Drilling machine state automatic identification and error correction method and system based on multi-source data

Abstract

The invention provides a multi-source data-based drilling machine state automatic identification and error correction method and system, and the method comprises the steps: generating a drilling machine state time sequence according to historical well drilling real-time data through a preset drilling state identification model, extracting a historical drilling machine state sequence with a time range as an index through a natural language recognition technology according to historical well drilling log data; constructing model training data according to the drilling machine state time sequence and the historical drilling machine state sequence, and training a preset Transform architecture natural language model through the model training data to obtain a sequence prediction model; and obtaining a drilling machine state identification result according to the real-time data of the current drilling well and the sequence prediction model. Therefore, the problems that due to the quality problem of drilling real-time data and drilling log data, the state of the drilling machine is mistakenly recognized, and due to the fact that drilling log records are not meticulous, drilling performance is difficult to accurately count and analyze are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of drilling engineering, and in particular to a drilling rig state automatic identification and error correction method and system based on multi-source data. BACKGROUND

[0002] Automatic identification of drilling rig state is an important basis for carrying out drilling performance analysis and real-time optimization of drilling, and the accuracy and real-time performance of its identification results are directly related to drilling efficiency and safety. However, the existing methods still have significant limitations in data quality dependence, identification granularity, model training cost, etc., which makes it difficult to achieve high-precision and high-efficiency drilling rig state identification in actual application.

[0003] At present, some researches have tried to use natural language processing technology to automatically extract working condition information from drilling logs. For example, Chinese invention patent CN202110850366.8 discloses a method for intelligently identifying drilling events based on drilling logs, which identifies the working condition of drilling logs and stores them in a structured manner by setting drilling event keyword group rules. Although this method achieves automation to some extent, its identification results are heavily dependent on the comprehensiveness and accuracy of keyword rules, and cannot cope with incomplete log records and non-standard expressions, resulting in rough identification results that cannot meet the needs of fine management.

[0004] On the other hand, machine learning methods based on real-time data have also been gradually applied to drilling rig state identification. Ma Lianrui [1] proposed a method for real-time identification of drilling conditions using logging data, which uses convolutional neural networks, residual networks and long short-term memory networks to build an identification model, and the reported identification accuracy can reach 95%. However, this kind of method highly depends on high-quality and large-scale labeled samples for model training, while in actual drilling environment, it is difficult and costly to obtain labeled data, and the model is sensitive to data noise, which easily leads to misidentification, limiting its promotion and application in actual production.

[0005] In summary, existing technologies either rely on manual rules, have coarse identification granularity and poor adaptability, or rely on a large number of labeled data, have high model training cost and weak noise resistance. Therefore, there is an urgent need for a high-precision drilling rig state identification method that can integrate multi-source data, combine the advantages of rules and deep learning, and have a cross-validation mechanism, to improve the reliability, automation and practical value of the identification results. SUMMARY

[0006] The purpose of this application is to provide a method and system for automatic identification and error correction of drilling rig status based on multi-source data. Based on two different data sources—drilling logs and real-time drilling data—natural language processing, deep learning algorithms, and rule-based algorithms are used respectively to extract and identify drilling rig status. Cross-validation is performed using drilling rig status data obtained from different sources and methods to improve the accuracy of drilling rig status identification. This solves the problems of misidentification of drilling rig status due to the quality issues of real-time drilling data and drilling logs, and the difficulty in accurately statistically analyzing drilling performance due to incomplete drilling log recording. Furthermore, it addresses the challenge of currently used pure data-driven models requiring the construction of complete, high-quality drilling rig status sample data for model training.

[0007] To achieve the above objectives, this application provides an automatic drilling rig status identification and correction method based on multi-source data. The method includes: generating a drilling rig status time series based on historical well drilling real-time data using a preset drilling status identification model; extracting historical drilling rig status sequences indexed by time ranges using natural language recognition technology based on historical well drilling log data; constructing model training data based on the drilling rig status time series and the historical drilling rig status sequences; training a preset Transformer architecture natural language model using the model training data to obtain a sequence prediction model; and obtaining a drilling rig status identification result based on the current drilling real-time data and the sequence prediction model.

