A method for identifying wellbore microleaks using distributed temperature gradient sensor data

Abstract

The present application relates to the technical field of oil and gas geothermal exploitation, and discloses a method for identifying wellbore micro leakage by using distributed temperature gradient sensing data, which comprises the following steps: S1: optical fiber laying; laying optical fiber sensors along the outer wall of the casing outside the tubing in the whole length of the wellbore; S2: data acquisition; S3: data processing; extracting low-frequency thermal strain data from DAS, processing temperature gradient by LF-DAS data, and processing temperature gradient by absolute temperature data of DTS data to obtain distributed temperature gradient data; and performing visual processing on the DTGS data to generate a heat map of temperature distribution and strain change; S4: leakage identification; performing threshold judgment on DTGS and DTS, establishing a leakage identification model, and combining multi-dimensional signal features to predict the state and development trend of leakage; and performing early identification and early warning of micro leakage. The present application uses DTGS data to identify wellbore micro leakage, and significantly improves the detection accuracy of temperature gradient change and the identification ability of low flow rate, small flow wellbore micro leakage.

Description

Technical Field

[0001] The present invention relates to the technical field of geothermal exploitation of oil and natural gas, and in particular to a method for identifying micro-leakage in a wellbore by utilizing distributed temperature gradient sensing data. Background Art

[0002] The growing demand for real-time monitoring of wellbore integrity in the oil, gas, and geothermal industries places higher demands on the real-time detection and precise location of wellbore leaks. Traditional leak detection methods, such as point sensors, noise logging tools (NLTs), and production logging tools (PLTs), are effective under certain conditions. However, they generally suffer from limitations such as high deployment costs, poor real-time performance, limited coverage, and difficulty or low accuracy in identifying micro-leaks. These limitations make it difficult to meet the sophisticated monitoring needs of modern, complex downhole environments. In recent years, the development of distributed fiber optic sensing (DFOS) technology has provided a new solution for wellbore micro-leak detection.

[0003] The existing literature Sidenko, E., Tertyshnikov, K., Lebedev, M. et al. 2022. Experimental Study of Temperature Change Effect on Distributed Acoustic Sensing Continuous Measurements. Geophysics 87(3): D111–D122. provides applications of distributed temperature sensing (DTS) and distributed acoustic sensing (DAS) in leak detection, production profile monitoring, and downhole fluid dynamics research. DTS can generally effectively capture large temperature anomalies of the order of 0.1°C, but its ability to identify low-flow, low-flow wellbore micro-leakages of the order of 0.01°C is still insufficient. Therefore, the ability to identify wellbore micro-leakage based on temperature data still needs to be further improved. Summary of the Invention

[0004] This invention addresses the shortcomings of existing wellbore microleak detection technologies, which suffer from insufficient accuracy, limited real-time performance, and poor adaptability to complex wellbore conditions. By proposing a method for identifying wellbore microleakages using distributed temperature gradient sensing data, this method extracts DTGS data from LF-DAS data and uses the processed DTGS data for high-precision identification of wellbore microleakages. This method significantly improves the accuracy of detecting temperature gradient changes and the ability to identify low-flow, low-volume microleakages. The method processes the collected DAS raw data using LF-DAS and DTGS to calculate temperature gradient changes. Combined with DTS absolute temperature analysis, this method accurately identifies the location, fluid velocity, and trend of wellbore microleakages. Furthermore, by leveraging the long-range coverage of fiber optic sensing and the advantages of multidimensional data analysis, this method enables real-time monitoring, precise identification and location, and early warning of leaks in complex downhole environments. This solution forms a complete technical implementation path, from fiber deployment to signal acquisition and processing, and then to multidimensional feature fusion and leak detection model construction. Combined with multi-parameter dynamic monitoring, it provides comprehensive support for wellbore integrity management.

[0005] The present invention provides the following technical solutions:

[0006] The present invention provides a method for identifying wellbore micro-leakage using distributed temperature gradient sensing data, which comprises the following steps:

[0007] S1: Fiber optic deployment

[0008] Fiber optic sensors are deployed along the entire length of the outer wall of the casing outside the oil pipe in the wellbore. The fiber optic sensors include single-mode optical fibers and multi-mode optical fibers. The single-mode optical fibers are used to collect DAS data, and the multi-mode optical fibers are used to collect DTS data. The ends of the single-mode optical fibers and the multi-mode optical fibers are connected to DAS demodulators and DTS demodulators respectively.

[0009] S2: Data Collection

[0010] The DTS demodulator is used to obtain temperature distribution data along the optical fiber path, namely the DTS data; the DAS demodulator is used to obtain dynamic strain and broadband vibration signal data, namely the DAS data;

[0011] S3: Data Processing

[0012] Extracting low-frequency thermal strain data, i.e., LF-DAS data, from the DAS, performing temperature gradient processing on the LF-DAS data and the absolute temperature data of the DTS data to obtain distributed temperature gradient data, i.e., DTGS data; and visualizing the DTGS data to generate a thermal map of temperature distribution and strain changes, which is then stored and backed up;

[0013] S4: Leakage Identification

[0014] Perform DTGS and DTS threshold judgments, establish a leak identification model based on the DTGS data extracted by LF-DAS, and combine multi-dimensional signal features to predict the leakage status and development trend; perform early identification and warning of micro leaks.

[0015] According to some embodiments, step S3 comprises the following steps:

[0016] S31: LF-DAS data extraction, including the following processing steps:

[0017] S311: Low-pass filtering: Apply a 0-0.5 Hz low-pass filter to remove high-frequency noise and retain low-frequency signals related to thermal strain. After 0-0.5 Hz low-pass filtering, the original DAS data is extracted as LF-DAS signals, and the plume-like signals on its two-dimensional image represent the fluid flow velocity.

[0018] S312: Downsampling: reducing the sampling rate of the original signal to 1 Hz;

[0019] S313: Median filtering: eliminating possible impulse noise;

[0020] S314: DC drift correction: Perform DC drift correction on the low-frequency components in the signal to eliminate signal offset caused by device noise;

[0021] S32: Distributed Temperature Gradient Data Extraction

[0022] The temperature gradient of the absolute temperature data in the DTS data is calculated to obtain the DTGS data. Assuming that the temperature T is a function of time t and position L, the rate of change of temperature with time is the time gradient of the temperature. The temperature gradient is calculated in a discrete form. The discrete time gradient formula is as follows:

[0023]

[0024] In the above formula, T is the temperature change, t 1、 t2 represents different time, Δt is the time change, Tt2 is the temperature at time t2, and Tt1 is the temperature at time t1;

[0025] The temperature gradient of LF-DAS data is calculated to obtain DTGS data to accurately characterize the temperature gradient change, thereby achieving early detection of small leaks or temperature fluctuations. The LF-DAS data is regarded as DTS data and the DTGS data is extracted using formula (1).

[0026] According to some embodiments, the following steps are further included:

[0027] S33: Data visualization and storage: Visualize the DTGS data and DTS data and draw a time series graph to reflect the dynamic process of leakage development in detail; and store and back up the DTGS data and DTS data.

[0028] According to some embodiments, step S4 specifically includes the following steps:

[0029] S41: DTGS and DTS threshold judgment

[0030] Based on the DTGS data extracted from the LF-DAS data, a leakage identification model is established, which specifically includes position identification and state prediction, that is, locating the leakage point x through the spatial distribution of the DTGS and DTS data signal peaks. leak , and combined with multi-dimensional signal characteristics, including leakage location, fluid velocity and absolute temperature change, to predict the leakage status and development trend. The leakage identification model is shown in the following formula:

[0031]

[0032] In the above formula, P leak That is, the peak amplitude of the leakage point, x leak The maximum value among the peak values ​​is taken;

[0033] S42: Early identification and warning of micro-leakage in wellbore

[0034] Set the warning threshold P threshold When the fusion signal strength exceeds the warning threshold, the recognition model is used to identify the wellbore micro-leakage phenomenon and location, and trigger an early warning. The recognition model is shown in the following formula:

[0035]

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] This invention proposes a method for identifying wellbore micro-leaks using distributed temperature gradient sensing data. This method utilizes DTGS data to identify wellbore micro-leaks, significantly improving the accuracy of detecting temperature gradient changes and the ability to identify micro-leaks in low-flow and low-flow wellbores. This method overcomes the limitations of traditional wellbore leak detection methods in terms of real-time performance, accuracy, and adaptability to complex well conditions, significantly improving the reliability, stability, and cost-effectiveness of monitoring and providing an innovative solution for wellbore integrity management. This is specifically reflected in the following aspects:

[0038] (1) High-precision leak detection

[0039] The present invention accurately identifies micro-leakages in wellbores through DTS absolute temperature data and DTGS data extracted from LF-DAS data. The temperature gradient calculation and analysis of DTGS significantly improves the spatial and temporal resolution of leak detection, and the accuracy of leak identification in complex well conditions is higher. The present invention uses a method for DTGS processing based on LF-DAS to extract low-frequency strain characteristics caused by temperature changes and use them to calculate the dynamic changes of temperature gradients, thereby achieving accurate identification and dynamic monitoring of micro-leakages in wellbores; an automatic leakage identification method based on DTS absolute temperature data and DTGS data extracted from LF-DAS data is established, which significantly improves the accuracy and reliability of identifying the location, fluid velocity and development trend of micro-leakages in wellbores, realizes real-time monitoring and early warning of micro-leakage phenomena, and provides efficient and reliable technical support for wellbore integrity management.

[0040] (2) Efficiency and economy

[0041] Compared with traditional point sensors and production logging tools (such as PLT and NLT), the present invention adopts distributed fiber optic sensing technology, which can provide continuous monitoring information along the entire fiber path without the need for additional hardware deployment, greatly reducing system construction and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A flow chart of a method for identifying wellbore micro-leakage using distributed temperature gradient sensing data provided by an embodiment of the present invention.

[0043] Figure 2 A waterfall chart comparison of DTGS data (a) extracted from LF-DAS and DTGS data (b) extracted from DTS provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] To address the problems of insufficient accuracy, poor real-time performance, and poor adaptability to complex well conditions in existing wellbore microleak detection technologies, the present invention proposes a method for identifying wellbore microleakages using distributed temperature gradient sensing (DTGS) data. Optical fiber can be laid along the entire length of the wellbore. First, low-frequency distributed acoustic sensing (LF-DAS) data is extracted from the original DAS data as temperature (thermal strain) data. DTGS data is then extracted from the LF-DAS data. Finally, threshold judgment is performed by combining DTGS and DTS data to perform real-time monitoring, accurate identification and positioning, and early warning of microleakages in the wellbore.

[0045] The present invention is described in detail below with reference to the embodiments and accompanying drawings. However, it should be understood that the embodiments and accompanying drawings are merely exemplary descriptions of the present invention and do not constitute any limitation on the scope of protection of the present invention. All reasonable variations and combinations within the scope of the inventive concept of the present invention fall within the scope of protection of the present invention.

[0046] Example 1

[0047] like Figure 1 This embodiment provides a method for identifying wellbore micro-leakage using distributed temperature gradient sensing data, which includes the following steps:

[0048] S1: Fiber optic deployment

[0049] Fiber optic sensors are deployed along the entire length of the wellbore to ensure that all potential leak areas are monitored. In this embodiment, the fiber optic sensors used for data acquisition are coupled to the outer wall of the casing outside the oil pipe. The fiber optic sensors include single-mode and multimode fibers. The single-mode fiber is used to collect DAS data, while the multimode fiber is used to collect DTS data. The ends of the single-mode and multimode fibers are connected to the DAS and DTS demodulators, respectively. The collected data is processed in the next step.

[0050] Fiber optic sensors are responsible for sensing, while the DAS and DTS interrogators primarily perform the following tasks: First, using DTS technology to record temperature distribution data along the fiber path, providing essential information for DTS-based DTGS processing; second, using DAS technology to collect dynamic strain and broadband vibration signals, extracting LF-DAS low-frequency thermal strain data from the raw DAS data and converting it into temperature change data, providing critical data support for LF-DAS-based DTGS processing. The fiber optic layout must ensure continuous and high-resolution data acquisition to accommodate long-term stable operation under complex wellbore conditions, thus meeting the requirements for high-precision leak detection. LF-DAS, with its high sensitivity to low-frequency signals and its ability to resist interference, can more accurately capture minute temperature fluctuations caused by micro-leaks. Combined with DTGS data, LF-DAS enables real-time monitoring and early warning of micro-leak location, fluid velocity, and development trends in the wellbore. Compared to traditional methods, it exhibits greater sensitivity and adaptability in complex downhole environments, providing an innovative and efficient solution for wellbore integrity management and oil and gas production optimization.

[0051] S2: Data Collection

[0052] Data acquisition is the foundation of this invention. It uses DTS and DAS demodulators to acquire monitoring data, achieve high-frequency acquisition of multi-dimensional signals, and provide basic support for subsequent DTGS processing. It includes the following steps:

[0053] S21: DAS data acquisition

[0054] DAS operates based on the Rayleigh scattering effect of optical fibers. By recording acoustic signals (dynamic strain), it reflects the vibration and strain changes caused by wellbore leaks. The DAS system utilizes a high sampling rate of 10kHz, enabling it to capture subtle strain changes along the optical fiber path while ensuring sufficient temporal resolution to meet the requirements of dynamic monitoring. The collected raw data, after LF-DAS processing, primarily includes vibration signals (frequency below 0.5Hz) and dynamic strain changes, providing critical data support for real-time leak monitoring and analysis.

[0055] S22: DTS data collection

[0056] The DTS system monitors the temperature distribution along the fiber path, capturing changes in temperature gradients in real time and identifying abnormal temperature behavior in leak areas. Its core principle is based on the scattering properties of Stokes and anti-Stokes light in optical fibers. The intensity of these light signals is directly correlated with temperature changes along the fiber path. By monitoring and analyzing these signals, the temperature distribution at every point along the fiber path can be determined.

[0057] To meet the needs of temperature monitoring in different scenarios, the DTS system uses an adjustable sampling rate to ensure data accuracy while achieving long-term stable temperature recording. High sampling rates are suitable for real-time monitoring of dynamic temperature changes, while lower sampling rates are suitable for long-term trend tracking and static environmental monitoring.

[0058] S23: Data storage and transmission

[0059] All collected data is transmitted to the ground control center via a real-time data transmission system, ensuring data integrity and real-time transmission. Once transmitted, the data is stored on high-capacity storage devices capable of supporting the large amounts of data generated during long-term, continuous collection and featuring high reliability to ensure secure storage. The system also incorporates a comprehensive data backup mechanism, effectively preventing data loss through multiple backup methods and ensuring smooth subsequent analysis. This comprehensive data management process provides reliable data support for real-time monitoring and subsequent analysis.

[0060] S3: Data Processing

[0061] Data processing is one of the core steps in achieving high-precision monitoring of wellbore micro-leakages in this invention. This step pre-processes the collected raw data, extracts features, and integrates them into multiple dimensions, ultimately extracting key information related to the leakage. Specifically, it includes the following steps:

[0062] S31: LF-DAS data extraction, including the following processing steps:

[0063] S311: Low-pass filtering: Apply a 0-0.5 Hz low-pass filter to remove high-frequency noise while retaining low-frequency signals related to thermal strain. After 0-0.5 Hz low-pass filtering, the raw DAS data is extracted as LF-DAS signals, and the plume-like signals on the 2D image represent the fluid flow velocity.

[0064] S312: Downsampling: reducing the sampling rate of the original signal to 1 Hz to improve the efficiency of subsequent data analysis while reducing the pressure of data storage and calculation;

[0065] S313: Median filtering: eliminates possible impulse noise to ensure the smoothness and stability of the data;

[0066] S314: DC drift correction: Perform DC drift correction on the low-frequency components in the signal to eliminate signal offset caused by device noise;

[0067] S32: Distributed Temperature Gradient (DTGS) Data Extraction

[0068] Data processing based on LF-DAS and DTS can further extract temperature gradient information. The core of DTGS technology is to infer the temperature gradient by analyzing the time derivative of optical fiber data in the time dimension.

[0069] DTS data records a complete, nearly continuous temperature curve in space and time. Therefore, the temperature gradient can be calculated for the DTS absolute temperature data to obtain DTGS data. Assuming that temperature T is a function of time t and location L, in practical applications, the rate of change of temperature over time, i.e., the time gradient of temperature, is usually of interest. In practical applications, the calculation of temperature gradient is usually done in discrete form. The discrete time gradient formula is as follows:

[0070]

[0071] In the above formula, T is the temperature change, t 1、 t2 represents a different time, Δt is the time change, Tt2 is the temperature at time t2, and Tt1 is the temperature at time t1.

[0072] LF-DAS can indirectly reflect relative temperature changes in the near-wellbore area along the optical fiber by detecting thermal strain signals related to temperature changes in low-frequency acoustic signals. By calculating the temperature gradient of LF-DAS, DTGS data can also be obtained to accurately characterize temperature gradient changes, thereby enabling early detection of small leaks or temperature fluctuations. Therefore, LF-DAS data can be treated as DTS data and DTGS data can be extracted using formula (1). The DTGS of DTS can resolve the absolute value of temperature, while the DTGS of DAS can detect temperature changes, with higher accuracy than DTS.

[0073] Compared with the existing DTS-based DTGS technology, the data processing process in step S3 provided in this embodiment has the following technical advantages:

[0074] ① Significantly improved detection sensitivity. LF-DAS technology demonstrates excellent sensitivity in detecting temperature gradient changes and can effectively capture minute temperature changes in complex downhole environments.

[0075] ② The temperature gradient resolution is greatly improved. The temperature gradient resolution of DTGS obtained using LF-DAS technology can reach 0.01°C / s, which is an order of magnitude higher than the 0.1°C / s of traditional DTS-DTGS.

[0076] ③ Furthermore, when using a DAS demodulator with optimized low-frequency response signal-to-noise ratio, the temperature gradient resolution of the obtained DTGS data can reach ±0.001°C / s, which can realize the refined analysis of temperature anomaly points and improve the reliability of anomaly detection.

[0077] The above technical solution not only overcomes the technical problem of insufficient sensitivity of DTS temperature gradient detection in the existing technology, but also realizes high-precision detection and characterization of tiny temperature changes in complex downhole environments through the low-frequency signal capture capability of LF-DAS technology, which has significant technical advantages.

[0078] S33: Data Visualization and Storage

[0079] After data processing, the resulting DTGS and DTS data are visualized using data visualization techniques to generate heat maps of temperature distribution and strain changes, visually demonstrating the leak location. Furthermore, time series graphs are created to detail the dynamics of the leak's development, providing a visual reference for further analysis. The processed data is stored on high-capacity storage devices to ensure long-term preservation and availability. A real-time backup mechanism is implemented to prevent data loss and ensure smooth subsequent analysis.

[0080] S4: Leak identification, which includes the following steps:

[0081] S41: DTGS and DTS threshold judgment

[0082] Based on the DTGS data extracted by LF-DAS, a leakage identification model is established, which specifically includes location identification and status prediction, that is, locating the leakage point x through the spatial distribution of the DTGS and DTS data signal peaks. leak , and combined with multi-dimensional signal characteristics (such as leak location, fluid velocity, and absolute temperature change) to predict the leak status and development trend. This model, shown below, can be optimized using machine learning algorithms to further improve detection accuracy.

[0083]

[0084] In the above formula, P leak That is, the peak amplitude of the leakage point, x leak The maximum value among the peak values ​​is taken.

[0085] S42: Early identification and warning of micro-leakage in wellbore

[0086] The system monitors the behavior and location changes of micro-leakages in the wellbore in real time through continuous dynamic analysis to ensure timely response.

[0087] Set the warning threshold P threshold When the fusion signal strength exceeds the set range, the wellbore micro-leakage phenomenon and location are automatically identified and early warning is triggered in time, as shown in the following formula:

[0088]

[0089] When the amplitude intensity of the monitored leakage point exceeds the preset value, a warning will be issued.

[0090] After the early warning is triggered, the system automatically generates a leakage report and provides decision support to managers.

[0091] Example 2

[0092] Through experiments and practical applications, the present invention verifies the accuracy and reliability of extracting DTGS data from LF-DAS low-frequency thermal strain data in identifying wellbore micro-leakage.

[0093] During gas production, the B annulus of a gas well was pressurized. Although pressure fluctuations were minimal, a micro-leakage in the wellbore was suspected. Using a permanently installed optical fiber deployed outside the well casing and connected to DAS and DTS demodulators, high-quality DAS and DTS data were collected during a gas injection test in the A annulus. According to the technical solution proposed in this invention, the DAS and DTS data were processed. LF-DAS low-frequency thermal strain (temperature) data was first extracted from the raw DAS data. High-precision and high-temperature-resolution DTGS data were then extracted from the LF-DAS data. Furthermore, DTGS data was extracted from the raw DTS temperature data.

[0094] Figure 2 (a) shows the waterfall plot of DTGS data extracted from LF-DAS. Figure 2 (b) shows the waterfall chart of DTGS data extracted from DTS, both using the same temperature color scale. It can be seen that:

[0095] (1) After 20:12:00, both the DTGS data extracted from LF-DAS and the DTGS data extracted from DTS showed a slight temperature change of only 0.65°C in the depth range of 0-100 m. This temperature change was caused by the temperature difference caused by the gas injected from annulus A entering annulus B, indicating that both can identify micro-leakage in the wellbore.

[0096] (2) The temperature changes in the waterfall chart of the DTGS data extracted from LF-DAS are relatively fine, and the resolution of temperature changes in the depth and time dimensions is on the order of 0.01°C; while the waterfall chart of the DTGS data extracted from DTS is relatively smooth and lacks rich details, indicating that its temperature change resolution is only maintained at the order of 0.1°C; therefore, compared with the DTGS data extracted from DTS, the temperature resolution of the DTGS data extracted from LF-DAS is higher and more suitable for identifying small temperature changes caused by micro leaks.

[0097] The above embodiments are merely preferred embodiments of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions that fall within the scope of protection of the present invention are within the scope of protection of the present invention. It should be noted that improvements and modifications that can be made by a person skilled in the art without departing from the principles of the present invention are also considered to be within the scope of protection of the present invention.

Claims

1. A method for identifying wellbore micro-leakage using distributed temperature gradient sensing data, characterized by: The following steps are involved: S1: Fiber optic deployment Fiber optic sensors are deployed along the entire length of the outer wall of the casing outside the oil pipe in the wellbore. The fiber optic sensors include single-mode optical fibers and multi-mode optical fibers. The single-mode optical fibers are used to collect DAS data, and the multi-mode optical fibers are used to collect DTS data. The ends of the single-mode optical fibers and the multi-mode optical fibers are connected to DAS demodulators and DTS demodulators respectively. S2: Data Collection The DTS demodulator is used to obtain temperature distribution data along the optical fiber path, namely the DTS data; the DAS demodulator is used to obtain dynamic strain and broadband vibration signal data, namely the DAS data; S3: Data Processing Extract low-frequency thermal strain data from the DAS, i.e., LF-DAS data, and process the temperature gradient using the LF-DAS data. and The absolute temperature data of the DTS data is subjected to temperature gradient processing to obtain distributed temperature gradient data, i.e., DTGS data; and the DTGS data is visualized to generate a thermal map of temperature distribution and strain change, and stored for backup; S4: Leakage Identification Perform DTGS and DTS threshold judgments, establish a leak identification model based on DTGS data extracted by LF-DAS, and combine multi-dimensional signal features to predict the status and development trend of leaks; perform early identification and early warning of micro leaks; Step S3 includes the following steps: S31: LF-DAS data extraction, including the following processing steps: S311: Low-pass filtering: Apply a 0-0.5 Hz low-pass filter to remove high-frequency noise and retain low-frequency signals related to thermal strain. After 0-0.5 Hz low-pass filtering, the original DAS data is extracted as LF-DAS signals, and the plume-like signals on its two-dimensional image represent the fluid flow velocity. S312: Downsampling: reducing the sampling rate of the original signal to 1 Hz; S313: Median filtering: eliminating possible impulse noise; S314: DC drift correction: Perform DC drift correction on the low-frequency components in the signal to eliminate signal offset caused by device noise; S32: Distributed Temperature Gradient Data Extraction The temperature gradient of the absolute temperature data in the DTS data is calculated to obtain the DTGS data. Assuming that the temperature T is a function of time t and position L, the rate of change of temperature with time is the time gradient of the temperature. The temperature gradient is calculated in a discrete form. The discrete time gradient formula is as follows: （1） In the above formula, T is the temperature change, t 1、 t2 represents different time, Δt is the time change, Tt2 is the temperature at time t2, and Tt1 is the temperature at time t1; The temperature gradient of LF-DAS data is calculated to obtain DTGS data to accurately characterize the temperature gradient change, thereby achieving early detection of small leaks or temperature fluctuations. The LF-DAS data is regarded as DTS data and the DTGS data is extracted using formula (1).

2. The method for identifying wellbore micro-leakage using distributed temperature gradient sensing data according to claim 1, characterized in that: The following steps are also included: S33: Data visualization and storage: Visualize the DTGS data and DTS data and draw a time series graph to reflect the dynamic process of leakage development in detail; and store and back up the DTGS data and DTS data.

3. The method for identifying wellbore micro-leakage using distributed temperature gradient sensing data according to claim 1, characterized in that: Step S4 specifically includes the following steps: S41: DTGS and DTS threshold judgment Based on the DTGS data extracted from the LF-DAS data, a leak identification model is established, specifically including location identification and status prediction, that is, locating the leak point through the spatial distribution of the DTGS and DTS data signal peaks , and combined with multi-dimensional signal characteristics, including leakage location, fluid velocity and absolute temperature change, to predict the leakage status and development trend. The leakage identification model is shown in the following formula: （2） In the above formula, P leak That is, the peak amplitude of the leakage point, x leak The maximum value among the peak values ​​is taken; S42: Early identification and warning of micro-leakage in wellbore Set the warning threshold P threshold When the fusion signal strength exceeds the warning threshold, the recognition model is used to identify the wellbore micro-leakage phenomenon and location, and trigger an early warning. The recognition model is shown in the following formula: （3）。

Images