Method for establishing prediction model of viscosity of heavy oil reservoir, prediction method and application

Abstract

The invention provides a heavy oil reservoir viscosity prediction model establishment method, a prediction method and application, and relates to the field of crude oil physical property detection. According to the method for establishing the prediction model of the viscosity of the heavy oil reservoir, the characteristic that rock debris pyrolysis derived parameters are associated with the physical property of underground crude oil (heavy oil) is utilized, derived parameters of sample data tested by a rock debris pyrolysis geochemical instrument are fitted with viscosity data of the heavy oil, and the prediction model of the viscosity of the heavy oil and the (rock debris pyrolysis) derived parameters is established; the prediction model can accurately predict the viscosity of the thickened oil in the same construction unit, effectively solves the problems of difficult measurement of the viscosity of the thickened oil and many interference conditions, and provides a new method for predicting the viscosity of the thickened oil. The invention also provides a method for predicting the viscosity of the heavy oil reservoir, the viscosity of the target heavy oil reservoir is predicted by adopting the prediction model established by the method for establishing the prediction model of the viscosity of the heavy oil reservoir, and the prediction method can improve the estimation and prediction capability of the viscosity of the heavy oil reservoir.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of physical property detection of crude oil, and relates to a method for establishing a prediction model of viscosity of a heavy oil reservoir, a prediction method and application. BACKGROUND

[0002] Heavy oil is a relatively viscous crude oil with high viscosity and poor flowability. The viscosity of heavy oil is an important parameter in oil exploration, which not only affects the oil recovery efficiency and yield prediction, but also relates to the oilfield development plan, oil refining process and product quality in the development and production of oilfields; in exploration, it affects the evaluation of the flowability, distribution and productivity potential of the crude oil flow in the reservoir.

[0003] The commonly used heavy oil viscosity measurement methods at present include the Brookfield viscosity method, the rotary viscosity method and the differential pressure method. However, due to the high viscosity and complexity of heavy oil, the measurement results are often affected by temperature changes, addition or removal of chemicals, oxidation and shear rate. Therefore, the above-mentioned traditional viscosity measurement methods may face some challenges. Therefore, how to accurately determine the viscosity of heavy oil is very important to ensure the accuracy of decision-making, reduce economic cost, improve efficiency and resource utilization.

[0004] In view of this, the present application is proposed. SUMMARY

[0005] The present application aims to provide a method for determining the viscosity of a heavy oil reservoir and application, to solve the above-mentioned deficiencies existing in the prior art.

[0006] In order to achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0007] The first object of the present application is to provide a method for establishing a prediction model of viscosity of a heavy oil reservoir, comprising the following steps:

[0008] (a) obtaining sample data tested by a pyrolysis geochemical instrument of cuttings of a plurality of heavy oil wells and heavy oil viscosity data corresponding to the heavy oil wells;

[0009] (b) preprocessing the heavy oil viscosity data obtained in step (a);

[0010] (c) calculating each derived parameter and performing feature extraction according to the sample data obtained in step (a) to obtain characteristic parameters;

[0011] (d) performing data analysis on a part of the heavy oil viscosity data preprocessed in step (b) and the characteristic parameters of the heavy oil wells corresponding thereto, and establishing a prediction model;

[0012] (e) verifying the prediction effect of the prediction model established in step (d) by using the remaining part of the heavy oil viscosity data preprocessed in step (b).

[0013] Further, on the basis of the above technical solutions of the present application, in step (a), the multiple heavy oil wells include different time and / or geographical distribution heavy oil wells in the same structural unit.

[0014] Further, on the basis of the above technical solutions of the present application, in step (a), the sample data includes S0, S 11 , S 21 , S 22 and S 23 , wherein S0 is the hydrocarbon content in unit mass of the oil-bearing rock detected at 90℃, S 11 is the hydrocarbon content in unit mass of the oil-bearing rock detected at 200℃, S 21 is the hydrocarbon content in unit mass of the oil-bearing rock detected at >200℃-350℃, S 22 is the hydrocarbon content in unit mass of the oil-bearing rock detected at >350℃-450℃, and S 23 is the hydrocarbon content in unit mass of the oil-bearing rock detected at >450℃-600℃.