[0008] In one embodiment of this application, optionally, the drilling rig status time series includes drilling rig status data arranged according to a preset time period sequence; the drilling rig status data includes the drilling rig's rotary drilling, sliding drilling, reaming, reverse reaming, tripping, lowering, jacking, and stationary states.

[0009] In one embodiment of this application, optionally, extracting a historical drilling rig status sequence indexed by time range from historical well drilling log data using natural language processing technology includes: segmenting the historical well drilling log data into sentences using a natural language processing tool loaded with a drilling rig status domain dictionary to obtain work content description information; extracting time entity data from the work content description information using named entity extraction technology; segmenting the remaining content in the work content description information into words based on the time entity data to obtain drilling data; analyzing the drilling data using the drilling rig status domain dictionary in the natural language processing tool to obtain drilling rig status data for each preset time period in the time entity data; and generating a historical drilling rig status sequence based on the drilling rig status data and the corresponding time period.

[0010] In an embodiment of the present application, optionally, the model training data is generated by cross-validation through time alignment according to the drilling rig state time sequence and the historical drilling rig state sequence.

[0011] In an embodiment of the present application, optionally, the model training data is generated by cross-validation through time alignment according to the drilling rig state time sequence and the historical drilling rig state sequence includes: correcting and perfecting the drilling data of the corresponding time period in the drilling rig state time sequence through the historical drilling rig state sequence; and generating the model training data according to the perfect drilling rig state time sequence.

[0012] In an embodiment of the present application, optionally, the model training data is generated by cross-validation through time alignment according to the drilling rig state time sequence and the historical drilling rig state sequence further includes: supplementing the corresponding drilling rig operation data in the historical drilling rig state sequence according to the perfect drilling rig state time sequence; and obtaining drilling statistical data through the supplemented historical drilling rig state sequence.

[0013] In an embodiment of the present application, optionally, the sequence prediction model is obtained by training the preset Transformer architecture natural language model through the model training data includes: converting the drilling rig state time sequence in the model training data into professional text data; and training the preset Transformer architecture natural language model through the sequence prediction model according to the professional text data, a preset professional vocabulary list and a preset professional vocabulary bias parameter.

[0014] In an embodiment of the present application, optionally, the drilling rig state recognition result is obtained according to the real-time data of the current drilling and the sequence prediction model includes: obtaining to-be-verified data by analyzing the real-time data of the current drilling through the drilling rig state recognition model; generating a predicted drilling rig state through the sequence prediction model according to the real-time data of the current drilling; and obtaining the drilling rig state recognition result by cross-validation of the predicted drilling rig state and the to-be-verified data.

[0015] The application also provides a drilling rig state automatic identification and error correction system based on multi-source data, which comprises an analysis module, a training module and an identification module; the analysis module is used for generating a drilling rig state time sequence according to historical well drilling real-time data through a preset drilling state identification model, and extracting a historical drilling rig state sequence indexed by a time range according to historical well drilling log data through natural language recognition technology; the training module is used for constructing model training data according to the drilling rig state time sequence and the historical drilling rig state sequence, and obtaining a sequence prediction model by training a preset Transformer architecture natural language model through the model training data; and the identification module is used for obtaining a drilling rig state identification result according to real-time data of current drilling and the sequence prediction model.

[0016] The application also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0017] The application also provides a computer readable storage medium storing a computer program for executing the above method.

[0018] The application also provides a computer program product comprising computer programs / instructions, which, when executed by a processor, implement the steps of the above method.

