An intelligent early warning system and method for drilling well control

Abstract

The present invention relates to the technical field of drilling in soil layers or rocks, and particularly to an intelligent early warning system and method for well control. The system includes a collection unit, a establishment unit, an identification unit, an analysis unit, an evaluation and optimization unit, and an early warning generation unit. It obtains well control-related parameters through a sensor cluster, conducts multi-source data fusion analysis by combining edge computing and a cloud platform, generates an optimized risk assessment result, and outputs a well control early warning instruction. The present invention can improve the accuracy and real-time performance of well control risk monitoring, effectively solve the problems of traditional sensors being easily interfered with and having a slow response speed, and provide intelligent safety guarantee for drilling operations.

Description

Technical Field

[0001] The present invention relates to the technical field of drilling in soil layers or rocks, and in particular to an intelligent early warning system for well control in drilling, and also to an intelligent early warning method. Background Art

[0002] With the continuous development of technology, data fusion analysis has gradually become an important tool in various fields, and the field of well control in drilling is no exception. In the past, well control in drilling mainly relied on manual monitoring and experience judgment, but this method has low efficiency, is easily affected by human factors, and is difficult to meet the requirements of modern high-efficiency and safe drilling. Most of the existing well control early warning methods analyze and judge based on single-sensor data, and have limited ability to predict risks such as blowouts in complex environments. At the same time, due to the complex environment at the drilling site, traditional electrochemical sensors are easily interfered with or poisoned, resulting in a decrease in measurement accuracy and affecting the accuracy of early warning.

[0003] In addition, there are certain limitations in the response speed and explosion-proof design of the existing systems, and they may not be able to fully meet the real-time and safety requirements in high-risk scenarios. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent early warning system for well control in drilling to solve the technical problems in the prior art that the monitoring accuracy of well control risks is low and the real-time performance is poor due to the susceptibility of single-sensor data to interference; and also to provide an intelligent early warning method for well control in drilling to solve the technical problems in the prior art that the monitoring accuracy of well control risks is low and the real-time performance is poor due to the susceptibility of single-sensor data to interference.

[0005] In one aspect of the present invention, an intelligent early warning system for well control in drilling is provided, which includes a processor, and also includes an acquisition unit, a establishment unit, an identification unit, an analysis unit, an evaluation and optimization unit, and an early warning generation unit that are data-connected to the processor; wherein, the acquisition unit is used to connect to the data acquisition layer of the drilling operation platform and call a sensor cluster to obtain well control-related parameters;

[0006] The establishment unit is used to obtain the well depth information and formation pressure information of the drilling operation platform, and establish an initial risk assessment model in combination with the well control-related parameters;

[0007] The identification unit is used to connect to the data acquisition layer and call an edge computing unit, identify abnormal patterns according to the well control-related parameters, and incorporate the patterns whose identification results meet the preset abnormal threshold into the abnormal pattern set;

[0008] The analysis unit is used to connect to the data acquisition layer and call the cloud platform decision-making layer, perform multi-source data fusion analysis after connecting the cloud platform decision-making layer and the edge computing unit, and obtain the predicted risk probability corresponding to each pattern in the abnormal pattern set;

[0009] The evaluation and optimization unit is used to optimize and adjust the initial risk assessment model according to the predicted risk probability, and output the optimized risk assessment result;

[0010] The early warning generation unit is used to generate a well control early warning instruction according to the optimized risk assessment result.

[0011] In some embodiments, the system further includes a benchmark risk model determination unit, which is used to connect to the historical operation database and determine a benchmark risk model according to the historical correlation relationship between the well depth information and the formation pressure information;

[0012] In some embodiments, the system further includes a risk distribution node determination unit, which is used to determine the risk distribution nodes of the drilling operation platform on the benchmark risk model, where the risk distribution nodes are key monitoring points that meet the changing trend of well control related parameters.

[0013] In some embodiments, the system further includes a segmented risk assessment model establishment unit, which is used to segment the benchmark risk model with identified risk nodes and output a segmented risk assessment model.