[0015] Further, on the basis of the above technical solutions of the present application, in step (b), the pretreatment is to exclude abnormal values or missing values in the heavy oil viscosity data.

[0016] Further, on the basis of the above technical solutions of the present application, in step (c), the derived parameters include total oil and gas content ST, condensate oil index P1, light oil index P2, medium oil index P3, heavy oil index P4 and crude oil light and heavy component ratio P S , which are calculated according to the following formulas using the sample data obtained in step (a):

[0017]

[0018]

[0019] ST = S0 + S 11 + S 21 + S 22 + S 23 (6)

[0020] In formulas (1)-(6), S0 is the hydrocarbon content in unit mass of the oil-bearing rock detected at 90℃; S 11 is the hydrocarbon content in unit mass of the oil-bearing rock detected at 200℃; S 21 is the hydrocarbon content in unit mass of the oil-bearing rock detected at >200℃-350℃; S 22 is the hydrocarbon content in unit mass of the oil-bearing rock detected at >350℃-450℃; and S 23The hydrocarbon content in the unit mass of oil storage rock detected at >450℃-600℃.

[0021] Further, on the basis of the technical scheme of the present application, in step (d), a prediction model is established by a regression equation as follows:

[0022] R=a+b*P1+c*P3+d*P4 (7)

[0023] R=e+f*P S (8)

[0024] In the equations (7)-(8), R is the predicted viscosity of the heavy oil; a, b, c, d, e and f are model coefficients; P1 is the condensate oil index; P3 is the medium oil index; P4 is the heavy oil index; P S is the ratio of light and heavy components of the crude oil.

[0025] Further, on the basis of the technical scheme of the present application, in step (d), 60%-80% of the heavy oil viscosity data after the pretreatment in step (b) is used to analyze the data of the characteristic parameters of the heavy oil well corresponding thereto and to establish a prediction model.

[0026] Further, on the basis of the technical scheme of the present application, in step (e), 20%-40% of the heavy oil viscosity data after the pretreatment in step (b) is used to test the prediction effect of the prediction model established in step (d).

[0027] The second object of the present application is to provide a prediction method of the viscosity of a heavy oil reservoir, which uses the prediction model of the viscosity of a heavy oil reservoir established by the method for establishing a prediction model of the viscosity of a heavy oil reservoir according to the first object of the present application.

[0028] Further, on the basis of the technical scheme of the present application, the target heavy oil reservoir and the multiple heavy oil wells used to provide sample data in the process of establishing the prediction model are located in the same structural unit.

[0029] The third object of the present application is to provide the application of the method for establishing a prediction model of the viscosity of a heavy oil reservoir or the prediction method of the viscosity of a heavy oil reservoir in the field of detection of the properties of crude oil.

[0030] Compared with the prior art, the technical scheme of the present application has at least the following technical effects:

[0031] (1) The present application provides a method for establishing a prediction model of the viscosity of a heavy oil reservoir, which utilizes the correlation between the derived parameters of pyrolysis of drill cuttings and the properties of underground crude oil (heavy oil), and adopts the derived parameters of sample data tested by a pyrolysis geochemical instrument to fit with the viscosity data of heavy oil, so as to establish a prediction model of heavy oil viscosity and derived parameters (pyrolysis), which can more accurately predict the viscosity of heavy oil in the same structural unit, effectively solve the problems of difficulty in determining heavy oil viscosity and many interference conditions, and also provide a new method for predicting heavy oil viscosity.

[0032] (2) The present application provides a method for predicting the viscosity of a heavy oil reservoir, which adopts the prediction model established by the method for establishing a prediction model of the viscosity of a heavy oil reservoir to predict the viscosity of a target heavy oil reservoir, which can improve the estimation and prediction ability of heavy oil viscosity, reduce the uncertainty of exploration affairs and economic cost, and provide a new method for predicting heavy oil viscosity. BRIEF DESCRIPTION OF DRAWINGS

[0033] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application, and are incorporated in and constitute a part of this application. Among these figures:

[0034] Figure 1 The overall flowchart of the method for establishing a prediction model of the viscosity of a heavy oil reservoir provided by the present application is shown in the figure.