[0019] The application has the beneficial technical effects that the information from different data sources is fully utilized, the reliability of the identification result is improved, and the misidentification caused by data quality is avoided to some extent; the semi-automatic deep learning sample construction technology is utilized to effectively improve the objectivity and efficiency of sample construction, and has important significance for improving the drilling rig state identification accuracy and the automation and intelligence level. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings, which are included to provide a further understanding of the application and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0021] Figure 1 A flowchart of a drilling rig state automatic identification and error correction method based on multi-source data provided by an embodiment of the application;

[0022] Figure 2 An acquisition flowchart of a historical drilling rig state sequence provided by an embodiment of the application;

[0023] Figure 3A A model training data generation flowchart provided by an embodiment of the application;

[0024] Figure 3BA drilling statistics data obtaining process schematic diagram provided by an embodiment of the present application is shown in the figure;

[0025] Figure 4 A sequence prediction model construction process schematic diagram provided by an embodiment of the present application is shown in the figure;

[0026] Figure 5 A drilling rig state recognition result obtaining process schematic diagram provided by an embodiment of the present application is shown in the figure;

[0027] Figure 6 A well depth and drill bit depth and time curve relationship schematic diagram provided by an embodiment of the present application is shown in the figure;

[0028] Figure 7 A structure schematic diagram of a drilling rig state automatic recognition and error correction system based on multi-source data provided by an embodiment of the present application is shown in the figure;

[0029] Figure 8 A structure schematic diagram of an electronic device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0030] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and embodiments, so that how the present application applies technical means to solve technical problems and achieves technical effects can be fully understood and implemented. It should be noted that, unless there is a conflict, each embodiment in the present application and each feature in each embodiment can be combined with each other, and the technical solutions formed thereby are all within the protection scope of the present application.

[0031] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown.

[0032] Please refer to Figure 1 The present application provides a drilling rig state automatic recognition and error correction method based on multi-source data, which specifically includes:

[0033] S101 generating a drilling rig state time sequence according to historical well drilling real-time data through a preset drilling state recognition model, and extracting a historical drilling rig state sequence indexed by a time range according to historical well drilling log data through natural language recognition technology;

[0034] S102 constructing model training data according to the drilling rig state time sequence and the historical drilling rig state sequence, and obtaining a sequence prediction model by training a preset Transformer architecture natural language model through model training data;

[0035] S103 obtains a rig state recognition result according to real-time data of the current drilling and the sequence prediction model.

[0036] The drilling state recognition model can be implemented by a drilling state recognition technology based on rules and threshold setting in the prior art, and will not be described in detail here. The rig state time sequence includes rig state data arranged in a preset time period sequence. The rig state data includes rotary drilling, sliding drilling, reaming, reverse reaming, tripping out, tripping in, sticking, and static, etc. In step S101, the rig state time sequence is mainly recognized according to real-time drilling data, and the rig state is automatically recognized. The recognized rig state includes rotary drilling, sliding drilling, reaming, reverse reaming, tripping out, tripping in, sticking, static, circulation, etc. The drilling state recognition adopts a rule-based method: if the drill bit is at the bottom of the well, the pump is on, and the drill string is rotating, it can be determined as rotary drilling; if the drill bit is at the bottom of the well, the pump is on, and the drill string is not rotating, it can be determined as sliding drilling; if the drill bit is moving upward and the drill string is not rotating, it is determined as tripping out; if the drill bit is moving downward and the drill string is not rotating, it is determined as tripping in; if the drill bit is moving upward and the drill string is rotating, it is determined as reverse reaming; if the drill bit is moving downward and the drill string is rotating, it is determined as reaming; if the drill bit is static and the pump is off, it is determined as static; if the drill bit is static and the pump is on, it is determined as circulation; and if the hook load is less than a preset threshold, it is determined as sticking. The rig state recognition result is a rig state time sequence indexed by time. In this way, the rig state is extracted and recognized in different ways based on multiple source data, and the recognition results are cross-verified, effectively improving the accuracy of rig state recognition and providing a more reliable data basis for the next step of drilling performance analysis. Secondly, the Transformer model is used to predict the latest rig state vocabulary of the rig state sequence, and the recognition result of the rule-based rig state recognition algorithm is combined to improve the reliability of rig state recognition and avoid misrecognition caused by data noise. The specific implementation logic of each step will be described in detail in subsequent embodiments, and will not be described here.

[0037] Please refer to Figure 2 In an embodiment of the present application, the historical rig state sequence indexed by a time range is extracted from historical well drilling log data by natural language recognition technology, as shown in the figure.