[0014] In some embodiments, the system further includes a sensor status generation unit, which is used to classify the sensor resources of the drilling operation platform according to the edge computing unit, obtain multiple types of sensors, and generate a sensor status library for the acquisition basic information of each type of sensor.

[0015] In some embodiments, the system further includes an operation status generation unit, which is used to collect basic information from the historical operation data of the drilling operation platform according to the cloud platform decision-making layer and generate an operation status library.

[0016] In some embodiments, the system further includes an optimization unit, which is used to optimize the abnormal mode set and the predicted risk probability based on the sensor status library and the operation status library.

[0017] Another aspect of the present invention provides an intelligent early warning method for drilling well control, which is executed by a processor and includes: connecting to the data acquisition layer of the drilling operation platform, invoking the sensor cluster to obtain well control related parameters, and collecting gas concentration, pressure value, flow value and mud density information in the well control related parameters; obtaining the well depth information and formation pressure information of the drilling operation platform, and combining the well control related parameters to establish an initial risk assessment model; connecting to the data acquisition layer and invoking the edge computing unit to identify abnormal patterns according to the well control related parameters, and incorporating the patterns whose identification results meet the preset abnormal threshold into the abnormal pattern set; connecting to the data acquisition layer and invoking the cloud platform decision layer, performing multi-source data fusion analysis after connecting the cloud platform decision layer and the edge computing unit to obtain the predicted risk probability corresponding to each pattern in the abnormal pattern set; optimizing and adjusting the initial risk assessment model according to the predicted risk probability, and outputting the optimized risk assessment result, wherein the optimized risk assessment result is the predicted value of the blowout probability within the next 30 minutes; generating a well control early warning instruction according to the optimized risk assessment result.

[0018] According to one or more technical solutions adopted by the present invention, the beneficial effects that can be achieved are as follows: The intelligent early warning system for drilling well control of the present invention realizes high-precision monitoring and real-time early warning of well control risks, and effectively solves the problems existing in the prior art that the single sensor data is easily interfered, and the traditional electrochemical sensors are easily poisoned and the response speed is insufficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings according to these drawings without creative efforts.

[0020] Figure 1 It is the structural diagram of the intelligent early warning system of the present invention;

[0021] Figure 2 It is the flow chart of the intelligent early warning method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The following will combine the attached drawings in the embodiments of the present invention Figure 1 - attached Figure 2 to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments.

[0023] Embodiment 1:

[0024] Figure 1The structural diagram of an intelligent early warning system for drilling well control provided by an embodiment of the present invention. The system includes a processor, where the processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application-Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.; the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0025] The system further includes an acquisition unit, a establishment unit, an identification unit, an analysis unit, an evaluation and optimization unit, and an early warning generation unit that are data-connected to the processor. The aforementioned units jointly implement the functions of the system through physical connection and data interaction methods.

[0026] Among them, the acquisition unit is used to connect to the data acquisition layer of the drilling operation platform and call the sensor cluster to obtain well control-related parameters. Further, as the first step of the entire system, the parameter acquisition unit is used to connect to the data acquisition layer of the drilling operation platform and call the sensor cluster to obtain well control-related parameters. The sensor cluster includes a gas concentration sensor, a pressure sensor, a flow sensor, and a mud density sensor, and these sensors are connected to the acquisition unit by wired or wireless means; the well control-related parameters at least include gas concentration, pressure value, flow value, and mud density information. It can be understood that the gas concentration sensor is installed at key gas monitoring points on the drilling platform, for example, near the wellhead and at the outlet of the mud circulation system; the pressure sensors are arranged in different depth areas of the wellbore and at the surface high-pressure pipe manifold; the flow sensors are set at the outlet of the mud pump and the return pipeline; and the mud density sensor is placed in the mud pit and the mud circulation system. The acquisition unit transmits the collected real-time dynamic data to the subsequent units through the signal transmission module to ensure the continuity and accuracy of the data.

