Multi-layer gas injection working condition intelligent early warning method and system based on real-time data, medium and equipment

Abstract

The invention relates to the field of petroleum and natural gas exploitation engineering, and discloses a multi-layer gas injection working condition intelligent early warning method and system based on real-time data, a medium and equipment. The method comprises the steps that a corresponding dynamic attenuation or dynamic increasing dynamic reference model is built for each gas injection layer based on historical real-time gas injection amount data, obtaining the predicted gas injection amount of the layer so as to determine the real-time flow interval of each gas injection layer; a moving average model of wellhead gas injection pressure is established for each gas injection position, a pressure moving average value changing along with time is obtained so as to predict a wellhead gas injection pressure value at the current moment, and a real-time pressure interval of each gas injection position is determined; according to the real-time flow interval and the real-time pressure interval, whether the real-time flow and the real-time pressure collected at the current moment exceed the corresponding interval ranges or not is determined, if the real-time flow and the real-time pressure exceed the ranges and set time, early warning is carried out, the early warning process of each layer is achieved through one thread, and multi-layer synchronous early warning is achieved through multiple threads.

Description

Technical Field

[0001] This invention relates to the field of oil and gas extraction engineering technology, and in particular to a method, system, medium and equipment for intelligent early warning of multi-layer gas injection conditions based on real-time data. Background Technology

[0002] Multi-layer gas injection is a complex and high-investment engineering technology, fraught with various technical challenges and risks. Without an effective early warning system, it can easily lead to serious technical failures, safety accidents, and economic losses. The fundamental purpose of early warning is to shift from "passively responding to failures" to "proactively preventing risks," ensuring the "safety, efficiency, and controllability" of gas injection operations.

[0003] A patent document with publication number CN 119357839 A discloses a gas well condition diagnosis method based on the gradient boosting decision tree (GBDT) algorithm model. This method predicts liquid accumulation or hydrate blockage in gas wells by preprocessing production data, feature filtering, and model training. However, the core of this method lies in the complex offline machine learning model training and classification. Its model construction relies on batch processing of historical datasets, making it difficult to meet the real-time early warning requirements for second-level response in gas injection conditions. Furthermore, its data dimensions mainly revolve around the gas and liquid production parameters of production wells, which are fundamentally different from the dynamic decreasing patterns of gas injection volume and pressure in injection wells. Therefore, it cannot be directly applied to dynamic benchmark modeling and rapid anomaly identification for multi-layered gas injection conditions.

[0004] A patent document with publication number CN 116220660 A discloses a multi-layer gas injection control hardware scheme. This scheme uses wireless wavecode communication between a surface controller and a downhole intelligent gas distributor to monitor layered temperature, pressure, and flow data in real time, and to control the CO2 injection volume. While this method directly involves "multi-layer gas injection" and "real-time data acquisition," it is essentially a hardware control system (relying on a downhole gas distributor and wireless communication module) rather than a software algorithm-based early warning scheme. It does not involve dynamic benchmark modeling of abnormal operating conditions or intelligent early warning logic. Its core function focuses on "gas injection volume control," lacking proactive risk identification and early warning capabilities, and thus failing to achieve the core objective of "proactive prevention."

[0005] A patent document with publication number CN 116663901 A discloses a layered gas injection gas channeling risk early warning scheme. This scheme identifies gas channeling risks by establishing a layered gas injection parallel model, plotting gas injection volume-pressure characteristic curves, and monitoring changes in start-up pressure and flow coefficient. While this method addresses gas injection anomalies, it is essentially a macroscopic reservoir engineering analysis method based on periodic well testing. Its data acquisition is discontinuous, and the analysis cycle is long, making it unable to capture anomalies in continuous real-time data at the second level during production. Furthermore, its early warning targets focus on the specific risk of "gas channeling," without covering broader daily operational anomalies such as sudden increases or decreases in injection volume and pressure anomalies.

