A reservoir fluid property identification method based on fracture local gas logging anomaly detection

Abstract

The application discloses a reservoir fluid property identification method based on fracture local gas logging anomaly detection, and comprises the following steps: picking up large-scale fractures and their through-hole depth positions by using logging data; calculating local relative anomalies of different light hydrocarbon components, and eliminating the influence of the matrix part by adjusting the scale range; based on the convection-diffusion mechanism of light hydrocarbon components at the fracture, a fracture fluid identification mode is constructed with the local relative anomalies of light hydrocarbon components C1 and C2 as the core; and the reservoir fluid property is determined by using the local relative anomalies of light hydrocarbon components at the large-scale fracture through hole. The application focuses on the more significant gas logging anomaly at the fracture than the reservoir matrix, avoids the weakening of the matrix logging response and the interference of the wellbore expansion / mud change, and effectively improves the reliability of the complex reservoir fluid identification.

Description

Technical Field

[0001] This invention belongs to the field of oil and gas exploration and development technology, specifically relating to a method for identifying reservoir fluid properties based on the detection of local gas logging anomalies in fractures. Background Technology

[0002] In recent years, the focus of domestic oil exploration and development has gradually shifted to complex lithology, deep and unconventional oil and gas reservoirs. These new areas and strata, such as the Mesozoic volcanic rocks and Archean metamorphic buried hills in the Bohai Sea, generally have low porosity and low permeability, complex pore structures and strong heterogeneity. In addition, the intrusion of drilling mud weakens the logging response characteristics of the reservoir fluids themselves, seriously affecting the accuracy of traditional logging fluid identification methods. This brings significant uncertainty to the determination of oil, gas and water interfaces, the determination of effective thickness, the calculation of reliable reserves, development plans and fracturing designs, and greatly increases the risks and costs of exploration and development decisions.

[0003] Domestic and international scholars have primarily developed two technical systems for well logging fluid identification in complex reservoirs: multi-parameter fusion analysis and artificial intelligence methods. The former uses sensitivity analysis of the target formation fluid response to screen key parameters (such as light hydrocarbon composition, resistivity, and sonic transit time) and construct a multi-dimensional comprehensive identification chart to improve accuracy. However, the low porosity and permeability of the reservoir, complex pore structure, and mud intrusion effects weaken fluid response characteristics, causing conventional methods such as the hydrocarbon moisture ratio / equilibrium ratio envelope method to fail. Furthermore, engineering interferences such as overpressure well conditions, mud system changes, and single-channel / after-effect gas introduce non-fluid factors into the raw data, significantly increasing identification uncertainty. The latter relies on cluster analysis and other techniques to establish a feature library of oil, gas, and water layer samples for intelligent identification. However, the strong heterogeneity of complex reservoirs leads to significant differences in well logging responses between single wells and between wells. Coupled with sample scarcity and insufficient generalization ability of the fluid identification model, it also faces a bottleneck in identification reliability. Summary of the Invention

[0004] This invention is proposed to address the problem of weakened logging fluid response in complex lithological reservoirs due to low porosity and permeability characteristics and mud invasion in existing technologies. Its purpose is to provide a method for identifying reservoir fluid properties based on the detection of local gas logging anomalies in fractures.

[0005] This invention is achieved through the following technical solution: A method for identifying reservoir fluid properties based on the detection of local gas logging anomalies in fractures includes the following steps: S1. Use well logging data to identify large-scale fractures and their depths through the well. S2. Calculate the local relative anomaly curves of different light hydrocarbon components in the well logging; S3. Adjust the scale range of the local relative anomaly curve of light hydrocarbon components that change with depth, and establish a fracture fluid identification mode. S4. Determine reservoir fluid properties by utilizing the local relative anomalies of light hydrocarbon components at the wellhead crossing of large-scale fractures.