[0027] The establishment unit is used to obtain the well depth information and formation pressure information of the drilling operation platform, and establish an initial risk assessment model in combination with well control related parameters. Further, the establishment unit receives the well control related parameters from the acquisition unit, and constructs an initial risk assessment model in combination with the well depth information and formation pressure information obtained from the drilling operation platform. Among them, the well depth information is provided by the depth measurement device of the drilling platform, and the formation pressure information is obtained through comprehensive calculation of the pressure sensor and the historical formation pressure database. The establishment process of the initial risk assessment model includes: preliminarily analyzing the well control related parameters to determine the correlation relationship between the parameters. For example, there may be a positive correlation between the gas concentration and the formation pressure, and the change of the mud density may affect the pressure balance in the wellbore. The establishment unit performs mathematical modeling on the above parameters through an algorithm module to generate a basic risk assessment framework.

[0028] The identification unit is used to connect to the data acquisition layer and call the edge computing unit to identify abnormal patterns according to the well control related parameters, and incorporate the patterns whose identification results meet the preset abnormal threshold into the abnormal pattern set. Further, the identification unit connects to the data acquisition layer and calls the edge computing unit to identify abnormal patterns of the well control related parameters by using edge computing technology. The edge computing unit is usually deployed near the data acquisition point to reduce data transmission delay and improve response speed. The identification unit judges whether there are abnormalities in the well control related parameters through a preset abnormal threshold. For example, when the gas concentration exceeds the preset safety range or the pressure value fluctuates violently, this pattern is incorporated into the abnormal pattern set. Each pattern in the abnormal pattern set corresponds to a specific parameter combination, and these combinations are updated in real time through the fast processing ability of the edge computing unit.

[0029] The analysis unit is used to connect to the data acquisition layer and call the cloud platform decision layer. After connecting the cloud platform decision layer and the edge computing unit, it performs multi-source data fusion analysis to obtain the predicted risk probability corresponding to each pattern in the abnormal pattern set. Further, the analysis unit connects to the cloud platform decision layer and conducts data interaction with the edge computing unit in the identification unit to perform multi-source data fusion analysis. The cloud platform decision layer is connected to the analysis unit through a network communication module to receive the abnormal pattern set data from the edge computing unit. The analysis unit calculates the predicted risk probability corresponding to each pattern in the abnormal pattern set through comprehensive analysis of historical operation data, real-time monitoring data, and external environment data. For example, a certain abnormal pattern may be associated with high-risk events in historical data, and its predicted risk probability is therefore given a higher weight. The analysis unit also uses machine learning algorithms to classify and regress the data to further improve the prediction accuracy.

[0030] The evaluation and optimization unit is used to optimize and adjust the initial risk assessment model according to the predicted risk probability, and output the optimized risk assessment result. Further, the evaluation and optimization unit optimizes and adjusts the initial risk assessment model according to the predicted risk probability output by the analysis unit. Among them, the optimization process includes: introducing a loss function to evaluate the key nodes in the model to reduce the false alarm rate and improve the prediction accuracy. The evaluation and optimization unit continuously adjusts the model parameters through an iterative algorithm, and finally outputs the optimized risk assessment result. The optimized risk assessment result is the predicted value of the blowout probability within the next 30 minutes, and this predicted value is presented to the operator through the numerical display module for timely taking countermeasures.

[0031] The warning generation unit is used to generate a well control warning instruction according to the optimized risk assessment result. Further, the warning generation unit generates a well control warning instruction according to the optimized risk assessment result. The warning generation unit judges the risk level through the logic control module and generates corresponding warning instructions according to different risk levels. For example, when the predicted blowout probability exceeds a certain threshold, the system will automatically trigger a first-level warning, close specific valves and start emergency equipment; when the predicted blowout probability is low but still needs attention, the system will generate a second-level warning to remind the operator to strengthen monitoring. The well control warning instruction is sent to the control system of the drilling operation platform through the communication interface to ensure that the instruction can be executed quickly.