[0006] In summary, existing technologies either fail to match the diagnostic targets with the gas injection conditions, are limited to hardware execution and lack an intelligent analysis kernel, or have overly simplistic warning targets with insufficient real-time performance. Currently, there is a lack of an intelligent early warning method that can build a dedicated dynamic benchmark for each gas injection layer based on real-time data streams, and achieve end-to-end, low-latency, multi-parameter collaborative monitoring through lightweight and efficient algorithms. Summary of the Invention

[0007] To address the aforementioned problems, the present invention aims to provide a multi-layer gas injection condition intelligent early warning method, system, medium, and device based on real-time data. Through dynamic benchmark modeling, multi-parameter correlation analysis, and multi-threaded parallel early warning mechanism, it achieves real-time monitoring and anomaly identification of key parameters such as gas injection volume and pressure, solving the problems of high false alarms and missed alarms, large response delays, and poor adaptability of traditional early warning methods.

[0008] To achieve the above objectives, in a first aspect, the technical solution adopted by the present invention is as follows: a multi-layer gas injection condition intelligent early warning method based on real-time data, comprising: for each gas injection layer, constructing a corresponding dynamic benchmark model with dynamic decay or dynamic increase based on historical real-time gas injection volume data, and obtaining the predicted gas injection volume of the layer through the dynamic benchmark model; for each gas injection layer, establishing a moving average model of wellhead gas injection pressure, and obtaining the pressure moving average value that changes over time; predicting the gas injection volume value at the current moment based on the dynamic benchmark model to determine the real-time flow range of each gas injection layer; predicting the wellhead gas injection pressure value at the current moment based on the pressure moving average value to determine the real-time pressure range of each gas injection layer; determining whether the real-time flow and real-time pressure collected at the current moment exceed the corresponding range according to the real-time flow range and real-time pressure range, and issuing an early warning if they exceed the range and exceed a set time, wherein the early warning process for each layer is implemented using a single thread, and multi-layer synchronous early warning is achieved through multi-threading.

[0009] Furthermore, a dynamic baseline model with corresponding dynamic decay or dynamic increase is constructed based on historical real-time gas injection data. The predicted gas injection volume for this layer is obtained through the dynamic baseline model, specifically: Based on historical real-time gas injection data, a dynamic benchmark model is obtained by fitting an exponentially decreasing or increasing function. The baseline for the dynamic decay or increase of gas injection volume at this layer, obtained through the dynamic baseline model, is as follows:

[0010] In the formula, This is a predicted value for the future gas injection volume of the stratum; This is the initial gas injection volume; It represents either a decreasing rate or an increasing rate.

[0011] Furthermore, based on the dynamic benchmark model, the current gas injection volume is predicted to determine the real-time flow range for each injection layer: .

[0012] Furthermore, for each injection layer, a moving average model of the wellhead injection pressure is established to obtain the moving average pressure over time, specifically: Set a length of The sliding window is used to add new pressure data points to the current sliding window as soon as they are obtained, while removing the oldest data point in the current sliding window. The moving average model continuously calculates the average value of the data within the current sliding window, which serves as both the current trend value and the predicted value for the next moment.

[0013] Furthermore, the average value of the data within the sliding window is always calculated using the moving average model:

[0014] In the formula, It is a point in time. The moving average, also The predicted value at any given time; It is a point in time. The actual pressure value; It is the size of the sliding window.

[0015] Furthermore, the current wellhead injection pressure is predicted based on the moving average pressure value to determine the real-time pressure range for each injection layer: .

[0016] Furthermore, based on the real-time traffic range and real-time pressure range, it is determined whether the real-time traffic and pressure exceed the corresponding range. If they exceed the range and exceed the set time, an alert is issued. Each level of the alert process is implemented using a single thread, and multi-threading is used to achieve multi-level synchronous alerts. Specifically: One thread per layer, starting with the layer gas injection volume and wellhead gas injection pressure. The total wellhead injection pressure is predicted in the future wellhead gas injection pressure using a moving average model to identify whether the injection pressure of a single layer is abnormal. At the same time, the stratified gas volume prediction is simultaneously fitted with an increasing or decreasing function based on the respective historical gas injection volume data to predict the future gas injection volume of each stratum, thereby identifying whether the gas injection volume of a single stratum is abnormal. Based on abnormal wellhead gas injection pressure and abnormal gas injection volume at each layer, structured early warning information is generated, which includes anomaly location, cause inference and handling suggestions, and is pushed to remote monitoring screens and mobile terminals in real time.

[0017] Secondly, the technical solution adopted by this invention is as follows: a multi-layer gas injection condition intelligent early warning system based on real-time data, comprising: a gas injection volume prediction module for each layer, which constructs a corresponding dynamic benchmark model with dynamic decay or dynamic increase based on historical real-time gas injection volume data for each gas injection layer, and obtains the predicted gas injection volume for that layer through the dynamic benchmark model; a wellhead gas injection pressure prediction module, which establishes a moving average model of wellhead gas injection pressure for each gas injection layer, and obtains the moving average pressure value that changes over time; a judgment interval determination module, which predicts the gas injection volume value at the current moment based on the dynamic benchmark model to determine the real-time flow interval of each gas injection layer, and predicts the wellhead gas injection pressure value at the current moment based on the moving average pressure value to determine the real-time pressure interval of each gas injection layer; and an early warning module, which determines whether the real-time flow and real-time pressure collected at the current moment exceed the corresponding interval range according to the real-time flow interval and the real-time pressure interval, respectively. If they exceed the range and exceed the set time, an early warning is issued. The early warning process for each layer is implemented using a single thread, and multi-layer synchronous early warning is achieved through multi-threading.

[0018] Thirdly, the technical solution adopted by the present invention is: a computer-readable storage medium for storing one or more programs, wherein the one or more programs include instructions, which, when executed by a computing device, cause the computing device to perform any of the methods described above.

[0019] Fourthly, the technical solution adopted by the present invention is: a computing device comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing any of the methods described above.

[0020] The present invention has the following advantages due to the adoption of the above technical solutions: 1. This invention constructs a dynamic benchmark for gas injection volume and a moving average trend of wellhead pressure specific to each layer, and combines it with a multi-threaded parallel early warning mechanism, which can significantly improve the accuracy and timeliness of anomaly identification and effectively overcome the shortcomings of traditional fixed threshold method with many false alarms and missed alarms and large response delay of complex machine learning model.

[0021] 2. This invention realizes intelligent monitoring with low latency at the second level across the entire chain from data acquisition to early warning push. It can not only adapt to the dynamic changes of the entire production stage of gas injection wells, but also provide decision support for the site, including anomaly location and handling suggestions, thereby significantly reducing production risks and powerfully promoting the intelligent management and control transformation of gas injection operations from "passive response" to "proactive prevention". Attached Figure Description

[0022] Figure 1 This is an overall flowchart of the intelligent early warning method for multi-layer gas injection conditions based on real-time data in an embodiment of the present invention; Figure 2 This is a detailed flowchart of the intelligent early warning method for multi-layer gas injection conditions based on real-time data in this embodiment of the invention; Figure 3 This is a multi-layer formation parameter diagram of the gas injection well configuration in this embodiment of the invention; Figure 4 This is a parameter anomaly baseline setting diagram in an embodiment of the present invention; Figure 5 This is a diagram of real-time injection parameter data collected at each layer in an embodiment of the present invention; Figure 6 This is a diagram showing the early warning results at time 1 in an embodiment of the present invention; Figure 7 This is a diagram showing the early warning result at time 2 in an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0024] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0025] In one embodiment of the present invention, a multi-layer gas injection operation condition intelligent early warning method based on real-time data is provided. Specifically, it relates to operation condition monitoring and risk early warning technology for multi-layer gas injection operations. This method is applicable to the identification of abnormal operation conditions throughout the entire production stage of multi-layer gas injection wells, from new well injection and mid-term operations to near shutdown. It can achieve proactive safety early warning for gas injection operations, ensuring the safety, efficiency, and controllability of the gas injection process. In this embodiment, as... Figure 1 , Figure 2 As shown, the method includes the following steps: 1) For each gas injection layer, a dynamic baseline model with dynamic decay or dynamic increase is constructed based on historical real-time gas injection data. The predicted gas injection volume for that layer is obtained through the dynamic baseline model. 2) For each gas injection layer, establish a moving average model of wellhead gas injection pressure to obtain the moving average pressure value over time; 3) Predict the current gas injection volume based on the dynamic benchmark model to determine the real-time flow range of each gas injection layer; predict the current wellhead gas injection pressure based on the pressure moving average to determine the real-time pressure range of each gas injection layer. 4) Determine whether the real-time traffic and pressure collected at the current moment exceed the corresponding range based on the real-time traffic range and real-time pressure range. If they exceed the range and exceed the set time, multiple warnings will be issued. Each level of warning process is implemented using a thread, and multi-level synchronous warnings are achieved through multi-threading.

[0026] In step 1) above, a pressure sensor and a flow meter are installed at the wellhead of each gas injection layer.

[0027] In this embodiment, after acquiring historical real-time gas injection volume data, data preprocessing is performed, including cleaning, filtering, and standardization.

[0028] In step 1) above, the dynamic benchmark model naturally depicts the rapid decline or growth characteristics of gas injection wells due to geological conditions and production history, avoiding the false alarm problems caused by traditional "one-size-fits-all" fixed thresholds. It sets simple deviation rules (such as continuous exceedances) and multi-parameter correlation logic (such as abnormal pressure-flow ratios) around the dynamic benchmark, resulting in low computational complexity. This meets the response speed requirements (within 10 seconds) for real-time alarms (non-predictive), and the results are easy for field personnel to understand and verify. Compared to machine learning classification models that require a large number of labeled samples, are complex to train, and have slightly weaker interpretability, this method is more direct, efficient, and robust in achieving the core objective (distinguishing between normal decline (increase) and anomalies).

[0029] Specifically, a dynamic baseline model with corresponding dynamic decay or dynamic increase is constructed based on historical real-time gas injection data. The predicted gas injection volume for this layer is obtained through the dynamic baseline model, which includes the following steps: 1.1) Based on historical real-time gas injection data, a dynamic benchmark model is obtained by fitting an exponentially decreasing or increasing function; 1.2) The baseline for the dynamic decay or increase of gas injection volume at this layer is obtained through the dynamic baseline model:

[0030] In the formula, This represents the predicted future gas injection volume for the stratum, in tens of thousands of cubic meters per day. This is the initial gas injection volume, in units of 10,000 cubic meters per day; The decreasing or increasing rate is obtained by fitting it from historical data using the least squares method.

[0031] In this embodiment, the current gas injection volume is predicted based on a dynamic baseline model to determine the real-time flow range for each gas injection layer: The recommended values ​​for the upper and lower limits of this range can be provided by learning from historical real-time data.

[0032] The dynamic benchmark model in this embodiment is applicable to gas injection wells at different production stages. Whether it is a newly injected well, a well that has been injected for a period of time, or a gas injection well that is about to be shut down, it can construct a dynamic benchmark based on its own historical data and accurately identify abnormal gas injection volume.

[0033] In step 2) above, this embodiment considers not only the abnormal injection volume but also the abnormal wellhead injection pressure. The production dynamics (injection pressure parameters) of a gas injection well are a typical time series data, and their changes have the following characteristics: (1) Time dependence: The current production status is strongly dependent on the status in the past period (e.g., yesterday's gas injection volume will affect today).

[0034] (2) Nonlinearity: There are complex nonlinear relationships among the factors affecting the gas injection volume (underground seepage, well flow, equipment conditions, etc.).

[0035] (3) Long-term effects: Effects such as injection reduction and changes in fracture conductivity will affect injection dynamics over a long time scale.

[0036] Therefore, in this embodiment, a simple moving average model is used for the wellhead gas injection pressure. This model captures the pressure trend by directly calculating the arithmetic mean of the most recent n periods of data. Compared with traditional complex models such as RNN, it is simple to calculate, does not have gradient vanishing or exploding problems, and can effectively capture long-term dynamic patterns from historical data, making it very suitable for real-time trend prediction and anomaly identification of gas injection well injection pressure.

[0037] Specifically, for each injection layer, a moving average model of the wellhead injection pressure is established to obtain the moving average pressure over time. This includes the following steps: 2.1) Set a length of The sliding window is used to add new pressure data points to the current sliding window as soon as they are obtained, while removing the oldest data point in the current sliding window. 2.2) The moving average model is used to continuously calculate the average value of the data within the current sliding window, which serves as the current trend value and the predicted value for the next moment.

[0038] In this embodiment, the average value of the data within the sliding window is always calculated using the moving average model:

[0039] In the formula, It is a point in time. The moving average, also The predicted value at any given time; It is a point in time. The actual pressure value, in MPa; It is the size of the sliding window (e.g., the past 5 minutes, 10 data points).

[0040] In this embodiment, the current wellhead gas injection pressure is predicted based on the moving average pressure value to determine the real-time pressure range for each injection layer: [ The recommended values ​​for the upper and lower limits of this range can be provided after learning from historical real-time data.

[0041] In step 4) above, it is determined whether the real-time traffic and real-time pressure exceed the corresponding range based on the real-time traffic range and real-time pressure range. If they exceed the range and exceed the set time, an early warning is issued. Each level of the early warning process is implemented using one thread, and multi-threaded synchronous early warning is achieved. Specifically, it includes the following steps: 4.1) One thread per layer, starting from the layer gas injection volume and wellhead gas injection pressure. The total wellhead injection pressure is predicted to predict the future wellhead gas injection pressure using a moving average model in order to identify whether the injection pressure of a single layer is abnormal. 4.2) Simultaneously, the stratified gas volume prediction is fitted with an increasing or decreasing function based on the respective historical gas injection volume data to predict the future gas injection volume of each stratum, thereby identifying whether the gas injection volume of a single stratum is abnormal. 4.3) Based on the abnormal wellhead gas injection pressure and the abnormal gas injection volume of each layer, a structured early warning information containing anomaly location, cause inference and handling suggestions is generated and pushed to the remote monitoring screen and mobile terminal in real time.

[0042] In this embodiment, the time is set to 5 minutes. (Regarding real-time traffic...) Exceeding the real-time flow range for 5 consecutive minutes or injecting pressure If the pressure exceeds the real-time pressure range for 5 consecutive minutes, it is determined to be an anomaly at that level.

[0043] In this embodiment, the gas injection well has five layers. First, a unified data acquisition layer aggregates data from the wellhead injection pressure sensor and the gas volume meters of each layer in real time, and performs cleaning, filtering, and standardization preprocessing. Then, the system activates a multi-threaded parallel prediction engine: one thread for each layer, starting with the gas injection volume and wellhead injection pressure. The total wellhead injection pressure is calculated using a moving average method to predict future wellhead injection pressure, keenly detecting pressure drops or overpressure risks. Simultaneously, the layered gas volume prediction is fitted with an increasing / decreasing function based on the historical gas injection volume data of each layer, and then further predicts the future gas injection volume of each layer, thereby accurately identifying single-layer injection anomalies (such as sudden increases or decreases). Based on the anomalies in wellhead injection pressure and gas volume of each layer, structured early warning information containing anomaly location, cause inference, and handling suggestions is generated and pushed to the monitoring screen and mobile terminals in real time, thereby achieving proactive safety early warning for the entire multi-layer gas injection process with low latency and high accuracy.

[0044] In one embodiment of the present invention, a multi-layer gas injection condition intelligent early warning system based on real-time data is provided, comprising: The gas injection volume prediction module for each injection layer constructs a dynamic baseline model based on historical real-time gas injection volume data, which can be dynamically decayed or dynamically increased. The predicted gas injection volume for that layer is obtained through the dynamic baseline model. The wellhead gas injection pressure prediction module establishes a moving average model of wellhead gas injection pressure for each gas injection layer, and obtains the moving average value of pressure changing over time. The interval determination module predicts the current gas injection volume based on the dynamic benchmark model to determine the real-time flow range of each gas injection layer, and predicts the current wellhead gas injection pressure based on the pressure moving average to determine the real-time pressure range of each gas injection layer. The early warning module determines whether the real-time traffic and pressure collected at the current moment exceed the corresponding range based on the real-time traffic range and real-time pressure range. If they exceed the range and exceed the set time, an early warning is issued. Each level of the early warning process is implemented using a thread, and multi-threading is used to achieve multi-level synchronous early warning.

[0045] In the above embodiments, a dynamic baseline model with corresponding dynamic decay or dynamic increase is constructed based on historical real-time gas injection data. The predicted gas injection volume for this layer is obtained through the dynamic baseline model, specifically as follows: Based on historical real-time gas injection data, a dynamic benchmark model is obtained by fitting an exponentially decreasing or increasing function. The baseline for the dynamic decay or increase of gas injection volume at this layer, obtained through the dynamic baseline model, is as follows:

[0046] In the formula, This represents the predicted future gas injection volume for the stratum. This is the initial gas injection volume; It represents either a decreasing rate or an increasing rate.

[0047] In the above embodiments, the current gas injection volume is predicted based on a dynamic benchmark model to determine the real-time flow range for each gas injection layer: .

[0048] In the above embodiments, a moving average model of the wellhead gas injection pressure is established for each gas injection layer to obtain the moving average pressure value over time, specifically: Set a length of The sliding window is used to add new pressure data points to the current sliding window as soon as they are obtained, while removing the oldest data point in the current sliding window. The moving average model continuously calculates the average value of the data within the current sliding window, which serves as both the current trend value and the predicted value for the next moment.

[0049] In the above embodiments, the average value of the data within the sliding window is always calculated using the moving average model:

[0050] In the formula, It is a point in time. The moving average, also The predicted value at any given time; It is a point in time. The actual pressure value; It is the size of the sliding window.

[0051] In the above embodiments, the current wellhead gas injection pressure is predicted based on the moving average pressure value to determine the real-time pressure range for each injection layer: .

[0052] In the above embodiments, it is determined whether the real-time traffic and real-time pressure exceed the corresponding range based on the real-time traffic range and real-time pressure range, respectively. If they exceed the range and exceed the set time, an early warning is issued. Each level of the early warning process is implemented using one thread, and multi-threading is used to achieve multi-level synchronous early warning, specifically as follows: One thread per layer, starting with the layer gas injection volume and wellhead gas injection pressure. The total wellhead injection pressure is predicted in the future wellhead gas injection pressure using a moving average model to identify whether the injection pressure of a single layer is abnormal. At the same time, the stratified gas volume prediction is simultaneously fitted with an increasing or decreasing function based on the respective historical gas injection volume data to predict the future gas injection volume of each stratum, thereby identifying whether the gas injection volume of a single stratum is abnormal. Based on abnormal wellhead gas injection pressure and abnormal gas injection volume at each layer, structured early warning information is generated, which includes anomaly location, cause inference and handling suggestions, and is pushed to remote monitoring screens and mobile terminals in real time.

[0053] The system provided in this embodiment is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0054] In this example, firstly, single-well formation parameters (including top and bottom depths, effective thickness, gas saturation, etc.) are configured. Then, upper and lower boundaries and durations for abnormal parameters are set. Next, based on the historical real-time injection data of each formation (automatically collected and saved to the database by monitoring instruments), the current injection flow rate and wellhead pressure of each formation are calculated and predicted. Then, the data is compared with the current real-time data. If an anomaly is found, the start time of the anomaly is recorded. If the anomaly persists and the duration exceeds the set duration, an abnormal operating condition alarm is triggered. The specific steps are as follows: (1) Configure gas injection well formation information, such as Figure 3 As shown.

[0055] (2) Set the injection volume warning range, such as Figure 4 As shown.

[0056] (3) Collect real-time data at each level, such as Figure 5 As shown.

[0057] (4) Warning results at different times: After setting, simulation calculation is performed. The warning result of 1 at a certain time is as follows: Figure 6 As shown; the warning result for 2 at a certain moment is as follows: Figure 7 As shown.

[0058] In one embodiment of the present invention, a computing device is provided, which can be a terminal and may include: a processor, a communication interface, memory, a display screen, and an input device. The processor, communication interface, and memory communicate with each other via a communication bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs, which, when executed by the processor, implement the methods described in the above embodiments. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, a management network, NFC (Near Field Communication), or other technologies. The display screen can be a liquid crystal display or an e-ink display. The input device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the casing of the computing device, or an external keyboard, touchpad, or mouse. The processor can call logical instructions stored in the memory.

[0059] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0060] In one embodiment of the present invention, a computer program product is provided, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, and when the program instructions are executed by a computer, the computer is able to perform the methods provided in the above-described method embodiments.

[0061] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided, which stores server instructions that cause a computer to perform the methods provided in the above embodiments.

[0062] The computer-readable storage medium provided in the above embodiments has a similar implementation principle and technical effect to the above method embodiments, and will not be described again here.

[0063] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0064] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0065] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-layer gas injection condition intelligent early warning method based on real-time data, characterized in that, include: For each gas injection layer, a dynamic baseline model with dynamic decay or dynamic increase is constructed based on historical real-time gas injection data. The predicted gas injection volume for that layer is obtained through the dynamic baseline model. For each gas injection layer, a moving average model of the wellhead gas injection pressure is established to obtain the moving average value of the pressure over time. The current gas injection volume is predicted based on the dynamic benchmark model to determine the real-time flow range of each gas injection layer. The current wellhead injection pressure is predicted based on the moving average pressure value to determine the real-time pressure range for each injection layer. Based on the real-time traffic range and the real-time pressure range, determine whether the real-time traffic and pressure collected at the current moment exceed the corresponding range. If they exceed the range and exceed the set time, an early warning will be issued. Each level of the early warning process is implemented using a single thread, and multi-threading is used to achieve multi-level synchronous early warning.

2. The intelligent early warning method for multi-layer gas injection conditions based on real-time data as described in claim 1, characterized in that, Based on historical real-time gas injection data, a dynamic baseline model is constructed to predict the gas injection volume for that layer, either dynamically decreasing or dynamically increasing. Based on historical real-time gas injection data, a dynamic benchmark model is obtained by fitting an exponentially decreasing or increasing function. The baseline for the dynamic decay or increase of gas injection volume at this layer, obtained through the dynamic baseline model, is as follows: In the formula, This represents the predicted future gas injection volume for the stratum. This is the initial gas injection volume; It represents either a decreasing rate or an increasing rate.

3. The intelligent early warning method for multi-layer gas injection conditions based on real-time data as described in claim 2, characterized in that, Based on the dynamic baseline model, the current gas injection volume is predicted to determine the real-time flow range for each gas injection layer: .

4. The intelligent early warning method for multi-layer gas injection conditions based on real-time data as described in claim 1, characterized in that, For each injection layer, a moving average model of the wellhead injection pressure is established to obtain the moving average pressure over time, specifically: Set a length of The sliding window is used to add new pressure data points to the current sliding window as soon as they are obtained, while removing the oldest data point in the current sliding window. The moving average model continuously calculates the average value of the data within the current sliding window, which serves as both the current trend value and the predicted value for the next moment.

5. The intelligent early warning method for multi-layer gas injection conditions based on real-time data as described in claim 4, characterized in that, The moving average model consistently calculates the average value of the data within the sliding window as follows: In the formula, It is a point in time. The moving average, also The predicted value at any given time; It is a point in time. The actual pressure value; It is the size of the sliding window.

6. The intelligent early warning method for multi-layer gas injection conditions based on real-time data as described in claim 4, characterized in that, The current wellhead injection pressure is predicted based on the moving average pressure value to determine the real-time pressure range for each injection layer: .

7. The intelligent early warning method for multi-layer gas injection conditions based on real-time data as described in claim 1, characterized in that, Based on the real-time traffic range and real-time pressure range, it is determined whether the real-time traffic and pressure exceed the corresponding range. If they exceed the range and exceed the set time, an alert is issued. Each level of the alert process is implemented using a single thread, and multi-threading is used to achieve multi-level synchronous alerts. Specifically: One thread per layer, starting with the layer gas injection volume and wellhead gas injection pressure. The total wellhead injection pressure is predicted in the future wellhead gas injection pressure using a moving average model to identify whether the injection pressure of a single layer is abnormal. At the same time, the stratified gas volume prediction is simultaneously fitted with an increasing or decreasing function based on the respective historical gas injection volume data to predict the future gas injection volume of each stratum, thereby identifying whether the gas injection volume of a single stratum is abnormal. Based on abnormal wellhead gas injection pressure and abnormal gas injection volume at each layer, structured early warning information is generated, which includes anomaly location, cause inference and handling suggestions, and is pushed to remote monitoring screens and mobile terminals in real time.

8. A multi-layer gas injection condition intelligent early warning system based on real-time data, characterized in that, include: The gas injection volume prediction module for each injection layer constructs a dynamic baseline model based on historical real-time gas injection volume data, which can be dynamically decayed or dynamically increased. The predicted gas injection volume for that layer is obtained through the dynamic baseline model. The wellhead gas injection pressure prediction module establishes a moving average model of wellhead gas injection pressure for each gas injection layer, and obtains the moving average value of pressure changing over time. The interval determination module predicts the current gas injection volume based on the dynamic benchmark model to determine the real-time flow range of each gas injection layer, and predicts the current wellhead gas injection pressure based on the pressure moving average to determine the real-time pressure range of each gas injection layer. The early warning module determines whether the real-time traffic and pressure collected at the current moment exceed the corresponding range based on the real-time traffic range and real-time pressure range. If they exceed the range and exceed the set time, an early warning is issued. Each level of the early warning process is implemented using a thread, and multi-threading is used to achieve multi-level synchronous early warning.

9. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any of the methods described in claims 1 to 7.

10. A computing device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing any of the methods described in claims 1 to 7.

Images