CN120402041B - AI-based intelligent control system for marine precision pressure control drilling - Google Patents
AI-based intelligent control system for marine precision pressure control drillingInfo
- Publication number
- CN120402041B CN120402041B CN202510762907.XA CN202510762907A CN120402041B CN 120402041 B CN120402041 B CN 120402041B CN 202510762907 A CN202510762907 A CN 202510762907A CN 120402041 B CN120402041 B CN 120402041B
- Authority
- CN
- China
- Prior art keywords
- control
- value
- time
- marking
- optimization
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B44/00—Automatic control systems specially adapted for drilling operations, i.e. self-operating systems which function to carry out or modify a drilling operation without intervention of a human operator, e.g. computer-controlled drilling systems; Systems specially adapted for monitoring a plurality of drilling variables or conditions
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B7/00—Special methods or apparatus for drilling
- E21B7/12—Underwater drilling
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- Physics & Mathematics (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Feedback Control In General (AREA)
Abstract
本发明属于海洋钻井领域，涉及数据分析技术，用于解决现有技术在同一控制参数出现设置数值分析结果相悖时无法对控制参数进行合理化设置的问题，具体是基于AI技术的海洋精细控压钻井智能控制系统，包括依次连接的参数控制模块、补偿控制模块、保护调节模块以及保护优化模块；参数控制模块用于对海底钻井设备进行参数控制分析：生成若干个控制组并计算控制组的偏离系数，通过偏离系数对设置优化值进行标记；将设置优化值发送至补偿控制模块；本发明有效解决多源监测参数导致的控制指令冲突问题，同时避免因控制滞后引发的井筒堵塞风险，提升深海钻井作业的安全性和控制精度。
This invention belongs to the field of marine drilling and relates to data analysis technology. It addresses the problem in existing technologies where conflicting numerical analysis results for the same control parameter prevent the rational setting of control parameters. Specifically, it is an AI-based intelligent control system for marine precision pressure control drilling, comprising a parameter control module, a compensation control module, a protection adjustment module, and a protection optimization module connected in sequence. The parameter control module performs parameter control analysis on the subsea drilling equipment: generating several control groups and calculating the deviation coefficients of the control groups, marking the optimized setting values based on the deviation coefficients, and sending the optimized setting values to the compensation control module. This invention effectively solves the problem of control command conflicts caused by multi-source monitoring parameters, while avoiding the risk of wellbore blockage due to control lag, thus improving the safety and control accuracy of deep-sea drilling operations.
Description
Technical Field
The invention belongs to the field of ocean drilling, relates to a data analysis technology, and particularly relates to an intelligent control system for ocean fine pressure control drilling based on an AI technology.
Background
The intelligent control system for the marine fine pressure control well drilling is a high-end technical system for deep sea or complex stratum well drilling operation, aims to ensure well drilling safety, improve efficiency and reduce risks by monitoring and dynamically regulating and controlling well bore pressure in real time, and is a key technology for deep water oil gas development, and the intelligent and automatic level of the intelligent control system directly determines operation safety.
The invention patent with the publication number of CN115822550B discloses an intelligent control system for ocean fine pressure control drilling, which is beneficial to reducing the difficulty of data acquisition and the working complexity of technicians in the ocean fine pressure control drilling process, realizing the effective utilization of various parameters and improving the accuracy and the working efficiency of control instructions, but the submarine drilling environment is complex, the setting values obtained by analyzing the same control parameters from different angles can have larger differences, and the control parameters cannot be rationally set under the condition in the prior art, so that the risk of submarine drilling is high.
The application provides a solution to the technical problem.
Disclosure of Invention
The invention aims to provide an intelligent control system for ocean fine pressure control drilling based on an AI technology, which is used for solving the problem that the control parameters cannot be reasonably set when the set numerical analysis results of the same control parameters are contradictory in the prior art;
The technical problem to be solved by the invention is how to provide the intelligent control system for the ocean fine pressure control drilling based on the AI technology, which can reasonably set the control parameters when the set numerical analysis results of the same control parameters are contradictory.
The aim of the invention can be achieved by the following technical scheme:
the intelligent control system for the marine fine pressure control well drilling based on the AI technology comprises a parameter control module, a compensation control module, a protection adjusting module and a protection optimizing module which are connected in sequence;
the parameter control module is used for performing parameter control analysis on the submarine drilling equipment, and generating a plurality of control groups, calculating deviation coefficients of the control groups, marking a set optimization value through the deviation coefficients, and sending the set optimization value to the compensation control module;
The compensation control module is used for carrying out calibration compensation control analysis on the submarine drilling equipment, wherein before the control of the control parameters, a compensation optimization value is generated by setting the optimization value, and the numerical value of the control parameters is set by the compensation optimization value;
The protection adjusting module is used for adjusting and analyzing the hydrate generation critical point of the submarine drilling equipment, namely, acquiring temperature data, fluid data and external data of the submarine drilling equipment, inputting the acquired temperature data, fluid data and external data into an LSTM neural network model for hydrate generation critical point analysis and obtaining a critical point threshold value;
the protection optimization module is used for carrying out optimization analysis on the protection liquid injection process of the submarine drilling equipment.