[0038] S201 obtains work content description information by using a natural language processing tool to load a rig state field dictionary to perform sentence segmentation on the historical well drilling log data;

[0039] S202 extracts time entity data in the work content description information by using a named entity extraction technology, and performs word segmentation on the remaining content in the work content description information to obtain drilling data according to the time entity data;

[0040] S203 obtaining drilling rig state data of each preset time period in the time entity data by analyzing the drilling rig state field dictionary in the natural language processing tool;

[0041] S204 generating a historical drilling rig state sequence according to the drilling rig state data and the corresponding time period.

[0042] In this embodiment, mainly natural language recognition technology is adopted to automatically extract keywords describing the drilling rig state in the drilling log, and the keywords are combined with the recognized time range to form a drilling rig state sequence indexed by time range. The specific implementation logic of the process is as follows: a drilling rig state field dictionary is constructed and loaded into a Chinese natural language processing tool LTP; LTP is applied to sentence segmentation of the work content description part in the drilling log, time entities in the sentence are extracted by using named entity extraction technology, and the remaining content in the sentence is segmented, and the drilling rig state corresponding to the time entity is extracted in combination with the field dictionary; based on the recognition result of the above steps, a drilling rig state sequence indexed by time range is constructed.

[0043] In an embodiment of the present application, constructing model training data according to the drilling rig state time sequence and the historical drilling rig state sequence includes: generating model training data by cross-validation through time alignment according to the drilling rig state time sequence and the historical drilling rig state sequence. Please refer to Figure 3A illustrated, wherein generating model training data by cross-validation through time alignment according to the drilling rig state time sequence and the historical drilling rig state sequence includes:

[0044] S301 correcting and perfecting the drilling data of the corresponding time period in the drilling rig state time sequence through the historical drilling rig state sequence;

[0045] S302 generating model training data according to the perfected drilling rig state time sequence.

[0046] Please refer to Figure 3B illustrated, further, generating model training data by cross-validation through time alignment according to the drilling rig state time sequence and the historical drilling rig state sequence further includes:

[0047] S303 supplementing the corresponding drilling rig operation data in the historical drilling rig state sequence according to the perfected drilling rig state time sequence;

[0048] S304 obtaining drilling statistics data by analyzing the supplemented historical drilling rig state sequence.

[0049] Specifically, in actual work, the cross-validation of the drilling rig state time sequence and the historical drilling rig state sequence is mainly realized by using time alignment, and the cross-validation content includes:

[0050] Correct the misidentification of the rig state in the rig state time series due to data errors using the historical rig state sequence in a time range extracted by natural language recognition technology. For example, if the rig state time series recognition result is "drilling" in the time range recorded in the drilling log as "running in", it can be corrected to "running in";

[0051] Supplement the recognition result of the rig state time series with information from the drilling log. For example, the drilling log describes a casing running operation, and the recognition result of the rig state time series is drilling down. The recognition result of the rig state time series can be supplemented to: casing running.

[0052] Use the corrected and supplemented rig state time series to refine the operations in the time range of the historical rig state sequence for further statistical analysis of rig operation. For example: the time spent in different rig states, including rotary drilling, sliding drilling, reaming, reverse reaming, tripping out, tripping in, sticking and static, can be counted for each well during drilling, which can be used to analyze effective drilling time and drilling operation optimization space. By counting the total time of rotary drilling and sliding drilling in different formations, the mechanical drilling speed of different formations can be calculated to provide data basis for formation drillability analysis.

[0053] Merge the rig state time series of different wells after cross-validation to generate a training data set for the rig state sequence prediction model.