[0032] Preferably, the system further includes a benchmark risk model determination unit, which is used to connect to the historical operation database and determine the benchmark risk model according to the historical correlation relationship between the well depth information and the formation pressure information; the system further includes a risk distribution node determination unit, which is used to determine the risk distribution nodes of the drilling operation platform on the benchmark risk model, where the risk distribution nodes are key monitoring points that meet the change trend of well control related parameters; the system further includes a segmented risk assessment model establishment unit, which is used to segment the benchmark risk model with the identified risk nodes and output the segmented risk assessment model. Further, the benchmark risk model determination unit and the segmented risk assessment model establishment unit work together to further improve the prediction ability of the system. Among them, the benchmark risk model determination unit connects to the historical operation database, analyzes the historical correlation relationship between the well depth information and the formation pressure information, and determines the benchmark risk model. The key monitoring points in the benchmark risk model are called risk distribution nodes, and these nodes are screened and identified through the algorithm module. The segmented risk assessment model establishment unit divides the benchmark risk model into N segmented models based on the identified risk nodes, and minimizes the error of each segmented model, thereby improving the overall prediction accuracy.

[0033] Preferably, the system further includes a sensor status generation unit, which is used to classify the sensor resources of the drilling operation platform according to the edge computing unit, obtain multiple types of sensors, and generate a sensor status library for the basic information collected by each type of sensor; the system further includes an operation status generation unit, which is used to collect basic information from the historical operation data of the drilling operation platform according to the decision-making layer of the cloud platform and generate an operation status library. Further, the sensor status generation unit and the operation status generation unit are respectively responsible for generating a sensor status memory library and an operation status memory library. The sensor status generation unit classifies the sensor resources of the drilling operation platform, collects the basic information of each type of sensor, including sensitivity, response time, and anti-interference ability, and establishes a sensor status memory library through a screening mechanism. The operation status generation unit, based on the historical operation data of the drilling operation platform, collects information such as operation depth, pressure fluctuation range, and mud density change trend, and generates an operation status memory library. These two memory libraries provide data support for the subsequent optimization process.

[0034] Preferably, the system further includes an optimization unit, which is used to optimize the abnormal mode set and the predicted risk probability based on the sensor status library and the operation status library. Further, a trigger probability generation unit is configured in the optimization unit, which jointly completes the weighted fusion and optimization of data with the optimization unit. The trigger probability generation unit identifies the probability of any mode in the abnormal mode set being triggered in the sensor status memory library according to the abnormal mode set and the sensor status memory library, and obtains a first trigger probability set; at the same time, according to the predicted risk probability and the operation status memory library, it identifies the probability of any probability in the predicted risk probability being triggered in the operation status memory library, and obtains a second trigger probability set. By performing weighted fusion calculation on the first trigger probability set and the second trigger probability set, a fused trigger probability set is output. Finally, the optimization unit performs optimization in the fused trigger probability set and outputs an optimized risk assessment result.

[0035] In some possible implementation manners, the early warning generation unit is further configured with a monitoring information generation unit that works in cooperation, which collaboratively generates segmented monitoring information and well control early warning instructions for the drilling operation platform. The monitoring information generation unit obtains the fused trigger probability set corresponding to each segmented model in the segmented risk assessment model, and extracts the identification sensors and identification operation parameters therefrom as the segmented monitoring information. The early warning generation unit automatically generates well control early warning instructions according to the segmented monitoring information to ensure that the early warning instructions can be accurately adjusted according to the specific conditions of different segments.

[0036] Through the above specific implementation manners, the intelligent well control early warning system provided by the present invention realizes high-precision monitoring and real-time early warning of well control risks, and effectively solves the problems of easy interference of single-sensor data and insufficient response speed existing in the prior art.

[0037] To better enable those in the technical field to fully understand and implement the present invention, the specific implementation principle of the present invention is supplemented and explained below in combination with a specific application scenario as follows:

[0038] In a high-risk drilling operation scenario, the drilling platform is located in an area with complex geological conditions, and the formation pressure fluctuates frequently and unpredictably. To ensure drilling safety, the operator deploys the intelligent early warning system for drilling well control provided by the present invention and gradually executes monitoring and early warning tasks according to the operation logic of the system.