Further, the generation process of the control group comprises the steps of marking control parameters of the submarine drilling equipment as control objects i, i=1, 2, & gt, n and n are positive integers, marking environmental parameters of the submarine drilling as monitoring objects e, e=1, 2, m and m are positive integers, wherein each control object i corresponds to one or more monitoring objects e, the control objects i and the corresponding monitoring objects e form a control group, and it can be understood that when the control object i is an opening of a throttle valve, the control objects i correspond to the temperature gradient of the shaft in the monitoring objects e (when the viscosity of the drilling fluid is reduced due to high temperature), the opening of the throttle valve is required to be increased to compensate the pressure loss, the sea current speed (when a strong sea current water isolation pipe vibrates, the throttle valve is triggered to inhibit the pressure fluctuation), the concentration of hydrogen sulfide (when the concentration exceeds the standard, the self-circulation rate of the sea current speed) and the bottom pressure of the well bore exceed the threshold value, and the bottom pressure fluctuation are set in the value of the well bore, and the bottom pressure fluctuation is set to be in the value of the throttle valve, and the bottom pressure fluctuation is set to be in the value of the well bore hole when the temperature fluctuation is controlled by the throttle valve, and the bottom pressure fluctuation is set to be in the value of the throttle valve is set to be in the value of 10 MPa when the throttle valve is exceeded.
Further, the obtaining process of the deviation coefficient of the control group includes that the monitoring objects e in the control group are arranged according to the sequence of response priority from first to last to obtain a priority sequence, the response priority represents the priority degree of control timeliness of corresponding parameters, for example, the priority sequence of the monitoring objects e in the control group corresponding to the throttle opening is hydrogen sulfide concentration-bottom hole pressure fluctuation-shaft temperature gradient-sea current speed, the setting value SZie of the control object i is obtained according to the real-time monitoring numerical analysis of the monitoring objects e, and then variance calculation is carried out on the setting values SZie of the control object i corresponding to all the monitoring objects e to obtain the deviation coefficient.
Further, the specific process of marking the set optimization value includes comparing the deviation coefficient with a preset deviation threshold, summing all the set values SZie to average to obtain the set optimization value if the deviation coefficient is smaller than the deviation threshold, and setting the set values SZie to be set optimization values in sequence according to the order of the priority sequence if the deviation coefficient is larger than or equal to the deviation threshold, wherein the set values SZie after the sequence is required to be set after the previous monitored object e is stable.
Further, the generation process of the compensation optimization value comprises the steps of obtaining the apparent numerical value and the actual numerical value of the control object i, marking the difference value of the apparent numerical value and the actual numerical value as the compensation value of the control object i, and marking the sum value of the set optimization value and the compensation value of the control object i as the compensation optimization value of the control object i.
Further, the temperature data acquisition process comprises the steps of acquiring temperature and pressure values of each section of a shaft and a pipeline through a temperature sensor, the fluid data acquisition process comprises the steps of detecting gas concentration, water content and supersaturation degree in fluid components, and the external data acquisition process comprises the step of acquiring the temperature and the flow rate of seawater in the region where the shaft is located.
Further, the specific process of optimizing and analyzing the protection liquid injection process of the submarine drilling equipment by the protection optimizing module comprises the steps of generating an optimizing conversion ratio, marking the product of the critical point threshold value and the optimizing conversion ratio as a critical point optimizing value after the subsequent LSTM neural network model outputs the critical point threshold value, comparing the real-time parameter of the current working condition of the shaft with the critical point optimizing value, and marking the subsequent injection time.
Further, the generation process of the optimized conversion ratio comprises the steps of marking the time difference between the effective time and the injection time after the concentration and the injection quantity of the protection liquid are regulated as time difference values, marking the injection optimizing time, marking the time difference between the injection optimizing time and the injection time before the injection time as time difference values, and marking the ratio of the real-time parameter of the injection optimizing time to the threshold value of the critical point as the optimized conversion ratio.
The invention has the following beneficial effects:
1. The problem of control command conflict caused by multisource monitoring parameters is effectively solved, the rationality and consistency of the control command are ensured by dynamically selecting a parameter optimization strategy through a deviation coefficient, meanwhile, the real-time prediction of a hydrate generation critical point and the prejudgment optimization of the injection time of the protective liquid are realized, the risk of shaft blockage caused by control hysteresis is avoided, and the safety and control precision of deep sea drilling operation are improved;
2. the problem of unreasonable setting caused by single control parameter setting angle in the prior art is solved, and the parameter setting difference under different analysis dimensions is reduced by establishing a dynamic association relation between the control parameters and the environmental parameters, so that the risk of blowout or equipment damage caused by parameter deviation in submarine drilling operation is reduced;
3. The method has the advantages that the generation risk of the hydrate can be more accurately identified, the injection mechanism of the protective liquid is triggered in advance when the working condition of the shaft is close to the critical point, the problem of hysteresis response caused by incomplete parameter monitoring is avoided, and therefore the potential risk that the shaft or a pipeline is blocked by the hydrate is effectively reduced;
4. The problem of the injection hysteresis of the protection liquid is effectively solved, the protection liquid is ensured to be regulated before the working condition of the shaft reaches the actual dangerous threshold value by dynamically correcting the threshold value of the critical point and combining a time difference value compensation mechanism, the generation risk of the hydrate is obviously reduced, and the timeliness of injection control is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a system block diagram of a first embodiment of the present invention;
Fig. 2 is a flowchart of a method according to a second embodiment of the invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the prior art, an intelligent control system for ocean fine pressure control drilling is a key technology for deep water oil gas development, and drilling safety is ensured by monitoring and dynamically regulating and controlling wellbore pressure in real time. However, the submarine drilling environment is complex, the setting values obtained by analyzing the same control parameters from different angles may have large differences, and the control parameters cannot be rationally set under the condition in the prior art, so that the risk of submarine drilling is high. For example, in a deep sea high pressure low temperature environment, wellbore pressure control parameters may produce conflicting analysis results due to temperature gradients, fluid composition differences, etc., and it is difficult for conventional systems to coordinate conflicts between multi-source data.