[0054] In this embodiment, the order of cross-validation is not strictly required. For example, the rig state time series result recognized based on real-time data can be used to refine and correct the rig state sequence indexed by time range in the historical rig state sequence. For example, if the drilling stop time is 2025.06.14 9:41, the drilling termination time in row 3 of Table 1 is corrected to 2025.06.14 9:41, and the cycle start time in row 4 is corrected to 2025.06.14 9:41. Then, the misidentification of the rig state time series is corrected according to the rig state sequence indexed by time range in the historical rig state sequence. For example, according to row 11 of Table 1, the rig state is drilling in the time range 2025.06.15 16:30-2025.06.16 6:30; the rig state recognition algorithm based on real-time data identifies 2025.06.15 17:14:19 as RotateInPumpOn, which is impossible in the drilling process. Therefore, it can be determined that it is misidentified, and the rig state at 2025.06.15 17:14:19 is corrected to rotary drilling.

[0055] Table 1

[0056]

[0057]

[0058] Please refer to Figure 4 In an embodiment of the present application, the sequence prediction model is obtained by training a preset Transformer architecture natural language model based on the model training data, which includes:

[0059] S401 converts the rig state time series in the model training data into professional text data;

[0060] S402 trains a preset Transformer architecture natural language model based on the professional text data, a preset professional vocabulary list, and a preset professional vocabulary bias parameter to obtain a sequence prediction model.

[0061] Specifically, in actual work, this embodiment converts the rig state time series data in the training data set into professional text data, generates a professional vocabulary list according to the identification logic of the drilling state identification model and the supplementary vocabulary of natural language recognition technology. A prediction model based on transformer architecture is constructed, a professional vocabulary bias mechanism is added to improve the prediction probability of professional vocabulary during inference; the model is trained with professional text data, the prediction accuracy of professional vocabulary is monitored during training, learning rate scheduling and gradient clipping are added, and the best model on the validation set is saved.

[0062] In an embodiment of the present application, the drilling rig state identification result obtained according to the real-time data of the current drilling and the sequence prediction model can further include: cross-verification of the calculated current rig state and the predicted rig state by the real-time data of the current drilling, the sequence prediction model, and the drilling state identification model, specifically, please refer to Figure 5 According to the real-time data of the current drilling and the sequence prediction model, the drilling rig state identification result includes:

[0063] S501 analyzes the real-time data of the current drilling to obtain verification data by the drilling state identification model;

[0064] S502 generates a predicted rig state by the sequence prediction model according to the real-time data of the current drilling;

[0065] S503 cross-verify the predicted drilling state and the verification data to obtain a drilling rig state identification result.

[0066] Specifically, the embodiment collects real-time data of the drilling in progress, performs real-time automatic identification of the rig state by using a preset drilling state identification model, forms a text sequence of the current rig state, predicts the probability of different rig states at the latest time point by using a sequence prediction model, and cross- validates the latest rig state identified by using the drilling state identification model, thereby improving the accuracy of rig state identification. In actual work, the embodiment can collect real-time data of the drilling in progress, perform real-time automatic identification of the rig state by using the drilling state identification model on the real-time data of well 2# 2025 / 06 / 19 2:50:04-3:05:34, and form a text sequence of the current rig state. The sequence prediction model is used to predict the probability of different rig states at the next time point (2025 / 6 / 19 3:06:04), and the prediction result is that the Still state is the highest and the EmptyWellBore is the lowest close to 0. Based on the real-time data at the time point of 2025 / 6 / 19 3:06:04, the rig state identified by using the drilling state identification model is EmptyWellBore. The identification result is cross-validated with the prediction result. Since the prediction probability of the identified state is close to 0, the identification result is corrected to the Still state with the highest prediction probability. Analysis of the real-time drilling data shows that there is a lot of noise in the drill bit position data of well 2# in this time period (as shown in Figure 6 The misidentification of the drilling state at the time point of 2025 / 6 / 19 3:06:04 is caused by the noise of the drill bit position at this time point. As can be seen from the example data, the cross-validation method can improve the accuracy of drilling state identification.

[0067] Therefore, the present application extracts and identifies the rig state in different ways based on multi-source data, cross-validates the identification results, effectively improves the accuracy of rig state identification, and provides a more reliable data basis for the next drilling performance analysis. At the same time, the Transformer model is used to predict the latest rig state vocabulary of the rig state sequence, the identification result of the rule-based rig state identification algorithm is combined, the reliability of rig state identification is improved, and misidentification caused by data noise is avoided.

