, unless otherwise explicitly specified / limited / restricted, the terms "set up," "install," "connect," "link," "provided with," "laid out," and "arranged" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to common connection methods in the art, such as welding, riveting, bolting, and threaded connections. Such connections can be mechanical, electrical, or communication connections; they can be direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components.

[0023] Example 1 The intelligent drilling rig operation system of the present invention includes: The acquisition module is used to acquire mechanical vibration signals, torque signals, spatial displacement signals, thermal field signals, acoustic signals, and image signals during the drilling process; The analysis module processes the signals acquired by the acquisition module to obtain the temporal evolution law of torque-displacement, displacement stagnation characteristics, vibration spectrum, thermal field distribution map, and acoustic time-frequency map. It then fuses these with a dynamic filtering algorithm and a machine learning model to generate a borehole condition feature vector. A stuck drill risk classification mechanism is established based on a historical working condition model library, and a stuck drill risk level is generated. Based on the abrupt change law of the vibration spectrum, the attenuation law of the thermal field distribution map, and the acoustic time-frequency map, the formation type and interface location are generated. A three-dimensional model of the borehole interior is constructed based on the image signals. The spatial coordinates of hard obstacles in the borehole path are located based on the energy reflection characteristics of the acoustic signals and the local temperature anomalies of the thermal field signals. A time-series prediction model of the borehole trajectory is constructed based on the nonlinear relationship between the torque growth rate and the displacement stagnation duration. Finally, borehole trajectory deviation characteristics are identified. The virtual simulation module establishes a coupled dynamic model of the drill pipe and the geological environment, optimizes the handling strategy through simulation and feeds it back to the decision-making module; it also simulates the equipment response under extreme working conditions, generates safety protection strategies and feeds them back to the decision-making module. The decision-making module determines whether to correct the deviation based on the risk level of the stuck drill; it generates a handling strategy and forms an adjustment instruction.

[0024] Specifically, the data acquisition module includes: Vibration detection sensors collect mechanical vibration signals, including axial vibration, radial vibration, and circumferential vibration signals of the drill pipe. A laser displacement sensor and an inertial measurement unit are used to acquire spatial displacement signals of the drill bit; The thermal field detection unit is used to collect data on the temperature field distribution inside the borehole wall; The acoustic signature acquisition unit is used to acquire acoustic signals during the drilling process; Imaging components are used to acquire image signals.

[0025] Specifically, triaxial accelerometers are placed at key nodes of the drill pipe to collect axial, radial, and circumferential vibration signals in real time, capturing the dynamic interaction characteristics between the drill bit and the formation; a combination of laser displacement sensors and inertial measurement units is used to monitor the three-dimensional spatial displacement changes of the drill bit, combined with Kalman filtering algorithms to eliminate measurement errors caused by drill pipe deflection; an infrared thermal imager array is integrated behind the drill bit to acquire real-time data on the temperature field distribution of the borehole wall, and abnormal friction areas are identified by changes in thermal radiation intensity; an acoustic sensor group is deployed along the drill pipe to collect acoustic signals during the drilling process, and the acoustic spectrum characteristics of different formations are extracted by combining wavelet transform; the imaging component can use a retractable optical imaging assembly, which is easy to extend into the borehole and can also adapt to the low-light environment inside the borehole.

[0026] The vibration spectrum is obtained by wavelet packet decomposition of the axial, radial and circumferential vibration signals of the drill pipe. The energy ratio of the low frequency band of 0-200Hz and the high frequency band of greater than 500Hz is extracted to calculate the spectral kurtosis and obtain the vibration spectrum abrupt change law.

[0027] The temperature field inside the borehole wall is divided into regions, the radial temperature gradient is calculated, and a thermal field distribution map is obtained. By extracting the formation resonance peak from the acoustic time-frequency diagram using the Mel spectrum cepstral coefficients, the acoustic energy attenuation rate is calculated, and the variation law of the attenuation rate is obtained.