In order to solve the above problems, a control method capable of coordinating the multi-dimensional parameter conflicts and realizing the dynamic optimization is required. Through analysis, the parameter setting conflict is derived from the differential influence of different monitoring objects on the same control parameter, and a priority ordering mechanism is required to be established to eliminate the contradiction. Meanwhile, the dynamic change of the hydrate generation critical point requires that the system has real-time prediction capability and needs to be comprehensively judged by combining the environmental parameters and the fluid state. In addition, the hysteresis of the injection timing of the protection liquid may cause potential safety hazards, and a pre-judging mechanism needs to be established to optimize the injection timing.
In the first embodiment, as shown in fig. 1, the application provides an intelligent control system for ocean fine pressure control drilling based on an AI technology, which comprises a parameter control module, a compensation control module, a protection adjusting module and a protection optimizing module which are sequentially connected. The parameter control module generates a control group and calculates a deviation coefficient, an optimized value is set through a deviation coefficient mark, the compensation control module generates a compensation optimized value to set parameters, the protection adjustment module acquires multidimensional data and inputs the multidimensional data into the LSTM model to analyze a critical point threshold value, the protection liquid parameter is adjusted when the real-time parameter reaches the threshold value, and the protection optimization module generates an optimized conversion ratio to adjust the critical point threshold value.
The deviation coefficient of the parameter control module refers to a set value variance value of the control object corresponding to a plurality of monitoring objects, and the deviation coefficient can be specifically realized through a variance calculation formula and is used for quantifying the comprehensive influence degree of different monitoring parameters on the same control object. The compensation optimization value of the compensation control module refers to an overlapped value of the deviation between the set optimization value and the display numerical value, and the method can be realized by acquiring the difference value between the actual numerical value and the feedback numerical value of the control system through a sensor, and is used for eliminating the execution error of equipment. The threshold value of the critical point of the protection and adjustment module refers to a boundary value of a hydrate generation condition predicted by an LSTM neural network, and the threshold value can be specifically realized by adopting a multidimensional data training model of temperature, pressure, fluid components and the like and is used for dynamically judging the injection triggering condition of the protection liquid. The optimized conversion ratio of the protection optimization module refers to the ratio of the pre-judging parameter at the injection time to the threshold value of the critical point, and can be specifically realized through historical effective time difference data modeling and is used for triggering the injection action of the protection liquid in advance.
Specifically, the parameter control module associates the control parameters with the environmental parameters to form a control group, and evaluates the degree of dispersion of the parameter settings by variance calculation. When the deviation coefficient is lower than the threshold value, the average value is adopted as an optimized value, and when the deviation coefficient is higher than the threshold value, the deviation coefficient is set step by step according to the priority sequence. And the compensation control module superimposes the error compensation value by the equipment to form a final control instruction. The protection and regulation module acquires wellbore temperature, fluid components and seawater environment data in real time, inputs a trained LSTM model prediction critical point, and immediately triggers the concentration regulation of the protection liquid when the real-time pressure or temperature reaches a prediction threshold value. The protection optimization module establishes an optimization conversion ratio according to the historical effective time difference, and adjusts a critical point trigger value in advance in the follow-up prediction so as to synchronize the injection action of the protection liquid with the change of the working condition.
Compared with the prior art, the prior system adopts a simple weighted average method when processing the multi-source parameter conflict, cannot identify the priority relation among the parameters, and easily causes the control instruction contradiction. According to the scheme, the degree of dispersion of the parameters is identified through the deviation coefficient, a hierarchical setting mechanism is established, and the mutual influence of the multidimensional parameters is effectively coordinated. The prior art relies on a fixed threshold value to judge the injection time of the protective liquid, and is difficult to adapt to dynamic environment changes. According to the scheme, the LSTM model is used for predicting the critical point in real time, the injection time is predicted by combining the optimized conversion ratio, and the control timeliness is remarkably improved.
Through the technical scheme, the control instruction conflict problem caused by the multisource monitoring parameters can be effectively solved, and the rationality and consistency of the control instructions are ensured by dynamically selecting the parameter optimization strategy through the deviation coefficient. Meanwhile, real-time prediction of a hydrate generation critical point and pre-judgment optimization of injection time of the protective liquid are realized, the risk of shaft blockage caused by control hysteresis is avoided, and the safety and control precision of deep sea drilling operation are improved.