[0006] In the above technical solution, step S1 specifically includes the following steps: S11. Use acoustic remote logging to obtain well perimeter reflected wave imaging results, and extract the fracture trajectory from the reflected wave imaging results. S12. Calculate the crack extension length; The formula for calculating the crack extension length is: In the formula: The length of the crack is expressed in meters (m). The depth of the crack initiation point, in meters; The depth of the crack termination point, in meters; The vertical distance from the fracture initiation point to the well, in meters; The vertical distance from the well to the termination point of the fracture is expressed in meters (m). S13. Based on the calculation results of step S12, large-scale fractures are picked out, and the well depth of the picked large-scale fractures is determined by combining the results of electrical imaging logging.

[0007] In the above technical solution, the standard for the large-scale crack is an extension length greater than 10m.

[0008] In the above technical solution, step S2 specifically includes the following steps: S21. Calculate the low-frequency trend curve of light hydrocarbon component content as a function of depth using a low-pass filter; The expression for the low-frequency trend curve of the light hydrocarbon component content varying with depth is: In the formula: The trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; lowfilt is a low-pass filter; f represents the content of light hydrocarbon components containing n carbon atoms at different depths z, expressed as %; z is the depth in meters (m); n is the number of carbon atoms in the light hydrocarbon component, where n is an integer less than or equal to 5; c f is the spatial cutoff frequency of the low-pass filter, expressed in 1 / m. s This represents the spatial sampling frequency of the low-pass filter, expressed in units of 1 / m. S22. The low-frequency trend of the content of light hydrocarbon components varying with depth is processed by median filtering to obtain the local trend curve of the content of light hydrocarbon components varying with depth. The expression for the local trend curve of the light hydrocarbon component content varying with depth is: In the formula: The local trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; midfilt is the median filter; The trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; w is the sliding window length of the median filter, odd number, dimensionless. S23. The relative difference between the low-frequency trend of light hydrocarbon component content with depth and the local trend of light hydrocarbon component content is used as a local relative anomaly of light hydrocarbon component with depth. The expression for the local relative anomaly curve of the light hydrocarbon component varying with depth is: Where: RC n (z) represents the local relative anomaly curves of the content of light hydrocarbon components containing n carbon atoms at different depths z, % The trend lines for the content of light hydrocarbon components containing n carbon atoms at different depths z are given by %. This represents the local trend line of the content of light hydrocarbon components containing n carbon atoms at different depths z.

[0009] In the above technical solution, step S3 specifically includes the following steps: S31. Establish the calculation formula for the content of light hydrocarbon components that diffuse into the wellbore in the reservoir matrix within time t, as shown in the following formula (5). In the formula: The content of light hydrocarbon components with n carbon atoms diffused into the wellbore within time t of the reservoir matrix, %; D n Let be the diffusion coefficient of the light hydrocarbon component with n carbon atoms in the reservoir matrix, %; c n t represents the concentration of light hydrocarbon components with n carbon atoms in the reservoir matrix, %; t represents the time length during which the light hydrocarbon components exchange with the wellbore, in seconds. S32. Establish the relationship between the content of light hydrocarbon components in the wellbore during the communication between the fracture and the wellbore within time t, as shown in the following formula (6). In the formula: The percentage is the content of light hydrocarbon component with n carbon atoms that diffuses into the wellbore within time t in the fractured portion, %; t is the time length during which the light hydrocarbon component exchanges with the wellbore, in seconds; in this application Instead of direct calculation, the value is represented by adjusting the scale. Therefore, an accurate calculation formula is not required. Formula (6) is only used as a methodological basis for discussion. in: Where: L is the crack extension length, %; μ is the crack fluid viscosity, in mPa·s; This refers to the permeability of cracks, expressed in meters (m). 2 ΔP is the pressure difference between the wellbore and the fracture, in Pa. S33. A fracture fluid identification model at the well crossing of large-scale fractures is established by using the local relative differences RC1 and RC2 between two typical light hydrocarbon components C1 and C2.