[0068] Please refer to Figure 7As shown, this application also provides an automatic drilling rig status identification and correction system based on multi-source data. The system includes an analysis module, a training module, and an identification module. The analysis module is used to generate a drilling rig status time series based on historical well drilling real-time data using a preset drilling status identification model, and to extract historical drilling rig status sequences indexed by time ranges based on historical well drilling log data using natural language recognition technology. The training module is used to construct model training data based on the drilling rig status time series and the historical drilling rig status sequences, and to train a preset Transformer architecture natural language model using the model training data to obtain a sequence prediction model. The identification module is used to obtain the drilling rig status identification result based on the current drilling real-time data and the sequence prediction model.

[0069] The beneficial technical effects of this application are as follows: it makes full use of information from different data sources, improves the reliability of the recognition results, and avoids misidentification due to data quality to a certain extent; by using semi-automatic deep learning sample construction technology, it effectively improves the objectivity and efficiency of sample construction, which is of great significance to improving the accuracy of drilling rig status recognition and the degree of automation and intelligence.

[0070] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.

[0071] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.

[0072] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0073] like Figure 8 As shown, the electronic device 600 may also include: a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily need to include these components. Figure 8 All components shown; in addition, the electronic device 600 may also include Figure 8 For components not shown, please refer to existing technologies.

[0074] like Figure 8 As shown, the central processing unit 100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operation of various components of the electronic device 600.

[0075] The memory 140, for example, can be one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The above-mentioned information related to failure can be stored, and in addition, a program for executing the information related to failure can be stored. The central processing unit 100 can execute the program stored in the memory 140 to implement information storage or processing, etc.

[0076] The input unit 120 provides input to the central processing unit 100. The input unit 120 is, for example, a key or a touch input device. The power supply 170 is used to supply power to the electronic device 600. The display 160 is used to display display objects such as images and characters. The display can be, for example, an LCD display, but is not limited thereto.

[0077] The memory 140 can be a solid state memory such as a read only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and is provided with more data, and examples of such a memory are sometimes referred to as an EPROM, etc. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 can include an application / function storage section 142 for storing application programs and function programs or a flow for executing the operation of the electronic device 600 by the central processing unit 100.

[0078] The memory 140 can also include a data storage section (data 143) for storing data such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. A driver storage section (driver 144) of the memory 140 can include various drivers of the electronic device for a communication function and / or for performing other functions of the electronic device such as a messaging application, a contact application, etc.

[0079] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processing unit 100 to provide input signals and receive output signals, which can be the same as in the case of a conventional mobile communication terminal.

[0080] Based on different communication technologies, multiple communication modules 110, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, can be provided in the same electronic device. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and to receive audio input from the microphone 132 to enable typical telecommunication functions. The audio processor 130 can include any suitable buffers, decoders, amplifiers, etc. In addition, the audio processor 130 is coupled to the central processor 100 to enable recording on-board via the microphone 132 and to enable playing on-board stored sounds via the speaker 131.

[0081] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, a system, or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0082] The present application is described in reference to the flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams block or blocks. Figure 1 means for performing one or more functions specified in the flowchart illustrations and / or block diagrams block or blocks.

[0083] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart illustrations and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams block or blocks. Figure 1 means for performing one or more functions specified in the flowchart illustrations and / or block diagrams block or blocks.

[0084] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are generated to realize the computer-implemented processes in the computer or other programmable devices, and the instructions executed in the computer or other programmable devices provide the processes for implementing the functions specified in the flowchart Figure 1 one flow or multiple flows and / or the functions specified in the block Figure 1 Figure 1 one block or multiple blocks.

[0085] The above-described specific embodiments have further detailed the purposes, technical solutions and beneficial effects of the present application. It should be understood that the above-described is only a specific embodiment of the present application and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for automatic identification and error correction of drilling rig status based on multi-source data, characterized in that, The method includes: Based on real-time drilling data from historical wells, a drilling rig status time series is generated using a pre-set drilling status recognition model. Based on historical well drilling log data, a historical drilling rig status sequence indexed by time range is extracted using natural language recognition technology. Model training data is constructed based on the drilling rig status time series and the historical drilling rig status series. A sequence prediction model is obtained by training a preset Transformer architecture natural language model using the model training data. The drilling rig status identification result is obtained based on the real-time data of the current drilling and the sequence prediction model.