Marking control parameters of the submarine drilling equipment as control objects i, i=1, 2, & gt, n, n being a positive integer, marking environmental parameters of the ocean drilling as monitoring objects e, e=1, 2, & gt, m, m being a positive integer, wherein each control object i corresponds to one or more monitoring objects e, and a control group is formed by the control objects i and the corresponding monitoring objects e.
The control object i refers to a parameter in the subsea drilling device, which needs to be dynamically adjusted, such as a wellbore pressure, a drilling fluid flow rate or a drill bit rotating speed, and specifically can be realized by acquiring data in real time through a pressure sensor, a flowmeter or a rotating speed sensor. The monitoring object e refers to an environmental parameter related to drilling safety, such as sea water temperature, flow rate or formation pressure, and specifically, a temperature sensor, a flow rate meter or a pressure sensor can be used for monitoring. The control group consists of a control object i and an associated monitoring object e, and is used for establishing a dynamic association relation between the control parameter and the environment parameter, so that one-sided analysis of a single parameter is avoided.
Specifically, when generating a control group, control parameters are first divided into independent control objects by functional category, for example, wellbore pressure is divided into one control object i, and drilling fluid flow is divided into another control object i. Each control object i is then assigned an environmental monitoring parameter associated with it, for example a wellbore pressure control object i may be associated with a monitoring object e of sea water temperature, flow rate etc. according to the actual requirements of the drilling operation. By means of the grouping mode, adjustment of control parameters and real-time environment change can be bound, for example, when the flow rate of seawater is suddenly increased, wellbore pressure control parameters related to the seawater are preferentially adjusted, and therefore dynamic optimization of parameter setting is achieved.
In contrast to the prior art, the control parameters are typically set based on only a single dimension, e.g., based on only historical wellbore pressure data, and the real-time impact of environmental parameters on the control effect is ignored. According to the scheme, the environment monitoring parameters and the equipment control parameters are associated by constructing the control group, so that the parameter setting can comprehensively consider the multi-dimensional dynamic change, for example, when the temperature of the seawater fluctuates, the associated drilling fluid flow parameters can be synchronously adjusted, and the control deviation caused by the environmental mutation is avoided.
Through the technical scheme, the problem of unreasonable setting caused by single control parameter setting angle in the prior art can be solved, and the parameter setting difference under different analysis dimensions is reduced by establishing the dynamic association relation between the control parameters and the environmental parameters, so that the risk of blowout or equipment damage caused by parameter deviation in submarine drilling operation is reduced.
The process for obtaining the deviation coefficient of the control group comprises the steps of arranging the monitoring objects e in the control group according to the response priority from first to last to obtain a priority sequence, analyzing the real-time monitoring values of the monitoring objects e to obtain the set value SZie of the control object i, and then performing variance calculation on the set values SZie of the control object i corresponding to all the monitoring objects e to obtain the deviation coefficient.
The priority sequence refers to an order formed by arranging the monitoring objects e according to response priorities, and can be specifically implemented by adopting a preset priority rule or an algorithm based on dynamic adjustment of environmental parameters, so as to determine the processing order of the influence of different monitoring objects on control parameters. The set value SZie refers to a parameter set value corresponding to the real-time data of the monitored object e, and specifically can be obtained by acquiring the real-time monitored value through a sensor and combining with a preset mapping relation or regression model. The deviation coefficient refers to a quantization index reflecting the set value dispersion degree of the same control object i under different monitoring objects e through variance calculation, and specifically, a variance formula in statistics can be adopted to calculate the value dispersion degree of a plurality SZie.
Specifically, after the priority sequence is generated, the real-time monitoring data of each monitoring object e is input into the analysis model, and converted into the setting value SZie of the corresponding control object i through a preset algorithm or mapping relation. Subsequently, for the same control object i, its corresponding plurality SZie is extracted and variance calculation is performed, and the variance result is defined as a deviation coefficient. For example, when the control object is wellbore pressure, the monitoring objects may include subsea flow rate, temperature and fluid composition, and after SZie corresponding to each monitoring object is generated by independent analysis, the variance calculation may reflect the degree of dispersion of the pressure set point by different monitoring objects.
Compared with the prior art, the prior art does not consider the problem of numerical value dispersion of the same control parameter in different monitoring dimensions, so that the parameter setting has a conflict risk. According to the scheme, the variance calculation is used for quantifying the discrete degree, and the priority sequence is combined for determining the processing sequence, so that the parameter setting process can process the data of the high-priority monitoring object preferentially, and meanwhile the rationality of parameter setting is judged through the deviation coefficient.
The specific process for marking the set optimized value comprises the steps of comparing the deviation coefficient with a preset deviation threshold, summing all set values to obtain an average value to obtain the set optimized value if the deviation coefficient is smaller than the deviation threshold, and sequentially setting the set values as the set optimized value according to the order of a priority sequence if the deviation coefficient is larger than or equal to the deviation threshold, wherein the set values after the sequence are required to be set after the previous monitoring object is stable.