[0010] In the above technical solution, the crack fluid identification mode is specifically as follows: When gas is present, RC1 and RC2 form a positive envelope; When oil is present, RC1 and RC2 form a positive or negative envelope, and the absolute values ​​and envelope areas of RC1 and RC2 are smaller than those when gas is present. When the layer is wet or dry, RC1 and RC2 coincide and their absolute values ​​tend to be 0.

[0011] In the above technical solution, step S4 specifically involves: processing each large-scale crack f that meets the requirements picked in step S1. i Pick its depth z=d i Local relative anomaly of light hydrocarbon components at RC n Then, based on the fracture fluid identification model established in step S3, the fluid properties are identified, and the fluid properties in the reservoir above the fracture crossing point can be determined as the fluid properties in the fracture.

[0012] The beneficial effects of this invention are: This invention provides a reservoir fluid property identification method based on the detection of local gas logging anomalies in fractures. It utilizes logging methods to identify large-scale, high-permeability fractures, focusing on the more significant gas logging responses at fracture crossings compared to low-porosity, low-permeability matrices. It also quantifies the local anomaly characteristics of different light hydrocarbon components at fracture crossings and constructs fracture fluid identification patterns based on the intensity of these local anomalies, thereby guiding fluid identification in the fractured reservoir. This method utilizes the initial logging gas logging response during fracture opening, reducing mud intrusion interference. Furthermore, it replaces the analysis of original gas logging values ​​across the entire well section with local anomaly analysis of light hydrocarbon components at large-scale fracture crossings. This effectively overcomes the weakening logging response in low-porosity, low-permeability matrices and the influence of non-fluid factors such as wellbore enlargement and mud system changes. This provides a new approach and method for comprehensive logging fluid identification in complex reservoirs, offering significant guidance for determining oil-gas-water interfaces, establishing effective thicknesses, calculating reliable reserves, development plans, and fracturing design. Attached Figure Description

[0013] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a large-scale fracture pickup image obtained from a well in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the crack fluid identification mode in Embodiment 1 of the present invention; Figure 4 This is a comprehensive result diagram of reservoir fluid property identification based on local gas measurement anomaly detection in fractures, according to Embodiment 1 of the present invention.

[0014] For those skilled in the art, other related figures can be obtained from the above figures without any creative effort. Detailed Implementation

[0015] To enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0016] like Figure 1 As shown, a method for identifying reservoir fluid properties based on the detection of local gas logging anomalies in fractures includes the following steps: S1. Use well logging data to identify large-scale fractures and their depths through the well. Low-porosity and low-permeability reservoirs often form large-scale fractures or cracks under tectonic activity. In addition, unconformities between different lithologies that are not filled with solids such as clay also serve as good high-permeability channels, which are also considered as fractures. Fractures with an extension length greater than 10m are usually saturated with fluids from the reservoir in which they are located, and can form considerable fluid exchange between the wellbore and the fractures at the moment the fractures are exposed during drilling, providing favorable conditions for fluid identification. Therefore, it is necessary to obtain information on such large-scale fractures through well logging data first. Specifically, the following steps are included: S11. Use acoustic remote logging to obtain well perimeter reflected wave imaging results, and extract the fracture trajectory from the reflected wave imaging results. Wellbore reflection imaging can be obtained using acoustic long-range logging, which can currently be fully implemented using existing commercial software. The reflected wave images obtained through commercial software processing are presented in image form, reflecting the reflection information around the well as varying radial distance *r*, i.e., fracture characteristics. The distance *r* from the wellbore to each point on the fracture surface can be directly read from the reflected wave image. The trajectory of the fracture is picked up on the reflected wave image; the trajectory of any fracture can be represented by the coordinates of its starting and ending points, i.e., the starting point (z... s r s ) and termination point (z e r e ); S12. Calculate the crack extension length; The formula for calculating the crack extension length is: In the formula: The length of the crack is expressed in meters (m). The depth of the crack initiation point, in meters; The depth of the crack termination point, in meters; The vertical distance from the fracture initiation point to the well, in meters; The vertical distance from the well to the termination point of the fracture is expressed in meters (m). S13. Based on the calculation results of step S12, select the large-scale cracks and denote them as f. i Simultaneously, the large-scale fracture f picked up was determined by combining the electrical imaging logging results. i Well depth position d i f i d represents the i-th large-scale crack with an extension length greater than 10m. i The depth through the well of the i-th large-scale fracture with an extension length greater than 10m is represented in meters. The standard for large-scale cracks is an extension length greater than 10m; Figure 2 This is the fracture picking result of the target layer in well X in this embodiment. Figure 2 The left side shows the processed results of acoustic long-range detection (low-angle imaging and high-angle imaging). The dark areas in the image represent reflective features near the well. Based on the imaging results, four large-scale fractures (f1, f2, f3, f4) were identified. Within the imaging range (with the well axis as the center, a diameter of approximately 60m), each fracture extends for more than 10m. Figure 2 The electrical imaging logging on the right further confirms the well depths of the aforementioned large-scale fractures as d1=3756.0m, d2=3785.0m, d3=3856.0m and d4=3955.0m.

[0017] S2. Calculate the local relative anomaly curves of different light hydrocarbon components in the well logging; Specifically, the following steps are included: S21. Let C be the logging curve showing the content of light hydrocarbon components as a function of depth z. n (z), where n represents the carbon atoms of the light hydrocarbon component, and n is an integer less than or equal to 5. Figure 4 The fifth passage presents the C1 and C2 curves for the light hydrocarbon components (red and blue solid lines). Using a low-pass filter (here denoted as lowfilt, which can be implemented using common algorithms, not elaborated here), the low-frequency trend line TC of the light hydrocarbon component content as a function of depth z is calculated.n (z); The low-frequency trend line TC of the light hydrocarbon component content varying with depth z n The expression for (z) is: In the formula: The trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; lowfilt is a low-pass filter; f represents the content of light hydrocarbon components containing n carbon atoms at different depths z, expressed as %; z is the depth in meters (m); n is the number of carbon atoms in the light hydrocarbon component, where n is an integer less than or equal to 5; c f is the spatial cutoff frequency of the low-pass filter, expressed in 1 / m. s The spatial sampling frequency of the low-pass filter is expressed in units of 1 / m; the sampling frequency f of the low-pass filter... s The depth interval of the actual light hydrocarbon components is related to the actual sampling depth, and is usually taken as 10~20. The spatial cutoff frequency f of the low-pass filter is... c Depending on the geological structure being studied, the value of f is usually taken for fractures. c It is 1~2; TC n The local trend of light hydrocarbon component content within the window length w was characterized. Figure 4 The fifth section presents the local trend curves of the light hydrocarbon component content of light hydrocarbon components C1 and C2, TC1 and TC2 (red and blue dashed lines). S22, Low-frequency trend line TC for the content of light hydrocarbon components varying with depth z. n (z) After median filtering, the local trend line MC of the light hydrocarbon component content as a function of depth z is obtained. n (z); The local trend line MC of the light hydrocarbon component content varying with depth z n The expression for (z) is: In the formula: The local trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; midfilt is the median filter; The trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; w is the sliding window length of the median filter, odd number, dimensionless. S23. Utilizing the low-frequency trend line TC of light hydrocarbon component content varying with depth z. n (z) and the local trend line MC of light hydrocarbon component content n The relative difference RC of (z) n (z) is a local relative anomaly curve of light hydrocarbon components that vary with depth z; The expression for the local relative anomaly curve of the light hydrocarbon component varying with depth z is: Where: RC n (z) represents the local relative anomaly curves of the content of light hydrocarbon components containing n carbon atoms at different depths z, % The trend lines for the content of light hydrocarbon components containing n carbon atoms at different depths z are given by %. This represents the local trend line of the content of light hydrocarbon components containing n carbon atoms at different depths z, % . Figure 4 The sixth section presents the local relative difference curves (RC1 and RC2) of the light hydrocarbon component content of C1 and C2 (red and blue solid lines). S3. Adjust the scale range of the local relative anomaly curve of the light hydrocarbon component that varies with depth z, and establish a fracture fluid identification mode. Specifically, the following steps are included: S31. Without considering chemical reactions, after drilling through the formation, the content (or concentration) of light hydrocarbon components satisfies the convection-diffusion equation. Under drilling mud equilibrium conditions, the matrix fluid in low-porosity, low-permeability reservoirs typically cannot form effective convection, and light hydrocarbon components mainly diffuse. The content of light hydrocarbons within the formation does not change significantly, and the concentration gradient along the diffusion path from the formation to the wellbore is approximately a constant c. n The content of light hydrocarbon components that diffuse into the wellbore from the reservoir matrix within time t can be approximated as: In the formula: The content of light hydrocarbon components with n carbon atoms diffused into the wellbore within time t of the reservoir matrix, %; D n Let be the diffusion coefficient of the light hydrocarbon component with n carbon atoms in the reservoir matrix, %; c n t represents the concentration of light hydrocarbon components with n carbon atoms in the reservoir matrix, %; t represents the time length during which the light hydrocarbon components exchange with the wellbore, in seconds. S32. At the instant when the fracture connects with the wellbore, convection is formed due to the pressure difference. Convection causes the fracture and the wellbore to re-establish pressure balance, and the convection decreases until it stops. During this period, light hydrocarbon components enter the wellbore with the convection. The content of light hydrocarbon components in the wellbore within time t has an exponential relationship with time, as specifically expressed by equation (6): In the formula: The percentage is the content of light hydrocarbon component with n carbon atoms that diffuses into the wellbore within time t in the fractured portion; t is the time length during which the light hydrocarbon component exchanges with the wellbore, in seconds. in: Where: L is the crack extension length, %; μ is the crack fluid viscosity, in mPa·s; This refers to the permeability of cracks, expressed in meters (m). 2 ΔP is the pressure difference between the wellbore and the fracture, in Pa. S33. Single-channel gas is a common phenomenon in oil drilling operations. It refers to the phenomenon that when connecting a new drill pipe, the bottom hole pressure is relatively reduced due to the mud pump stopping. In addition, the suction effect generated by the drill string being pulled up causes light hydrocarbon components in the formation to seep into the well. During this process, different light hydrocarbon components enter the wellbore due to their concentration differences and diffusion rates. According to Equation (5), the content of each light hydrocarbon component maintains an approximately fixed ratio. By adjusting the local relative anomaly curve RC of the light hydrocarbon component content... n The scale range of (z) is made to coincide at the depth of single gas generation to correct for the diffusion effect of matrix light hydrocarbon components (e.g. Figure 3 As shown), after the scale range adjustment, RC1 and RC2 of well X coincide at a single gas line (as shown). Figure 4 As shown in section 8), according to equation (6), the convection of different light hydrocarbon components at the crack is nonlinear, and the lighter the component, the smaller the viscosity μ, C n The larger the RC n The larger the value, the more likely a fracture fluid identification pattern can be established by considering the local relative differences (RC1 and RC2) between two typical light hydrocarbon components (C1 and C2). Figure 3 As shown, at the wellhead of a large-scale fracture, when gas is present, RC1 and RC2 have a positive envelope; when oil is present, RC1 and RC2 have a positive or negative envelope, depending on the oil quality. However, the absolute values ​​of RC1 and RC2 and the envelope area are smaller than those in the gas-containing case. A more refined regional identification model can be established by combining actual sampling data. When water is present or in a dry layer, RC1 and RC2 overlap and their absolute values ​​tend to be 0.

[0018] S4. Determine reservoir fluid properties by utilizing the local relative anomalies of light hydrocarbon components at the wellhead crossing of large-scale fractures.

[0019] Specifically: For each large-scale crack f that meets the requirements picked in step S1 i Pick its depth z=d i Local relative anomaly of light hydrocarbon components at RC n ,like Figure 4 As shown in the 8th question.

[0020] Then, based on the crack fluid identification model established in step S3, fluid properties are identified: Among the fractures, f1 and f2 have relatively high RC1 and RC2 values, and after calibration, they exhibit a positive envelope, consistent with the characteristics of gas-bearing fractures, indicating that the reservoir is a gas-bearing layer. Fractury f3 has lower RC1 and RC2 values ​​than f1 and f2, but still exhibits a certain positive envelope, consistent with the characteristics of oil-bearing fractures, indicating that the reservoir is an oil-bearing layer. Fractury f4 has relatively low RC1 and RC1 values, approaching 0, consistent with the characteristics of water-bearing fractures, indicating that the reservoir is a water-bearing layer. This well's test section ( Figure 4 (Last step) Daily oil production: 132.66m 3 Daily gas production: 223,417 m³ 3 / day, filtrate: 88.38m³ 3 / day, for a condensate gas reservoir, verifying the accuracy of fluid identification using this method. For comparison, Figure 4 The penultimate result shows the fluid identification results of the pulsed neutron method. By superimposing the count rates of long and short source distances (red envelope), the gas layer features in the range of 3715.0m to 3795.0m and the oil layer features in the range of 3795.0m to 3900.0m (green envelope) can be effectively identified. However, the features in the range of 3860.0m to 3865.0m are misidentified as gas layer features due to the effect of diameter expansion. In addition, the water layer below 3925.0m is also misidentified as gas layer features due to the effect of diameter expansion.

[0021] This invention utilizes the logging gas response at the initial stage of fracture drilling to reduce mud intrusion interference. Furthermore, it replaces the original gas logging analysis of the entire well section with local anomaly analysis of light hydrocarbon components at large-scale fracture crossings. This effectively overcomes the weakening logging response in low-porosity, low-permeability matrices, as well as the influence of non-fluid factors such as wellbore enlargement and mud system changes. It provides new ideas and methods for comprehensive fluid identification in logging of complex reservoirs, and has significant guiding significance for determining oil, gas, and water interfaces, establishing effective thicknesses, calculating reliable reserves, development plans, and fracturing design.

[0022] The applicant declares that the above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention fall within the protection and disclosure scope of the present invention.

Claims

1. A method for identifying reservoir fluid properties based on the detection of local gas logging anomalies in fractures, characterized in that: Includes the following steps: S1. Use well logging data to identify large-scale fractures and their depths through the well. S2. Calculate the local relative anomaly curves of different light hydrocarbon components in the well logging; S3. Adjust the scale range of the local relative anomaly curve of light hydrocarbon components that change with depth, and establish a fracture fluid identification mode. S4. Determine reservoir fluid properties by utilizing the local relative anomalies of light hydrocarbon components at the wellhead crossing of large-scale fractures.

2. The reservoir fluid property identification method based on local gas logging anomaly detection in fractures according to claim 1, characterized in that: Step S1 specifically includes the following steps: S11. Use acoustic remote logging to obtain well perimeter reflected wave imaging results, and extract the fracture trajectory from the reflected wave imaging results. S12. Calculate the crack extension length; The formula for calculating the crack extension length is: In the formula: The length of the crack is expressed in meters (m). The depth of the crack initiation point, in meters; The depth of the crack termination point, in meters; The vertical distance from the fracture initiation point to the well, in meters; The vertical distance from the well to the termination point of the fracture is expressed in meters (m). S13. Based on the calculation results of step S12, large-scale fractures are picked out, and the well depth of the picked large-scale fractures is determined by combining the results of electrical imaging logging.

3. The reservoir fluid property identification method based on local gas logging anomaly detection in fractures according to claim 2, characterized in that: The standard for large-scale cracks is an extension length greater than 10m.