2. The method for automatic identification and error correction of drilling rig status based on multi-source data according to claim 1, characterized in that, The drilling rig status time series includes drilling rig status data arranged according to a preset time period sequence; the drilling rig status data includes the drilling rig's rotary drilling, sliding drilling, reaming, reverse reaming, tripping, jacking up, jacking up, jacking up, and stationary states.

3. The method for automatic identification and error correction of drilling rig status based on multi-source data according to claim 1, characterized in that, Based on historical well drilling log data, historical drilling rig status sequences indexed by time ranges were extracted using natural language processing technology, including: The historical well drilling log data is segmented using a natural language processing tool that loads a drilling rig status domain dictionary to obtain descriptive information about the work content; Time entity data is extracted from the work content description information using named entity extraction technology, and drilling data is obtained by word segmentation of the remaining content in the work content description information based on the time entity data. The drilling data for each preset time period in the time entity data is obtained by analyzing the drilling data using the drilling rig status domain dictionary in the natural language processing tool. A historical drilling rig status sequence is generated based on the drilling rig status data and the corresponding time period.

4. The method for automatic identification and error correction of drilling rig status based on multi-source data according to claim 1, characterized in that, Constructing model training data based on the drilling rig status time series and the historical drilling rig status series includes: generating model training data by cross-validating the drilling rig status time series and the historical drilling rig status series through time alignment.

5. The method for automatic identification and error correction of drilling rig status based on multi-source data according to claim 4, characterized in that, Model training data is generated by cross-validating the drilling rig status time series and the historical drilling rig status series through time alignment, including: The drilling data for the corresponding time period in the drilling rig status time series is corrected and improved by the historical drilling rig status sequence. Model training data is generated based on the improved drilling rig status time series.

6. The method for automatic identification and error correction of drilling rig status based on multi-source data according to claim 5, characterized in that, The model training data generated by cross-validating the drilling rig status time series and the historical drilling rig status series through time alignment also includes: Supplement the corresponding drilling rig operation data in the historical drilling rig status sequence based on the improved drilling rig status time sequence; Drilling statistics were obtained through the supplemented historical drilling rig status sequence analysis.

7. The method for automatic identification and error correction of drilling rig status based on multi-source data according to claim 1, characterized in that, Sequence prediction models are obtained by training a pre-defined Transformer architecture natural language model using model training data, including: Convert the drilling rig status time series in the model training data into professional text data; A sequence prediction model is obtained by training a pre-defined Transformer architecture natural language model based on the professional text data, a pre-defined list of professional terms, and pre-defined professional term bias parameters.

8. The method for automatic identification and error correction of drilling rig status based on multi-source data according to claim 1, characterized in that, The drilling rig status identification results obtained based on the current real-time drilling data and the sequence prediction model include: The data to be verified is obtained by analyzing the real-time data of the current drilling through the drilling status identification model; The predicted drilling rig status is generated based on the real-time data of the current drilling operation using the sequence prediction model; The predicted drilling status and the data to be verified are cross-validated to obtain the drilling rig status identification result.

9. A drilling rig status automatic identification and error correction system based on multi-source data, characterized in that, The system includes: an analysis module, a training module, and a recognition module; The analysis module is used to generate a drilling rig status time series based on historical well drilling real-time data through a preset drilling status identification model, and to extract historical drilling rig status sequences indexed by time range based on historical well drilling log data through natural language recognition technology. The training module is used to construct model training data based on the drilling rig status time series and the historical drilling rig status series, and to train a preset Transformer architecture natural language model through the model training data to obtain a sequence prediction model. The identification module is used to obtain the drilling rig status identification result based on the real-time data of the current drilling and the sequence prediction model.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 8.

12. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 8.

Images