The deviation coefficient refers to a quantization index reflecting the fluctuation degree of the control parameter setting value, and can be specifically realized by calculating the variance of the corresponding setting value of each monitoring object, so as to evaluate the stability of the control parameter setting.
The deviation threshold value refers to a preset reference value for judging whether the parameter setting is stable or not, and can be specifically realized by adopting a critical value obtained by statistical analysis of historical operation data, and is used for dividing the application conditions of the parameter optimization strategy.
The priority sequence refers to an operation sequence formed by sequencing the monitoring objects according to the response priority, and can be specifically realized by setting weight coefficients of different monitoring parameters, and is used for determining the sequence logic of parameter adjustment.
Specifically, when the deviation coefficient of the control group is lower than the preset threshold, it is indicated that the fluctuation of the set values corresponding to the monitoring objects is small, and at this time, the average value of the set values is used as the optimized value, so that the equalization processing of the parameter setting is realized. When the deviation coefficient exceeds the threshold value, the significant parameter fluctuation is indicated, the parameters are gradually set according to the priority order of the monitoring objects, the adjustment of the follow-up parameters is required to be executed after the running state of the previous monitoring object is stable, and therefore system oscillation caused by simultaneous adjustment of multiple parameters is avoided.
Compared with the prior art, the prior art lacks an effective integration mechanism for different analysis results of the same control parameter, so that the parameter setting has a collision risk. According to the scheme, the fluctuation degree of the parameter is quantitatively estimated through the deviation coefficient, and the average optimization or step-by-step optimization strategy is dynamically selected according to the estimation result, so that the parameter setting efficiency in a low fluctuation scene is ensured, and the adjustment stability in a high fluctuation scene is realized.
The generation process of the compensation optimization value comprises the steps of obtaining a display value and an actual value of a control object i, marking a difference value between the display value and the actual value as a compensation value of the control object i, and marking a sum value of a set optimization value and the compensation value of the control object i as the compensation optimization value of the control object i.
The display numerical value refers to data displayed by the control object in an equipment operation interface or a sensor, and can be specifically realized by adopting a numerical value acquired by real-time monitoring equipment such as a pressure sensor, a flowmeter and the like and transmitted to a control system.
The actual numerical value refers to a parameter state of the control object actually existing in the physical environment, and the control object can be realized by performing secondary verification through independent calibration equipment or a redundant measurement device.
The compensation value refers to an error amount between the apparent numerical value and the actual numerical value, and specifically can be achieved by comparing the apparent numerical value and the actual numerical value in real time by a difference value calculation module and generating a correction parameter.
The compensation optimization value refers to a final set value of the control parameter after error correction, and the final set value can be realized by performing superposition operation on the set optimization value and the compensation value through an adder.
In particular, during control parameter setting, the apparent values may deviate due to sensor drift, environmental disturbances or transmission delays. By acquiring the actual value and calculating the compensation value, the influence of the system error on the control accuracy can be eliminated. The setting optimization value is an initial control parameter generated based on environmental parameter analysis, and the compensation optimization value obtained after the compensation value is overlapped is output to the submarine drilling equipment as a final execution instruction. For example, when the apparent value of the pressure control parameter is 10MPa and the actual value is 9.8MPa, the compensation value of 0.2MPa will be superimposed on the set optimization value, ensuring that the control command matches the actual operating condition.
Compared with the prior art, the prior art only depends on a single data source for parameter setting, and does not consider systematic deviation of the device display value and the true value. The application can dynamically correct the control parameter setting error by introducing a compensation value calculation mechanism. For example, in the patent with publication number CN115822550B, the control parameters are directly generated by adopting monitoring data, and an error compensation link is not involved, but the application effectively avoids the problem of control misalignment caused by equipment errors through double verification of the apparent numerical value and the actual numerical value.
Through the technical scheme, the problem of drilling risk caused by deviation of control parameter setting in the prior art is solved. By generating the compensation optimization value in real time, the control parameter can be ensured to be matched with the actual working condition accurately, for example, in the control of the well bore pressure, the application of the compensation optimization value can reduce the pressure fluctuation range to be within the safety threshold, so that the generation of hydrate or blowout accidents caused by parameter setting errors are avoided.
The temperature data acquisition process comprises the steps of acquiring temperature and pressure values of each section of a shaft and a pipeline through a temperature sensor, detecting gas concentration, water content and supersaturation degree in fluid components, acquiring sea water temperature and flow rate of a region where the shaft is located through an external data acquisition process, specifically, acquiring temperature and pressure distribution of each section of the shaft through a distributed sensor network to form a thermodynamic state section, synchronously detecting the gas concentration, the water content and the supersaturation degree in the fluid through a plurality of types of sensors through the fluid data acquisition process, reflecting the phase stability of the fluid in the shaft in real time, and acquiring sea water temperature and flow rate around the shaft through marine environment monitoring equipment through the external data acquisition process, wherein the external data acquisition process is used for evaluating the influence of external environment on heat exchange of the shaft. The three types of data are integrated and input into an LSTM neural network model, and a dynamic prediction model of a hydrate generation critical point is established through time sequence feature learning. When the real-time parameter reaches the critical point threshold, the system triggers the judgment logic of the injection time of the protection liquid, and dynamically adjusts the concentration and the injection amount of the protection liquid according to the fluid components and the environmental parameters.