[0039] First, the acquisition unit obtains real-time dynamic data through a sensor cluster. Gas concentration sensors are installed near the wellhead and at the outlet of the mud circulation system to monitor the concentration changes of combustible gases such as methane; pressure sensors are arranged in different depth areas of the wellbore and at the surface high-pressure pipe manifold to capture the subtle fluctuations of the formation pressure; flow sensors are set at the outlet of the mud pump and the return pipeline to monitor the changes in mud flow; mud density sensors are placed in the mud pit and the mud circulation system to evaluate whether the mud density meets the requirements of the current downhole environment. These sensors transmit the collected data to the subsequent unit through the signal transmission module to ensure the continuity and accuracy of the data. During this process, the sensor status generation unit classifies and stores the basic information of each type of sensor, including sensitivity, response time, and anti-interference ability, so as to provide a reliable reference basis for subsequent abnormal pattern recognition.

[0040] Secondly, the establishment unit receives the data transmitted by the acquisition unit and constructs an initial risk assessment model in combination with the well depth information and formation pressure information of the drilling platform. The well depth information is provided by the depth measurement device of the drilling platform, and the formation pressure information is obtained by real-time collection through pressure sensors and comprehensively calculated with the historical formation pressure database. Through the correlation analysis of parameters such as gas concentration, pressure value, flow value, and mud density, the establishment unit determines the correlation relationship between the parameters. For example, when the gas concentration increases significantly, it may indicate an increase in formation pressure and a potential blowout risk; while the decrease in mud density may lead to pressure imbalance in the wellbore and further exacerbate the risk. Based on the above analysis, the establishment unit generates a basic risk assessment framework through the algorithm module, providing a benchmark for subsequent abnormal pattern recognition and risk assessment optimization.

[0041] Subsequently, the recognition unit invokes the edge computing unit to perform real-time anomaly detection on well control related parameters. The edge computing unit is deployed close to the data acquisition point to reduce data transmission latency and improve response speed. For example, when the gas concentration exceeds the preset safety range (such as the methane concentration reaching 50% of the lower explosion limit) or the pressure value fluctuates violently (such as the instantaneous pressure difference exceeding 10 MPa), the recognition unit includes this mode in the anomaly mode set. Each mode in the anomaly mode set corresponds to a specific parameter combination, and these combinations are updated in real time through the fast processing ability of the edge computing unit. At the same time, the calibration information of the sensor is verified through the auxiliary calibration recognition channel to ensure the reliability of the sensor data, which is stored in the classification database.

[0042] Next, the analysis unit connects to the cloud platform decision layer and conducts data interaction with the edge computing unit in the recognition unit to perform multi-source data fusion analysis. The cloud platform decision layer receives the anomaly mode set data from the edge computing unit through the network communication module and conducts comprehensive analysis in combination with historical operation data, real-time monitoring data, and external environment data. For example, a certain anomaly mode may be associated with a high-risk event in historical data, and thus a higher weight is assigned to its predicted risk probability. The analysis unit also uses machine learning algorithms to classify and perform regression analysis on the data to further improve the prediction accuracy. During this process, the trigger probability generation unit identifies the probability of any mode in the anomaly mode set being triggered in the sensor state memory bank based on the anomaly mode set and the sensor state memory bank, obtaining the first trigger probability set; at the same time, based on the predicted risk probability and the operation state memory bank, it identifies the probability of any probability in the predicted risk probability being triggered in the operation state memory bank, obtaining the second trigger probability set. By performing weighted fusion calculation on the first trigger probability set and the second trigger probability set, the fused trigger probability set is output.

[0043] The evaluation and optimization unit optimizes and adjusts the initial risk assessment model according to the predicted risk probability output by the analysis unit. The optimization process includes introducing a loss function to evaluate the key nodes in the model to reduce the false alarm rate and improve the prediction accuracy. For example, for a certain risk distribution node, if there is a large deviation between its prediction result and the actual monitoring data, the false alarm rate loss is calculated through the loss function, and the model parameters are adjusted to reduce the error. The evaluation and optimization unit continuously adjusts the model parameters through an iterative algorithm and finally outputs the optimized risk assessment result. The optimized risk assessment result is the predicted value of the well blowout probability within the next 30 minutes, and this predicted value is presented to the operator through the numerical display module for timely response measures.