[0035] Figure 2 The scatter plot of heavy oil viscosity and pyrolysis derived parameters P1, P3 and P4 corresponding to the well section in the Hasi area of Example 2 of the present application is shown in the figure.

[0036] Figure 3 The scatter plot of heavy oil viscosity and pyrolysis derived parameter of the ratio of light and heavy components of crude oil (P S ) corresponding to the well section in the Hasi area of Example 2 of the present application is shown in the figure. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the embodiments of the present application. It should be understood by those skilled in the art that the embodiments are only used to help understand the present application, and should not be regarded as a specific limitation on the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application. The process parameters not specified in the following embodiments are usually in accordance with conventional conditions.

[0038] The endpoints of the ranges and any values disclosed herein are not limited to the precise values recited as the exact dimensions are not critical to the invention. Any numeric range recited is intended to include all values between the recited upper and lower values. In this sense, the phrase "between X and Y" is intended to capture all values X to Y, not just the values between the endpoints of the range. The value of the endpoints of the ranges and the individual values can be combined with each other to create one or more new ranges of values, which are to be considered disclosed within the present invention.

[0039] According to a first aspect of the present invention, there is provided a method for establishing a prediction model of viscosity of heavy oil reservoirs, comprising the following steps:

[0040] (a) obtaining sample data of rock pyrolysis geochemical instrument test and corresponding viscosity data of heavy oil of multiple heavy oil wells;

[0041] (b) preprocessing the heavy oil viscosity data obtained in step (a);

[0042] (c) calculating derived parameters and extracting features according to the sample data obtained in step (a) to obtain characteristic parameters;

[0043] (d) performing data analysis on part of the heavy oil viscosity data preprocessed in step (b) and the characteristic parameters of the heavy oil wells corresponding thereto, and establishing a prediction model;

[0044] (e) testing or verifying the prediction effect of the prediction model established in step (d) by using the remaining part of the heavy oil viscosity data preprocessed in step (b).

[0045] Specifically, in step (a), multiple heavy oil wells in the same structural unit are used as the source of sample collection to ensure the diversity and representativeness of the samples. The sample data includes rock pyrolysis geochemical instrument parameters S0, S 11 , S 21 , S 22 and S 23 , and viscosity data (represented by R1, R2, R3, etc.) corresponding to the heavy oil wells.

[0046] Wherein, S0 (natural gas peak) is the content of hydrocarbon (natural gas) in unit mass of oil-bearing rock detected at 90°C, S 11 (gasoline peak) is the content of hydrocarbon (gasoline) in unit mass of oil-bearing rock detected at 200°C, S 21 (kerosene + diesel peak) is the content of hydrocarbon (kerosene + diesel) in unit mass of oil-bearing rock detected at >200°C-350°C, S 22 (wax + heavy oil peak) is the content of hydrocarbon (wax + heavy oil) in unit mass of oil-bearing rock detected at >350°C-450°C, S 23The (gum+asphaltene peak) is the hydrocarbon (gum+asphaltene) content in the unit mass of the oil storage rock detected at >450 DEG C to 600 DEG C.

[0047] Step (b) is to pretreat the heavy oil viscosity data to ensure the quality and reliability of the subsequent data analysis and the established model.

[0048] Step (c) is to convert the sample data (direct parameters) tested by the rock pyrolysis instrument into derived parameters (indirect parameters), and then to extract features of each derived parameter, which is a process of removing redundant information, can remove redundant data, facilitate data analysis, reduce model calculation time and improve model prediction accuracy.

[0049] The pretreated heavy oil viscosity data of step (b) and the characteristic parameters in step (c) are divided into two parts, one part of the pretreated heavy oil viscosity data and the corresponding characteristic parameters in step (c) are mainly used for the establishment of the prediction model, that is, step (d), and the other part of the pretreated heavy oil viscosity data and the corresponding characteristic parameters in step (c) are mainly used for the verification of the prediction model, that is, step (e).