The specific process of the protection optimization module for carrying out optimization analysis on the protection liquid injection process of the submarine drilling equipment comprises the steps of generating an optimization conversion ratio, marking the product of a critical point threshold value and the optimization conversion ratio as a critical point optimization value after a subsequent LSTM neural network model outputs the critical point threshold value, comparing the real-time parameter of the current working condition of a shaft with the critical point optimization value, and marking the subsequent injection time. The generation process of the optimized conversion ratio comprises the steps of marking the time difference value between the effective time and the injection time after the concentration and the injection quantity of the protection liquid are regulated as time difference values, marking the injection optimizing time, marking the time difference value between the injection optimizing time and the injection time before the injection time as time difference values, and marking the ratio of the real-time parameter of the injection optimizing time and the threshold value of the critical point as the optimized conversion ratio.
The optimized conversion ratio is an adjustment coefficient established by the dynamic relation between the historical adjustment effective time difference value and the critical point threshold value, and can be specifically realized by adopting the calculation of the ratio of the time difference value to the real-time parameter, and is used for introducing a dynamic correction factor on the basis of the critical point threshold value. The critical point optimization value refers to an adjustment result of the original critical point threshold based on optimization conversion comparison, and can be specifically realized through multiplication operation and used for triggering the protection liquid injection action in advance. The injection optimization time is a preset starting time point of injection of the protection liquid according to the effective time difference, and can be specifically realized through forward calculation of the time difference, and is used for reserving a response period required by the protection liquid to be effective.
Specifically, after the LSTM neural network model outputs the hydrate generation critical point threshold, the optimized conversion ratio is dynamically calculated and applied to generate the critical point optimized value. And continuously comparing the real-time parameters of the shaft with the optimized values of the critical points, and triggering the injection time mark if the real-time parameters reach the optimized values. At this time, the injection action of the protection liquid is performed in advance to the injection optimization time, so that the concentration and the injection amount of the protection liquid are adjusted before the actual reaching of the critical working condition, and the generation risk of the hydrate caused by the delay of the effectiveness is avoided.
The generation process of the optimized conversion ratio comprises the steps of marking the time difference value between the effective time and the injection time after the concentration and the injection quantity of the protection liquid are regulated as time difference values, marking the injection optimizing time, marking the time difference value between the injection optimizing time and the injection time before the injection time as time difference values, and marking the ratio of the real-time parameter of the injection optimizing time and the threshold value of the critical point as the optimized conversion ratio.
The time difference value refers to a time interval required from execution to actual effectiveness of the adjustment operation of the concentration or injection amount of the protection liquid, and specifically can be realized by recording a time difference between the execution time of the adjustment instruction and the stabilization time of the feedback parameter of the sensor by using a timer module, and the characteristic is used for quantifying the response delay characteristic of the adjustment operation of the protection liquid.
The injection optimization time is an optimal starting time point of the protection liquid adjusting operation calculated based on the time difference value, and can be specifically determined by a time period of a time difference value shifted forward on the basis of the injection time, and the characteristic is used for establishing dynamic time correlation between the protection liquid adjusting operation and a critical point threshold value.
The optimization conversion ratio refers to a dynamic proportionality coefficient of a real-time working condition parameter and a critical point threshold value at the injection optimization moment, and can be specifically realized by acquiring wellbore temperature and pressure data in real time and carrying out ratio operation on the critical point threshold value output by an LSTM model, and the characteristic is used for constructing a dynamic matching mechanism of the injection timing of the protection liquid and the change of the working condition parameter.
Specifically, in the protection liquid injection optimization process, the time difference from execution to effect of the protection liquid concentration or injection amount adjustment operation is recorded as a time difference value through history data. When the LSTM neural network model outputs a critical point threshold of the current working condition, the system automatically marks the time length of the difference value when the injection time is shifted forwards as the injection optimization time when the protection liquid adjusting operation needs to be started. At the moment, the wellbore temperature and pressure parameters at the injection optimization moment are acquired in real time, and the ratio operation is carried out on the parameters and the threshold value of the critical point to generate the optimization conversion ratio. The conversion ratio is applied to the dynamic correction of the subsequent critical point threshold value, so that the protection liquid adjusting operation can be finished to be effective before the actual working condition reaches the critical point threshold value.
In the second embodiment, as shown in fig. 2, the intelligent control method for ocean fine pressure control drilling based on the AI technology comprises the following steps:
The method comprises the steps of performing parameter control analysis on the submarine drilling equipment, namely forming a control group by a control object i and a corresponding monitoring object e, performing variance calculation on set values SZie of the control object i corresponding to all the monitoring objects e in the control group to obtain a deviation coefficient, and controlling the value of the control object i through the deviation coefficient;
step two, calibrating compensation control analysis is carried out on the submarine drilling equipment, namely a compensation optimization value is obtained through the apparent numerical value and the actual numerical value of the control object, and parameter setting is carried out on the control object i through the compensation optimization value;
step three, adjusting and analyzing the hydrate generation critical point of the submarine drilling equipment, namely acquiring temperature data, fluid data and external data of the submarine drilling equipment, inputting the acquired temperature data, fluid data and external data into an LSTM neural network model for hydrate generation critical point analysis and obtaining a critical point threshold value;
And fourthly, optimizing and analyzing the protection liquid injection process of the submarine drilling equipment, namely marking the time difference value between the effective time and the injection time after the concentration and the injection quantity of the protection liquid are regulated as a time difference value, and marking the critical point optimizing value through the time difference value.