4. The reservoir fluid property identification method based on local gas logging anomaly detection in fractures according to claim 1, characterized in that: Step S2 specifically includes the following steps: S21. Calculate the low-frequency trend curve of light hydrocarbon component content as a function of depth using a low-pass filter; The expression for the low-frequency trend curve of the light hydrocarbon component content varying with depth is: In the formula: The trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; lowfilt is a low-pass filter; f represents the content of light hydrocarbon components containing n carbon atoms at different depths z, expressed as %; z is the depth in meters (m); n is the number of carbon atoms in the light hydrocarbon component, where n is an integer less than or equal to 5; c f is the spatial cutoff frequency of the low-pass filter, expressed in 1 / m. s This represents the spatial sampling frequency of the low-pass filter, expressed in units of 1 / m. S22. The low-frequency trend of the content of light hydrocarbon components varying with depth is processed by median filtering to obtain the local trend curve of the content of light hydrocarbon components varying with depth. The expression for the local trend curve of the light hydrocarbon component content varying with depth is: In the formula: The local trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; midfilt is the median filter; The trend line represents the content of light hydrocarbon components containing n carbon atoms at different depths z, %; w is the sliding window length of the median filter, odd number, dimensionless. S23. The relative difference between the low-frequency trend of light hydrocarbon component content with depth and the local trend of light hydrocarbon component content is used as a local relative anomaly of light hydrocarbon component with depth. The expression for the local relative anomaly curve of the light hydrocarbon component varying with depth is: Where: RC n (z) represents the local relative anomaly curves of the content of light hydrocarbon components containing n carbon atoms at different depths z, % The trend lines for the content of light hydrocarbon components containing n carbon atoms at different depths z are given by %. This represents the local trend line of the content of light hydrocarbon components containing n carbon atoms at different depths z.

5. The reservoir fluid property identification method based on local gas logging anomaly detection in fractures according to claim 1, characterized in that: Step S3 specifically includes the following steps: S31. Establish the calculation formula for the content of light hydrocarbon components that diffuse into the wellbore in the reservoir matrix within time t, as shown in the following formula (5). In the formula: The content of light hydrocarbon components with n carbon atoms diffused into the wellbore within time t of the reservoir matrix, %; D n Let be the diffusion coefficient of the light hydrocarbon component with n carbon atoms in the reservoir matrix, %; c n t represents the concentration of light hydrocarbon components with n carbon atoms in the reservoir matrix, %; t represents the time length during which the light hydrocarbon components exchange with the wellbore, in seconds. S32. Establish the relationship between the content of light hydrocarbon components in the wellbore during the communication between the fracture and the wellbore within time t, as shown in the following formula (6). In the formula: The percentage is the content of light hydrocarbon component with n carbon atoms that diffuses into the wellbore within time t in the fractured portion; t is the time length during which the light hydrocarbon component exchanges with the wellbore, in seconds. in: Where: L is the crack extension length, %; μ is the crack fluid viscosity, in mPa·s; This refers to the permeability of cracks, expressed in meters (m). 2 ΔP is the pressure difference between the wellbore and the fracture, in Pa. S33. A fracture fluid identification model at the well crossing of large-scale fractures is established by using the local relative differences RC1 and RC2 between two typical light hydrocarbon components C1 and C2.

6. The reservoir fluid property identification method based on local gas logging anomaly detection in fractures according to claim 5, characterized in that: The specific crack fluid identification mode is as follows: When gas is present, RC1 and RC2 form a positive envelope; When oil is present, RC1 and RC2 form a positive or negative envelope, and the absolute values ​​and envelope areas of RC1 and RC2 are smaller than those when gas is present. When the layer is wet or dry, RC1 and RC2 coincide and their absolute values ​​tend to be 0.

7. The reservoir fluid property identification method based on local gas logging anomaly detection in fractures according to claim 1, characterized in that: Step S4 specifically involves: processing each large-scale crack f that meets the requirements identified in step S1. i Pick its depth z=d i Local relative anomaly of light hydrocarbon components at RC n Then, based on the fracture fluid identification model established in step S3, the fluid properties are identified, and the fluid properties in the reservoir above the fracture crossing point can be determined as the fluid properties in the fracture.

Images