[0050] The prediction model establishment method of the heavy oil reservoir viscosity provided by the application is based on the derived parameters of rock pyrolysis and the prediction method of heavy oil viscosity, which not only avoids the interference factors in the determination process of heavy oil viscosity, but also reasonably reflects the different response degrees of different pyrolysis derived parameters to heavy oil viscosity.

[0051] As an optional embodiment of the application, in step (a), the multiple heavy oil wells include different (production) time and / or different geographical distribution of heavy oil wells in the same structural unit, that is, the multiple heavy oil wells can be the same (geographical distribution) heavy oil well in the same structural unit at different (production) times, or the same (production) time heavy oil well in the same structural unit with different geographical distribution, or the heavy oil well in the same structural unit with different (production) time and different geographical distribution. Through the further limitation of the multiple heavy oil wells, the diversity and representativeness of the data are ensured.

[0052] As an optional embodiment of the application, in step (a), the sample data includes S0, S 11 , S 21 , S 22 and S 23 .

[0053] Because there is a certain range of error in the analysis and test process, and the measured value is abnormal due to the different geological conditions of crude oil in the ground, it is necessary to pretreat the heavy oil viscosity data.

[0054] As an optional embodiment of the present application, in step (b), the pretreatment is to exclude abnormal values or missing values in the thickened oil viscosity data. The abnormal thickened oil viscosity data is identified and removed to ensure the accuracy and reliability of subsequent data analysis and model establishment.

[0055] As an optional embodiment of the present application, in step (c), the derived parameters include total oil and gas content ST, condensate oil index P1, light oil index P2, medium oil index P3, heavy oil index P4 and crude oil light and heavy component ratio P S The sample data obtained in step (a) is calculated according to the following formula:

[0056]

[0057]

[0058] ST = S0 + S 11 + S 21 + S 22 + S 23 (6)

[0059] In formula (1)-(6), S0 (natural gas peak) is the content of hydrocarbon (natural gas) in unit mass of oil reservoir rock detected at 90°C, S 11 (gasoline peak) is the content of hydrocarbon (gasoline) in unit mass of oil reservoir rock detected at 200°C, S 21 (kerosene + diesel peak) is the content of hydrocarbon (kerosene + diesel) in unit mass of oil reservoir rock detected at >200°C-350°C, S 22 (wax + heavy oil peak) is the content of hydrocarbon (wax + heavy oil) in unit mass of oil reservoir rock detected at >350°C-450°C, S 23 (gel + asphaltene peak) is the content of hydrocarbon (gel + asphaltene) in unit mass of oil reservoir rock detected at >450°C-600°C.

[0060] The thickened oil viscosity data after pretreatment in step (b) is divided into two parts, one part of the number of thickened oil viscosity data after pretreatment in step (b) and its corresponding characteristic parameters of thickened oil well are analyzed and a prediction model is established, and the other part of the number of thickened oil viscosity data after pretreatment in step (b) is used to test the prediction effect of the established prediction model.

[0061] As an optional embodiment of the present application, in step (d), 60-80% of the thickened oil viscosity data after pretreatment in step (b) and its corresponding characteristic parameters of thickened oil well are used for data analysis and establishment of a prediction model.

[0062] As an optional embodiment of the present application, in step (d), the prediction model is established by the following regression equation:

[0063] R = a + b x P1 + c x P3 + d x P4 (7)

[0064] R = e + f x P S (8)

[0065] wherein, in the formula (7)-(8), R is the predicted viscosity of the heavy oil; a, b, c, d, e and f are all model coefficients; P1 is the condensate oil index; P3 is the medium oil index; P4 is the heavy oil index; P S is the ratio of light and heavy components of the crude oil.

[0066] It should be noted that the prediction model of the viscosity of the heavy oil reservoir in different regions (different tectonic units) can be established by using the prediction model establishment method provided by the present application. The geological conditions, the composition of the heavy oil in the heavy oil well and the viscosity characteristics of different regions (different tectonic units) are different, and the prediction model established can be represented by the formula (7) and the formula (8), and the difference mainly lies in that the model coefficients a, b, c, d, e and f are different.

[0067] As an optional embodiment of the present application, in step (e), the prediction effect of the prediction model established in step (d) is tested by using 20%-40% of the heavy oil viscosity data after the pretreatment in step (b).

[0068] According to the second aspect of the present application, a prediction method of the viscosity of the heavy oil reservoir is also provided, which uses the prediction model established by using the prediction model establishment method of the viscosity of the heavy oil reservoir provided by the first aspect of the present application to predict the viscosity of the target heavy oil reservoir.

[0069] Regular heavy oil viscosity testing, collection of sufficient heavy oil samples, use of accurate experimental methods, combination of advanced prediction models and techniques can improve the estimation and prediction ability of the heavy oil viscosity, reduce the uncertainty of exploration affairs and economic cost, and provide a new method for the prediction of the viscosity of the heavy oil.

[0070] Since the geological conditions, the composition of the heavy oil in the heavy oil well and the viscosity characteristics of different regions (different tectonic units) are different, when the viscosity of the heavy oil reservoir of other heavy oil wells in a tectonic unit (for example, the Hashan region) is predicted, it is preferred to use the viscosity prediction model established by the sample data and the viscosity data of the heavy oil well in the tectonic unit (for example, the Hashan region) to predict, so as to ensure the accuracy of the prediction result.

[0071] According to the third aspect of the present application, the application of the above-mentioned prediction model establishment method of the viscosity of the heavy oil reservoir or the above-mentioned prediction method of the viscosity of the heavy oil reservoir in the field of detection of the properties of the crude oil is also provided.

[0072] The prediction model of the heavy oil reservoir viscosity and the prediction method of the heavy oil reservoir viscosity have the advantages that they have good application in the fields of crude oil property detection and exploration.

[0073] The application will be further described in detail below in combination with specific examples and comparative examples.

[0074] Example 1

[0075] The embodiment provides a prediction model establishing method of heavy oil reservoir viscosity, and a whole flowchart is shown in FIG. 1. Figure 1 The method comprises the following steps:

[0076] (a) obtaining sample data of a rock debris pyrolysis geochemical instrument (rock debris logging pyrolysis geochemical instrument) test of a production well section of a plurality of known oil (heavy oil) wells in the Bohai region and heavy oil viscosity data of the production well section;

[0077] (b) collating the heavy oil viscosity data obtained in step (a), and screening and removing abnormal heavy oil viscosity data;

[0078] (c) calculating pyrolysis derived parameters according to the sample data obtained in step (a), wherein each derived parameter comprises total oil and gas content (ST), condensate oil index (P1), light oil index (P2), medium oil index (P3), heavy oil index (P4) and crude oil light and heavy component ratio (P S ), and each derived parameter calculation method is shown in formulas (1)-(6):

[0079]

[0080] ST = S0 + S 11 + S 21 + S 22 + S 23 (6)

[0081] Then, the derived parameters are subjected to feature extraction to obtain feature parameters;

[0082] (d) performing data analysis and making a scatter plot on part (80%) of the heavy oil viscosity data after the pretreatment in step (b) and the feature parameters of the heavy oil well corresponding to the heavy oil viscosity data by using SPSS, and establishing a prediction (correlation) model, and a specific regression equation of the established prediction model is as follows:

[0083] R = a + b × P1 + c × P3 + d × P4 = 22789.1 - 5420.623 × P1 - 7739.120 × P3 - 17200.154 × P4

[0084] R = e + f × P S = 5736.978 + 876.567 × P S

[0085] In the formula, R is the predicted viscosity of heavy oil; P1, P3, P4, P S The derived parameters are condensate index, medium oil index, heavy oil index, and light and heavy component ratio of crude oil, respectively, based on the sample data tested by the cuttings pyrolysis geochemical analyzer; the model coefficients are a = 22789.1, b = -5420.623, c = -7739.120, d = -17200.154, e = 5736.978, and f = 876.567.

[0086] (e) The prediction effect of the prediction model established in step (d) is tested using the remaining amount (20%) of the preprocessed heavy oil viscosity data from step (b).

[0087] Example 2

[0088] This embodiment provides a method for establishing a predictive model for the viscosity of heavy oil reservoirs. The overall process diagram is shown below. Figure 1 As shown, it includes the following steps:

[0089] (a) Obtain sample data from cuttings pyrolysis geochemistry instrument (cuttings logging pyrolysis geochemistry instrument) tests on the production well sections of multiple known oil-producing (heavy oil) wells in the Hashan area, as well as the viscosity data of the heavy oil produced in that well section;

[0090] (b) Organize the heavy oil viscosity data obtained in step (a), identify and remove abnormal heavy oil viscosity data.

[0091] (c) Based on the sample data obtained in step (a), calculate each derived parameter. The calculation method for each derived parameter is as shown in formulas (1)-(6) in Example 1, and will not be repeated here.

[0092] Then, feature extraction is performed on each derived parameter to obtain the feature parameters;

[0093] (d) Using SPSS, a portion (80%) of the preprocessed heavy oil viscosity data from step (b) and the corresponding characteristic parameters of the heavy oil wells are analyzed and scatter plots are generated. Specifically, as follows... Figure 2 and Figure 3 As shown, a predictive model is established, and the specific regression equation is as follows:

[0094] R=a+b×P1+c×P3+d×P4=23281.537-6420.637×P1-8529.120×P3-15300.154×P4

[0095] R = e + f × P S =6947.998 + 568.647 × P S

[0096] In the formula, R is the predicted viscosity of heavy oil; P1, P3, P4, P S are respectively the derived parameters condensate oil index, medium oil index, heavy oil index and the ratio of light and heavy components of crude oil of the sample data tested by the rock pyrolysis geochemical instrument; the model coefficients a is 23281.537, b is -6420.637, c is -8529.120, d is -15300.154, e is 6947.998, and f is 568.647;

[0097] (e) using the remaining part (20%) of the heavy oil viscosity data of step (b) after pretreatment to test the prediction effect of the prediction model established in step (d).

[0098] It is found through comparison of the data of Example 1 and Example 2 that the error between the predicted value and the measured value of viscosity is basically about 10%, and the highest is 14%, thereby proving that there is no significant difference between the measured value and the predicted value, and the viscosity prediction model established can accurately predict the viscosity of heavy oil.

[0099] In order to further verify the prediction effect of the viscosity prediction model provided in the examples, the following experimental examples are specially set.

[0100] Experimental Example 1

[0101] Ten heavy oil wells in Hashan area are randomly selected, the sample data are tested by the rock pyrolysis geochemical instrument according to GB / T 18602-2012, the viscosity of heavy oil wells is detected by a viscometer according to GB / T 10247-2008 (the detection temperature is 40℃), then the derived parameters calculated from the sample data tested by the rock pyrolysis geochemical instrument are input into the viscosity prediction model established in Example 2 to predict the viscosity, and the specific results are shown in Table 1.

[0102] Table 1 Comparison of the measured value and the predicted value of viscosity content

[0103] Number Measured value / mPa s Predicted value (prediction model) / mPa s Error / % Table 1 : Measured and predicted values of the viscosity of the samples 1 7004.8627 8155.5825 16.40 2 7289.186 8464.85 16.13 3 7061.7274 7945.7332 12.51 4 7118.59 7057.781 0.85 5 7078.78681 7022.74 0.79 6 7175.4568 7479.478 4.23 7 7402.9156 8226.66 11.13 8 7016.235 5949.46 15.20 9 7459.78 7251.1 2.79 10 7488.21 9616.5175 14.22

[0104] As can be seen from the data in Table 1, the error between the predicted value and the measured value of viscosity is 0.7-16.4%, which indicates that there is no significant difference between the measured value and the predicted value, and the error of a few predicted values (14.22-16.40%) is large due to the influence of overpressure and other factors. Overall, the viscosity prediction model established by the present application is very robust.

[0105] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for establishing a predictive model for the viscosity of heavy oil reservoirs, characterized in that, Includes the following steps: (a) Obtain sample data from cuttings pyrolysis geochemical analyzer tests of multiple heavy oil wells and corresponding heavy oil viscosity data for the heavy oil wells; (b) Preprocess the heavy oil viscosity data obtained in step (a); (c) Based on the sample data obtained in step (a), calculate each derived parameter and extract features to obtain the feature parameters; (d) Perform data analysis on a portion of the preprocessed heavy oil viscosity data from step (b) and the corresponding characteristic parameters of the heavy oil wells, and establish a prediction model; (e) The prediction effect of the prediction model established in step (d) is tested using the remaining amount of heavy oil viscosity data after preprocessing in step (b).

2. The method for establishing a prediction model for the viscosity of heavy oil reservoirs according to claim 1, characterized in that, In step (a), the multiple heavy oil wells include heavy oil wells located within the same structural unit but distributed at different times and / or geographically.

3. The method for establishing a prediction model for the viscosity of heavy oil reservoirs according to claim 1, characterized in that, In step (a), the sample data includes S0, S 11 S 21 S 22 and S 23 Where S0 is the hydrocarbon content per unit mass of oil-bearing rock detected at 90℃, and S 11 S represents the hydrocarbon content per unit mass of oil-bearing rock measured at 200℃. 21 S represents the hydrocarbon content per unit mass of oil-bearing rock measured at temperatures ranging from >200℃ to 350℃. 22 S represents the hydrocarbon content per unit mass of oil-bearing rock measured at temperatures >350℃~450℃. 23 Hydrocarbon content per unit mass of oil-bearing rock detected at temperatures >450℃~600℃.

4. The method for establishing a prediction model for the viscosity of heavy oil reservoirs according to claim 1, characterized in that, In step (b), the preprocessing is to remove outliers or missing values ​​from the heavy oil viscosity data.

5. The method for establishing a prediction model for the viscosity of heavy oil reservoirs according to claim 1, characterized in that, In step (c), the derived parameters include total oil and gas content ST, condensate index P1, light oil index P2, medium oil index P3, heavy oil index P4, and the ratio of light to heavy components in crude oil P. S The sample data obtained in step (a) is calculated using the following formula: ST=S0+S 11 +S 21 +S 22 +S 23 (6) In formulas (1)-(6), S0 represents the hydrocarbon content per unit mass of oil-bearing rock detected at 90℃; S 11 Hydrocarbon content per unit mass of oil-bearing rock measured at 200℃; S 21 Hydrocarbon content per unit mass of oil-bearing rock measured at temperatures >200℃~350℃; S 22 Hydrocarbon content per unit mass of oil-bearing rock measured at temperatures >350℃~450℃; S 23 Hydrocarbon content per unit mass of oil-bearing rock detected at temperatures >450℃~600℃.

6. The method for establishing a prediction model for the viscosity of heavy oil reservoirs according to claim 5, characterized in that, In step (d), a prediction model is established using the following regression equation: R = a + b × P1 + c × P3 + d × P4 (7) R= e + f × P S (8) In formulas (7)-(8), R is the predicted viscosity of heavy oil; a, b, c, d, e, and f are model coefficients; P1 is the condensate index; P3 is the medium oil index; P4 is the heavy oil index; P S This represents the ratio of light to heavy components in crude oil.

7. The method for establishing a prediction model for the viscosity of heavy oil reservoirs according to claim 1, characterized in that, In step (d), 60% to 80% of the heavy oil viscosity data after preprocessing in step (b) and the corresponding characteristic parameters of the heavy oil wells are used for data analysis and a prediction model is established.

8. The method for establishing a prediction model for the viscosity of heavy oil reservoirs according to claim 6, characterized in that, In step (e), 20% to 40% of the heavy oil viscosity data after preprocessing in step (b) are used to test the prediction effect of the prediction model established in step (d).

9. A method for predicting the viscosity of heavy oil reservoirs, characterized in that, A prediction model is established using the method for establishing a prediction model for the viscosity of a heavy oil reservoir as described in any one of claims 1-8, to predict the viscosity of the target heavy oil reservoir; Preferably, the target heavy oil reservoir and the multiple heavy oil wells used to provide sample data during the establishment of the prediction model are located in the same structural unit.

10. The application of the method for establishing a prediction model for the viscosity of heavy oil reservoirs according to any one of claims 1-8 or the method for predicting the viscosity of heavy oil reservoirs according to claim 9 in the field of crude oil physical property testing.

Images