The intelligent control system for the ocean fine pressure control well drilling based on the AI technology is characterized in that when the intelligent control system for the ocean fine pressure control well drilling based on the AI technology works, a parameter control module generates a control group and calculates a deviation coefficient, an optimized value is set through a deviation coefficient mark, a compensation control module generates a compensation optimized value to set parameters, a protection adjusting module collects multi-dimensional data and inputs the multi-dimensional data into an LSTM model to analyze a critical point threshold value, when a real-time parameter reaches the threshold value, a protection liquid parameter is adjusted, and a protection optimizing module generates an optimized conversion ratio to adjust the critical point threshold value.
The foregoing is merely illustrative of the structures of this invention and various modifications, additions and substitutions for those skilled in the art can be made to the described embodiments without departing from the scope of the invention or from the scope of the invention as defined in the accompanying claims.
In the description of the present specification, the descriptions of the terms "one embodiment," "example," "specific example," and the like, mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (4)
1. The intelligent control system for the marine fine pressure control well drilling based on the AI technology is characterized by comprising a parameter control module, a compensation control module, a protection regulation module and a protection optimization module which are connected in sequence;
the parameter control module is used for performing parameter control analysis on the submarine drilling equipment, and generating a plurality of control groups, calculating deviation coefficients of the control groups, marking a set optimization value through the deviation coefficients, and sending the set optimization value to the compensation control module;
The compensation control module is used for carrying out calibration compensation control analysis on the submarine drilling equipment, wherein before the control of the control parameters, a compensation optimization value is generated by setting the optimization value, and the numerical value of the control parameters is set by the compensation optimization value;
The protection adjusting module is used for adjusting and analyzing the hydrate generation critical point of the submarine drilling equipment, namely, acquiring temperature data, fluid data and external data of the submarine drilling equipment, inputting the acquired temperature data, fluid data and external data into an LSTM neural network model for hydrate generation critical point analysis and obtaining a critical point threshold value;
the protection optimization module is used for carrying out optimization analysis on the protection liquid injection process of the submarine drilling equipment;
The acquisition process of the deviation coefficient of the control group comprises the steps of arranging the monitoring objects e in the control group according to the response priority from first to last to obtain a priority sequence, analyzing the real-time monitoring values of the monitoring objects e to obtain a set value SZie of the control object i, and performing variance calculation on the set values SZie of the control object i corresponding to all the monitoring objects e to obtain the deviation coefficient;
The specific process for marking the set optimized value comprises the steps of comparing the deviation coefficient with a preset deviation threshold value, summing all set values SZie and taking an average value to obtain the set optimized value if the deviation coefficient is smaller than the deviation threshold value, and setting the set values SZie as the set optimized values in sequence according to the order of a priority sequence if the deviation coefficient is larger than or equal to the deviation threshold value, wherein the set values SZie after the sequence is required to be set after the previous monitored object e is stable;
The generation process of the compensation optimization value comprises the steps of obtaining a display value and an actual value of a control object i, marking a difference value between the display value and the actual value as a compensation value of the control object i, and marking a sum value of a set optimization value and the compensation value of the control object i as the compensation optimization value of the control object i.
2. The intelligent control system for marine fine pressure control drilling based on the AI technology as set forth in claim 1, wherein the generation process of the control group comprises marking control parameters of the subsea drilling equipment as control objects i, i=1, 2, where n, n is a positive integer, marking environmental parameters of the marine drilling as monitoring objects e, e=1, 2, where m, m is a positive integer, each control object i corresponds to one or more monitoring objects e, and the control group is formed by the control objects i and the corresponding monitoring objects e.
3. The intelligent control system for marine fine pressure control drilling based on the AI technology as set forth in claim 2, wherein the temperature data acquisition process comprises the steps of acquiring temperature and pressure values of each section of a shaft and a pipeline through a temperature sensor, the fluid data acquisition process comprises the steps of detecting gas concentration, water content and supersaturation degree in fluid components, and the external data acquisition process comprises the step of acquiring the temperature and flow rate of sea water in an area where the shaft is located.