[0028] The processing of the acquired signals also includes: for example, using an adaptive band-stop filter to suppress interference in specific frequency bands for mechanical noise in vibration signals; performing environmental noise compensation on acoustic signals to enhance the significance of formation characteristic frequencies; and eliminating signal delays caused by differences in sampling frequencies of different sensors through a timestamp synchronization mechanism to ensure the spatiotemporal consistency of vibration, displacement, thermal field and acoustic signals.

[0029] Multi-sensor timestamp synchronization achieves signal alignment through Lagrange interpolation: ,in: Indicates the k-th sensor at The sampled value at time 10:00. Indicates the target time. This represents the time of the i-th known data point. This represents the time of the j-th known data point; n represents the interpolation order, which can be selected according to the actual situation, such as 3-5.

[0030] Based on a feature weighting allocation mechanism, cross-dimensional correlated features such as vibration spectrum characteristics, thermal radiation distribution patterns, and acoustic energy attenuation laws are extracted to generate high-information-density operational condition characterization vectors. For example, vibration spectrum characteristics are extracted by performing wavelet packet decomposition on the triaxial acceleration signal to extract the energy ratio between the 0-200Hz low-frequency band and the >500Hz high-frequency band, and the spectral kurtosis index is calculated to identify impact vibration anomalies. Thermal radiation distribution patterns are determined by segmenting the high-temperature region of the borehole wall using data from an infrared thermal imager and calculating the radial temperature gradient to determine heat diffusion anomalies. Acoustic energy attenuation features are extracted by using Mel-spectrum cepstral coefficients to extract formation resonance peaks and calculating the acoustic energy attenuation rate.

[0031] Feature fusion vector obtained by dynamic weight allocation:

[0032] in: Dynamic weights representing the vibration spectrum characteristics; Represents the kurtosis of the vibration spectrum; Represents the dynamic weights of the thermal field distribution; represents the maximum radial temperature gradient; represents the dynamic weight of the acoustic attenuation rate; represents the acoustic attenuation rate; The dynamic weight distribution table is shown in Table 1: Table 1 Dynamic Weight Distribution Table

[0033] Use the long short-term memory network to analyze the temporal evolution law of torque and displacement data, and identify the composite abnormal pattern of sudden torque increase accompanied by displacement stagnation; establish a three-level sticking risk classification mechanism for low, medium and high levels according to the vibration spectrum mutation intensity, the area of the thermal field anomaly region and the acoustic signal energy attenuation rate combined with the historical working condition pattern library. The sticking risk assessment function is expressed as ; Where: represents the vibration spectrum kurtosis; represents the maximum kurtosis; represents the area of the thermal anomaly region; represents the maximum area of the thermal anomaly; represents the acoustic attenuation rate; represents the maximum attenuation rate; The risk level division can be set as: R < 0.4, judged as low risk; 0.4 < R < 0.7, judged as medium risk; R > 0.7, judged as high risk.

[0034] As shown in Table 2: Table 2 Sticking Risk Classification Assessment and Parameter Boundaries

[0035] Monitor the non-linear relationship between the torque growth rate and the displacement stagnation duration through the sliding window mechanism, and early warn of the progressive sticking risk. By constructing a time series prediction model, combined with the current bit weight on bit, rotary speed and formation type, predict the resistance change curve in the future drilling depth interval; generate multiple sets of potential risk evolution paths through Monte Carlo simulation, and evaluate the sticking probability and trajectory deviation risk under different control strategies.

[0036] The trajectory dynamic deviation correction calculates the azimuth deviation and deflection distance between the actual trajectory of the drill bit and the target axis according to the displacement sensor data and the hole wall deformation characteristics output by the three-dimensional modeling module; constructs a drill pipe - formation coupling dynamics model, and takes minimizing the trajectory deviation as the objective function to roll and optimize the multi-degree-of-freedom adjustment instructions of the hydraulic actuator (including bit weight on bit distribution, azimuth fine-tuning, rotary speed adaptation); The virtual simulation module enables digital twin modeling and optimization: It establishes a flexible drill pipe model based on the finite element method, and generates a layered formation mechanical parameter database by combining geological exploration data; it synchronizes sensor data from the physical system with the virtual model's state in real time, simulating drill bit-formation interaction forces, drill pipe bending stress distribution, and borehole wall stability evolution. Multiple correction schemes (such as combinations of different drill pressure gradients and correction angles) are executed in parallel within the virtual environment, evaluating the trajectory convergence speed and energy consumption under each scheme; the actual execution effect of the optimal strategy is fed back to the virtual model, reversing the estimation values ​​of formation mechanical parameters and improving model prediction accuracy. Historical construction data (including formation characteristics, correction strategies, and execution effects) is stored, and the decision weights of control strategies are optimized through reinforcement learning algorithms. The virtual simulation module can also establish safety protection and anomaly response mechanisms, rehearsing equipment responses under extreme conditions in the virtual environment, generating safety protection strategies in advance, and generating emergency protection commands (such as progressive decompression and drill string retraction, and grouting to reinforce the borehole wall) by calling similar historical cases to address emergencies such as borehole wall collapse and abnormal drill string vibration, reducing the risk of equipment damage under abnormal conditions.

[0037] Based on the characteristics of borehole trajectory deviation, multi-degree-of-freedom hydraulic actuators can be used to adjust the drill rod. Model predictive control algorithms can dynamically adjust drilling pressure, rotational speed, and azimuth angle to achieve sub-degree-level accuracy in dynamic trajectory correction. For large-angle deviation conditions, adaptive control strategies can be used to dynamically adjust hydraulic system parameters, suppressing overshoot oscillations and the risk of secondary deviation during the correction process.

[0038] The stratigraphic type (such as clay, rock, and gravel) and interface location can be obtained by using a convolutional neural network to jointly learn the features of the vibration spectrum, thermal field distribution map, and acoustic time-frequency map. Based on the energy reflection characteristics of the acoustic signal and the local temperature anomaly in the thermal field, the spatial coordinates of hard obstacles (such as boulders and concrete blocks) in the borehole path can be located.

[0039] This embodiment employs a drilling operation correction control method based on a drilling rig intelligent operation system, referring to... Figure 1 As shown, it includes the following steps: Real-time acquisition of mechanical vibration signals, torque signals, spatial displacement signals, thermal field signals, and acoustic signals during drilling operations; The above signals are subjected to spatiotemporal fusion processing to extract the operating condition feature vector; Determine the risk level of a stuck drill bit; When the preset risk level is reached, image signals are acquired to construct a three-dimensional model of the borehole interior and determine the offset of the borehole trajectory. The optimal solution is obtained by simulating different handling strategies in a virtual environment, and adjustment instructions are generated.

[0040] When the preset risk level is reached, refer to Table 2 for processing. For example, when a high-risk stuck drill warning is triggered, the drive mechanism pushes the telescopic high-definition camera to the stuck drill position and simultaneously turns on the active supplementary lighting source to adapt to the low-light environment inside the borehole. Obstacle point cloud data is obtained based on binocular vision and structured light scanning technology. The obstacle type (such as rock blocks and metal foreign objects) is distinguished by semantic segmentation algorithm, and its geometric size and embedding depth are calculated. Preset processing schemes are matched for obstacle types (such as adjusting the drill bit speed for impact crushing and starting lateral vibration to assist in getting out of trouble), and the feasibility of the strategy is verified by the virtual simulation module.

[0041] For low-risk and medium-risk cases, refer to Table 2 for handling.

[0042] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A drilling rig intelligent operation system, characterized in that, include: The acquisition module is used to acquire mechanical vibration signals, torque signals, spatial displacement signals, thermal field signals, acoustic signals, and image signals during the drilling process; The analysis module processes the signals acquired by the acquisition module to obtain the temporal evolution law of torque-displacement, displacement stagnation characteristics, vibration spectrum, thermal field distribution map, and acoustic time-frequency map. It then fuses these with a dynamic filtering algorithm and a machine learning model to generate a borehole condition feature vector. A stuck drill risk classification mechanism is established based on a historical working condition model library, and a stuck drill risk level is generated. Based on the abrupt change law of the vibration spectrum, the attenuation law of the thermal field distribution map, and the acoustic time-frequency map, the formation type and interface location are generated. A three-dimensional model of the borehole interior is constructed based on the image signals. The spatial coordinates of hard obstacles in the borehole path are located based on the energy reflection characteristics of the acoustic signals and the local temperature anomalies of the thermal field signals. A time-series prediction model of the borehole trajectory is constructed based on the nonlinear relationship between the torque growth rate and the displacement stagnation duration. Finally, borehole trajectory deviation characteristics are identified. The virtual simulation module establishes a coupled dynamic model of the drill pipe and the geological environment, optimizes the handling strategy through simulation and feeds it back to the decision-making module; it also simulates the equipment response under extreme working conditions, generates safety protection strategies and feeds them back to the decision-making module. The decision-making module determines whether to correct the deviation based on the risk level of the stuck drill; it generates a handling strategy and forms an adjustment instruction.

2. The intelligent drilling rig operation system according to claim 1, characterized in that, The data acquisition module includes: Vibration detection sensors collect mechanical vibration signals, including axial vibration, radial vibration, and circumferential vibration signals of the drill pipe. A laser displacement sensor and an inertial measurement unit are used to acquire spatial displacement signals of the drill bit; The thermal field detection unit is used to collect data on the temperature field distribution inside the borehole wall; The acoustic signature acquisition unit is used to acquire acoustic signals during the drilling process; Imaging components are used to acquire image signals.

3. The intelligent drilling rig operation system according to claim 2, characterized in that, The vibration spectrum is obtained by wavelet packet decomposition of the axial, radial and circumferential vibration signals of the drill pipe. The energy ratio of the low frequency band of 0-200Hz and the high frequency band of greater than 500Hz is extracted to calculate the spectral kurtosis and obtain the vibration spectrum change law. The temperature field inside the borehole wall is divided into regions, the radial temperature gradient is calculated, and a thermal field distribution map is obtained. By extracting the formation resonance peak from the acoustic time-frequency diagram using the Mel spectrum cepstral coefficients, the acoustic energy attenuation rate is calculated, and the attenuation rate variation law is obtained.

4. The intelligent drilling rig operation system according to claim 3, characterized in that, A feature weight allocation model based on an attention mechanism is constructed to fuse vibration spectrum energy, thermal field distribution, and acoustic energy attenuation rate across dimensions to generate a multi-dimensional working condition feature vector.

5. The intelligent drilling rig operation system according to claim 4, characterized in that, The dynamic weighting range for vibration spectrum energy is 0.4-0.6, the dynamic weighting range for thermal field distribution is 0.3-0.5, and the dynamic weighting range for acoustic energy attenuation rate is 0.2-0.

4.

6. The intelligent drilling rig operation system according to claim 3, characterized in that, Based on the intensity of vibration spectrum mutation, the area of ​​abnormal thermal field region, and the acoustic signal energy attenuation rate, combined with a historical working condition pattern database, a three-level stuck drill risk classification mechanism is established, consisting of low, medium, and high risk levels.

7. A drilling rig intelligent operation system according to any one of claims 1-6, characterized in that, A convolutional neural network was used to perform joint feature learning on the vibration spectrum, thermal field distribution map, and acoustic time-frequency map to obtain the stratigraphic type and its interface location.

8. A method for correcting deviations during drilling operations, characterized in that, The drilling rig intelligent operation system as described in any one of claims 1-7 includes: Real-time acquisition of mechanical vibration signals, torque signals, spatial displacement signals, thermal field signals, and acoustic signals during drilling operations; The above signals are subjected to spatiotemporal fusion processing to extract the operating condition feature vector; Determine the risk level of a stuck drill bit; When the preset risk level is reached, image signals are acquired to construct a three-dimensional model of the borehole interior and determine the offset of the borehole trajectory. The optimal solution is obtained by simulating different handling strategies in a virtual environment, and adjustment instructions are generated.

9. The drilling operation correction control method according to claim 8, characterized in that, Spatiotemporal fusion processing includes signal denoising and feature enhancement operations; virtual environment simulation includes dynamic response prediction of geological conditions.

10. The drilling operation correction control method according to claim 8, characterized in that, The execution effect data of the adjustment instructions is sent back to the virtual simulation module to form learning samples, and the decision weight parameters of the machine learning model are updated regularly.

Images