[0044] The warning generation unit generates well control warning instructions based on the optimized risk assessment results. The risk level is judged by the logic control module, and corresponding warning instructions are generated according to different risk levels. For example, when the predicted blowout probability exceeds a certain threshold (such as 80%), the system will automatically trigger a first-level warning, close specific valves and start emergency equipment; when the predicted blowout probability is low but still needs attention (such as 30%-50%), the system will generate a second-level warning to remind the operator to strengthen monitoring. The well control warning instructions are sent to the control system of the drilling operation platform through the communication interface to ensure that the instructions can be executed quickly.

[0045] In addition, the establishment unit divides the benchmark risk model into multiple segmented models based on identifying risk nodes as the segmentation basis, and minimizes the error of each segmented model, thereby improving the overall prediction accuracy. The monitoring information generation unit obtains the set of trigger probabilities corresponding to each segmented model in the segmented risk assessment model, and extracts the identification sensors and identification operation parameters therefrom as segmented monitoring information. The warning generation unit automatically generates well control warning instructions according to the segmented monitoring information to ensure that the warning instructions can be accurately adjusted according to the specific conditions of different segments.

[0046] Finally, the optimization unit optimizes the set of risk probabilities and outputs the optimized risk assessment results. The risk assessment results reflect the overall risk level of the current downhole environment and also provide clear response guidance for the operator. For example, in a certain high-risk segment, if the identification sensor shows that the gas concentration continues to rise and the mud density decreases significantly, the system will prompt the operator to immediately adjust the mud density and prepare for emergency well shut-in operations.

[0047] Test item Standard Actual measurement result Early warning accuracy rate (simulated blowout) >95% 97.2%

[0048] Embodiment 2:

[0049] Based on the same inventive concept as the drilling well control intelligent warning system in the foregoing Embodiment 1, as Figure 2 shown, the present invention also provides a drilling well control intelligent warning method. A drilling well control intelligent warning method, which is executed by a processor. The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0050] Specifically, it includes: a data acquisition layer connected to the drilling operation platform, which calls a sensor cluster to obtain well control-related parameters, and collects information on gas concentration, pressure value, flow value, and mud density in the well control-related parameters; obtains the well depth information and formation pressure information of the drilling operation platform, and combines the well control-related parameters to establish an initial risk assessment model; connects to the data acquisition layer and calls the edge computing unit to identify abnormal patterns according to the well control-related parameters, and incorporates the patterns whose identification results meet the preset abnormal threshold into the abnormal pattern set; connects to the data acquisition layer and calls the cloud platform decision-making layer, performs multi-source data fusion analysis after connecting the cloud platform decision-making layer and the edge computing unit, and obtains the predicted risk probability corresponding to each pattern in the abnormal pattern set; optimizes and adjusts the initial risk assessment model according to the predicted risk probability, and outputs the optimized risk assessment result, where the optimized risk assessment result is the predicted value of the blowout probability within the next 30 minutes; generates a well control early warning instruction according to the optimized risk assessment result.

[0051] The specific example of the drilling well control intelligent early warning system in the foregoing Embodiment 1 is equally applicable to the drilling well control intelligent early warning method of this embodiment. Through the foregoing detailed description of the drilling well control intelligent early warning system, those skilled in the art can clearly know the drilling well control intelligent early warning method in this embodiment. Therefore, for the sake of brevity of the specification, it will not be elaborated here.

[0052] The above shows and describes the basic principles, main features, and advantages of the present invention. For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to include all changes falling within the meaning and scope of the equivalent elements of the claims in the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

[0053] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An intelligent early warning system for drilling well control, comprising a processor, characterized in that, It further includes an acquisition unit, a establishment unit, an identification unit, an analysis unit, an evaluation and optimization unit, and an early warning generation unit that are data-connected to the processor; wherein, the acquisition unit is used to connect to the data acquisition layer of the drilling operation platform and call the sensor cluster to obtain well control-related parameters; The establishment unit is used to obtain the well depth information and formation pressure information of the drilling operation platform, and establish an initial risk assessment model in combination with the well control-related parameters; The identification unit is used to connect to the data acquisition layer and call the edge computing unit, identify abnormal patterns according to the well control-related parameters, and incorporate the patterns whose identification results meet the preset abnormal threshold into the abnormal pattern set; The analysis unit is used to connect to the data acquisition layer and call the cloud platform decision-making layer. After connecting the cloud platform decision-making layer and the edge computing unit, it performs multi-source data fusion analysis to obtain the predicted risk probability corresponding to each pattern in the abnormal pattern set; The evaluation and optimization unit is used to optimize and adjust the initial risk assessment model according to the predicted risk probability, and output the optimized risk assessment result; The early warning generation unit is used to generate a well control early warning instruction according to the optimized risk assessment result; The system further includes an optimization unit, and the optimization unit is used to optimize the abnormal pattern set and the predicted risk probability based on the sensor status memory library and the operation status memory library; wherein, a trigger probability generation unit is configured in the optimization unit, and it jointly completes the weighted fusion and optimization of data with the optimization unit; the trigger probability generation unit identifies the probability of any pattern in the abnormal pattern set being triggered in the sensor status memory library according to the abnormal pattern set and the sensor status memory library, and obtains a first trigger probability set; according to the predicted risk probability and the operation status memory library, it identifies the probability of any probability in the predicted risk probability being triggered in the operation status memory library, and obtains a second trigger probability set; By performing weighted fusion calculation on the first trigger probability set and the second trigger probability set, it outputs a fused trigger probability set; the optimization unit performs optimization in the fused trigger probability set and outputs the optimized risk assessment result, wherein the sensor status memory library at least includes sensor sensitivity, response time, and anti-interference ability; the operation status memory library at least includes historical operation data, the acquisition operation depth, the pressure fluctuation range, and the mud density change trend.

2. The system according to claim 1, wherein The system further includes a benchmark risk model determination unit, and the benchmark risk model determination unit is used to connect to the historical operation database and determine the benchmark risk model according to the historical correlation relationship between the well depth information and the formation pressure information.

3. The system according to claim 2, wherein The system further includes a risk distribution node determination unit, and the risk distribution node determination unit is used to determine the risk distribution nodes of the drilling operation platform on the benchmark risk model, wherein the risk distribution nodes are key monitoring points that meet the change trend of the well control-related parameters.

4. The system according to claim 3, characterized in that The system further includes a segmented risk assessment model establishment unit, and the segmented risk assessment model establishment unit is used to segment the benchmark risk model with the identified risk nodes and output a segmented risk assessment model.

5. The system according to claim 4, wherein: The system further includes a sensor status generation unit, which is configured to classify the sensor resources of the drilling operation platform according to the edge computing unit, obtain multiple types of sensors, and generate a sensor status memory library for the acquisition basic information of each type of sensor.

6. The system according to claim 5, characterized in that, The system further includes an operation status generation unit, which is configured to collect basic information from the historical operation data of the drilling operation platform according to the cloud platform decision layer and generate an operation status memory library.

7. A drilling well control intelligent early warning method, the drilling well control intelligent early warning method uses the drilling well control intelligent early warning system described in any one of claims 1-6, and this method is executed by a processor, and is characterized in that, It includes: A data acquisition layer connected to the drilling operation platform, which calls the sensor cluster to obtain well control related parameters, and collects gas concentration, pressure value, flow value and mud density information in the well control related parameters; Obtain the well depth information and formation pressure information of the drilling operation platform, and establish an initial risk assessment model in combination with the well control related parameters; Connect the data acquisition layer and call the edge computing unit to identify abnormal patterns according to the well control related parameters, and incorporate the patterns whose identification results meet the preset abnormal threshold into the abnormal pattern set; Connect the data acquisition layer and call the cloud platform decision layer. After connecting the cloud platform decision layer and the edge computing unit, perform multi-source data fusion analysis to obtain the predicted risk probability corresponding to each pattern in the abnormal pattern set; Optimize and adjust the initial risk assessment model according to the predicted risk probability, and output the optimized risk assessment result, where the optimized risk assessment result is the predicted value of the well blowout probability within the next 30 minutes; Generate a well control early warning instruction according to the optimized risk assessment result.

Images