4. The intelligent control system for ocean fine pressure control drilling based on AI technology as set forth in claim 3, wherein the specific process of optimizing the injection process of the protection fluid into the subsea drilling equipment by the protection optimizing module comprises generating an optimized conversion ratio, marking the product of the critical point threshold and the optimized conversion ratio as a critical point optimized value after the subsequent LSTM neural network model outputs the critical point threshold, comparing the real-time parameter of the current working condition of the well bore with the critical point optimized value, and marking the subsequent injection time;
the generation process of the optimized conversion ratio comprises the steps of marking the time difference between the effective time and the injection time after the concentration and the injection quantity of the protection liquid are regulated as time difference values, marking the injection optimizing time, marking the time difference between the injection optimizing time and the injection time before the injection time as time difference values, and marking the ratio of the real-time parameter of the injection optimizing time and the threshold value of the critical point as the optimized conversion ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510762907.XA CN120402041B (en) | 2025-06-09 | 2025-06-09 | AI-based intelligent control system for marine precision pressure control drilling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510762907.XA CN120402041B (en) | 2025-06-09 | 2025-06-09 | AI-based intelligent control system for marine precision pressure control drilling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120402041A CN120402041A (en) | 2025-08-01 |
| CN120402041B true CN120402041B (en) | 2026-01-06 |
Family
ID=96519001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510762907.XA Active CN120402041B (en) | 2025-06-09 | 2025-06-09 | AI-based intelligent control system for marine precision pressure control drilling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120402041B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120575835B (en) * | 2025-08-06 | 2025-11-11 | 吉林大学 | Exploration drilling parameter control method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115822550A (en) * | 2022-07-01 | 2023-03-21 | 中国石油天然气集团有限公司 | Intelligent control system for ocean fine pressure control drilling |
| CN220562897U (en) * | 2023-08-28 | 2024-03-08 | 山东禹越电气自动化有限公司 | Anchor machine is equipped in ocean well drilling |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007092956A2 (en) * | 2006-02-09 | 2007-08-16 | Weatherford/Lamb, Inc. | Managed pressure and/or temperature drilling system and method |
| EA033474B1 (en) * | 2012-11-13 | 2019-10-31 | Exxonmobil Upstream Res Co | Method to detect drilling dysfunctions |
| US11939859B2 (en) * | 2017-10-02 | 2024-03-26 | Schlumberger Technology Corporation | Performance based condition monitoring |
| RU2723805C9 (en) * | 2019-08-20 | 2020-07-27 | Общество с ограниченной ответственностью "Диджитал Петролеум" (ООО "ДП") | Method and computer system for control of drilling of the wells |
| WO2022204723A1 (en) * | 2021-03-26 | 2022-09-29 | Schlumberger Technology Corporation | Field equipment data system |
| CN119809058A (en) * | 2025-03-11 | 2025-04-11 | 广东石油化工学院 | A method and system for predicting the production capacity of natural gas hydrate mining |
-
2025
- 2025-06-09 CN CN202510762907.XA patent/CN120402041B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115822550A (en) * | 2022-07-01 | 2023-03-21 | 中国石油天然气集团有限公司 | Intelligent control system for ocean fine pressure control drilling |
| CN220562897U (en) * | 2023-08-28 | 2024-03-08 | 山东禹越电气自动化有限公司 | Anchor machine is equipped in ocean well drilling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120402041A (en) | 2025-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7611617B2 (en) | A method for constructing a digital twin for a multi-physical facility system by combining reduced-order models, measurement data, and machine learning techniques | |
| CN120402041B (en) | AI-based intelligent control system for marine precision pressure control drilling | |
| JP6482699B2 (en) | Use of predictive means in process control systems with wireless or intermittent process measurements | |
| NO340159B1 (en) | Methods, systems and computer-readable media for real-time oil and gas field production optimization using proxy simulator | |
| Andrade et al. | Virtual flow metering of production flow rates of individual wells in oil and gas platforms through data reconciliation | |
| CN105045233B (en) | The Optimization Design of PID controller based on time metric in Power Plant Thermal system | |
| CN120197778A (en) | Intelligent prediction method for urban pipe network operation based on physical information graph neural network | |
| TWI696057B (en) | Method of determining output flow rate of flow rate control device | |
| CN114036875A (en) | Method and system for acquiring thermophysical parameters of supercritical fluid working medium | |
| CN118050992B (en) | Throttle valve opening control quantity optimization method, device, equipment and storage medium | |
| CN120650774A (en) | Indoor temperature real-time regulation and control method and control system for heating power pipeline | |
| Persechini et al. | Centralized and distributed control architectures under Foundation Fieldbus network | |
| CN113778141A (en) | A Segmental PID Controlled Fluid Pipeline Flow Control System | |
| CN114326395B (en) | An online update method for intelligent generator set control model based on working condition discrimination | |
| CN113341694B (en) | Equivalent verification system and method for flow control valve control strategy | |
| CN119847248B (en) | Intelligent automatic temperature adjustment method and system for temperature controller | |
| CN120162554A (en) | A method for optimizing energy efficiency of diving equipment based on environmental data prediction | |
| CN119826161A (en) | Steam generator water level control method, system and device based on digital twin | |
| CN114912364A (en) | Natural gas well flow prediction method, device, equipment and computer readable medium | |
| CN120295378A (en) | Flow control method, device, electronic device and storage medium of mass flow meter | |
| CN120449691A (en) | Automatic scoring method and system for thermal power generator simulation operation console | |
| CN118780777A (en) | Intelligent management system and method for water conservancy engineering equipment data based on digital twin | |
| CN115877811A (en) | Process technology treatment method, device and equipment | |
| Nakaya et al. | On-line simulator for plant operation | |
| CN106950835A (en) | Succinct Robust Two-Degree-of-Freedom proportional plus integral control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |