A method for predicting remaining oil using time-lapse seismic

Abstract

The present invention provides a method for predicting remaining oil by time-lapse seismic, which is used to solve the problem that the difference in amplitude of conventional time-lapse seismic is greatly affected by acquisition and processing and cannot accurately predict the distribution of remaining oil. The specific steps include: obtaining measured seismic data of the early and late periods through two field seismic acquisitions in the early and late periods; establishing the relationship between the drilling lithofacies interpretation conclusion and the wave impedance curve based on drilling and core data; performing well-seismic calibration on the wave impedance curve and seismic data of the same period to obtain the average seismic wavelet of the early and late periods; using a multi-point geostatistical simulation method to establish a lithofacies model; using the lithofacies model as a constraint and the relationship between lithofacies and wave impedance curves as a bridge, using a truncated Gaussian simulation algorithm to respectively establish the final wave impedance models of the early and late periods; performing difference calculation on the final wave impedance models of the early and late periods to obtain a wave impedance difference model; using the wave impedance difference model, a wave impedance difference profile and a wave impedance difference plane can be extracted, and the part with a smaller difference is the part where the remaining oil is developed, thereby improving the remaining oil prediction accuracy.

Description

Technical Field

[0001] The invention relates to the field of oil field development, and in particular to a method for predicting remaining oil through time-lapse seismic prediction. Background Art

[0002] Accurate prediction of the remaining oil distribution in oil and gas basins is of great significance for improving oilfield recovery and efficient development.

[0003] Time-lapse seismic, also known as four-dimensional seismic, refers to the acquisition of three-dimensional seismic data at different times in the same block, and then using the differences in seismic data collected at different times to describe the changing characteristics of underground oil layers as production and development occur, thereby predicting the dominant locations for the development of remaining oil. The existing method is to predict the dominant locations for the development of remaining oil based on the difference in amplitude of time-lapse seismic, but the amplitude difference is greatly affected by acquisition and processing, resulting in large errors in the remaining oil prediction results, which affects oil and gas development.

[0004] Through research, it was found that wave impedance can reflect the changes in reservoir physical properties and fluids. Based on this, an innovative phase-controlled high-resolution time-lapse seismic wave impedance inversion method was established to obtain a wave impedance difference model to predict the distribution of remaining oil. This solved the problem that the difference in time-lapse seismic amplitude is greatly affected by acquisition and processing, and the distribution of remaining oil cannot be accurately predicted, thus improving the accuracy of remaining oil prediction. Summary of the invention

[0005] The purpose of the present invention is to invent a method for predicting remaining oil by time-lapse seismic, which uses seismic data collected twice and a phase-controlled high-resolution time-lapse seismic inversion method to describe the changing characteristics of oil layers with water injection development, thereby accurately predicting favorable locations for the development of remaining oil and improving the effect of oil field development.

[0006] The specific technical solution is as follows: A method for predicting remaining oil by time-lapse seismic includes the following steps:

[0007] Step 1: Obtain early measured seismic data and late measured seismic data through two field seismic acquisitions;

[0008] Step 2: Obtain the drilling lithofacies interpretation conclusion through drilling and core data;

[0009] Step 3: Obtain the early wave impedance curve and the late wave impedance curve through two well loggings, and establish the relationship between the drilling lithofacies interpretation conclusion and the early wave impedance curve, and between the drilling lithofacies interpretation conclusion and the late wave impedance curve respectively;

[0010] Step 4: Perform well-seismic calibration on the wave impedance curves and seismic data of the same period to obtain the early average seismic wavelet and the late average seismic wavelet;

[0011] Step 5: Based on the drilling lithofacies interpretation data, a lithofacies model is established using a multi-point geostatistical simulation method;

[0012] Step 6: Taking the lithofacies model as a constraint and the relationship between lithofacies and wave impedance curve as a basis, a truncated Gaussian simulation algorithm is used to obtain the early final wave impedance model and the late final wave impedance model;

[0013] Step 7: Calculate the difference between the early final wave impedance model and the late final wave impedance model to obtain a wave impedance difference model.

[0014] The specific method of step 3 is: through two loggings in the early and late periods, an early acoustic wave velocity curve, an early density curve, a late acoustic wave velocity curve, and a late density curve are obtained, and the early acoustic wave velocity curve and the early density curve are multiplied, and the late acoustic wave velocity curve and the late density curve are multiplied to obtain the wave impedance curves of the early and late periods, and the drilling lithofacies interpretation conclusion obtained in step 2 is used to establish the relationship between the drilling lithofacies interpretation conclusion and the early wave impedance curve, and the relationship between the drilling lithofacies interpretation conclusion and the late wave impedance curve.

[0015] The specific method of step 4 is: perform well-seismic calibration on the early wave impedance curve and the early seismic data to obtain the early seismic wavelet, perform well-seismic calibration on the late wave impedance curve and the late seismic data to obtain the late seismic wavelet, use the early seismic wavelets of multiple wells to obtain the early average seismic wavelet, and use the late seismic wavelets of multiple wells to obtain the late average seismic wavelet.

[0016] The specific method of step 5 is: using the drilling lithofacies interpretation data obtained in step 2 to establish a spatial training image, using the spatial training image as a constraint, and using a multi-point geostatistical simulation algorithm to establish a lithofacies model, including two lithofacies, one is sandstone and the other is mudstone.

[0017] The specific method of step 6 is: using the lithofacies model obtained in step 5 as a constraint, based on the relationship between lithofacies and wave impedance curves, using the truncated Gaussian simulation algorithm to respectively establish an early initial wave impedance model and a late initial wave impedance model, and convolve the early initial wave impedance model with the early average seismic wavelet to generate early synthetic seismic data, and convolve the late initial wave impedance model with the late average seismic wavelet to generate late synthetic seismic data; performing residual operations on the early synthetic seismic data and the early measured seismic data, and performing residual operations on the late synthetic seismic data and the late measured seismic data, and taking the residual satisfying the cutoff value as the iterative operation termination condition, repeatedly iterate to obtain the early final wave impedance model and the late final wave impedance model.

[0018] The specific method of step 7 is: using the early final wave impedance model and the late final wave impedance model obtained in step 6 to perform difference calculation to obtain a wave impedance difference model, and using the wave impedance difference model to extract a wave impedance difference profile and a wave impedance difference plane diagram, and the part with smaller difference is the part with residual oil development.

[0019] The beneficial effects of the present invention are:

[0020] The present invention can reflect the change of underground oil layer through the change of wave impedance in the area with time-lapse seismic data, and accurately predict the distribution position of remaining oil, which can improve the effect of oil field development. The method can be applied to different types of oil fields, and the calculation results are reasonable and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of the method of predicting remaining oil by time-lapse seismic;

[0022] Figures 2(a) and 2(b) are cross-sections of measured seismic data obtained from early and late field seismic acquisitions, respectively;

[0023] Figure 3 It is the relationship diagram between the drilling lithofacies interpretation conclusion and the early wave impedance curve and the late wave impedance curve;

[0024] Figures 4(a) and 4(b) are graphs of the early average seismic wavelet and the late average seismic wavelet, respectively;

[0025] Figure 5 It is a cross-section of the lithofacies model;

[0026] Figure 6 (a) and Figure 6 (b) are cross-sectional views of the early final wave impedance model and the late final wave impedance model, respectively;

[0027] Figure 7 It is the cross-sectional diagram of the wave impedance difference model;

[0028] Figure 8 It is the plane diagram of the wave impedance difference model. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings, taking the SN block of Daqing Oilfield as an example.

[0030] like Figure 1 As shown, a method for predicting remaining oil through time-lapse seismic includes the following steps:

[0031] Step 1: Obtain early measured seismic data and late measured seismic data through two field seismic acquisitions;

[0032] Step 2: Obtain the drilling lithofacies interpretation conclusion through drilling and core data;

[0033] Step 3: Obtain the early wave impedance curve and the late wave impedance curve through two well loggings, and establish the relationship between the drilling lithofacies interpretation conclusion and the early wave impedance curve, and between the drilling lithofacies interpretation conclusion and the late wave impedance curve respectively;

[0034] Step 4: Perform well-seismic calibration on the wave impedance curves and seismic data of the same period to obtain the early average seismic wavelet and the late average seismic wavelet;

[0035] Step 5: Based on the drilling lithofacies interpretation data, a lithofacies model is established using a multi-point geostatistical simulation method;

[0036] Step 6: Taking the lithofacies model as a constraint and the relationship between lithofacies and wave impedance curve as a basis, a truncated Gaussian simulation algorithm is used to obtain the early final wave impedance model and the late final wave impedance model;

[0037] Step 7: Calculate the difference between the early final wave impedance model and the late final wave impedance model to obtain a wave impedance difference model.

[0038] The specific method of step 3 is: through two loggings in the early and late periods, an early acoustic wave velocity curve, an early density curve, a late acoustic wave velocity curve, and a late density curve are obtained, and the early acoustic wave velocity curve and the early density curve are multiplied, and the late acoustic wave velocity curve and the late density curve are multiplied to obtain the wave impedance curves of the early and late periods, and the drilling lithofacies interpretation conclusion obtained in step 2 is used to establish the relationship between the drilling lithofacies interpretation conclusion and the early wave impedance curve, and the relationship between the drilling lithofacies interpretation conclusion and the late wave impedance curve.

[0039] The specific method of step 4 is: perform well-seismic calibration on the early wave impedance curve and the early seismic data to obtain the early seismic wavelet, perform well-seismic calibration on the late wave impedance curve and the late seismic data to obtain the late seismic wavelet, use the early seismic wavelets of multiple wells to obtain the early average seismic wavelet, and use the late seismic wavelets of multiple wells to obtain the late average seismic wavelet.

[0040] The specific method of step 5 is: using the drilling lithofacies interpretation data obtained in step 2 to establish a spatial training image, using the spatial training image as a constraint, and using a multi-point geostatistical simulation algorithm to establish a lithofacies model, including two lithofacies, one is sandstone and the other is mudstone.

[0041] The specific method of step 6 is: using the lithofacies model obtained in step 5 as a constraint, based on the relationship between lithofacies and wave impedance curves, using the truncated Gaussian simulation algorithm to respectively establish an early initial wave impedance model and a late initial wave impedance model, and convolve the early initial wave impedance model with the early average seismic wavelet to generate early synthetic seismic data, and convolve the late initial wave impedance model with the late average seismic wavelet to generate late synthetic seismic data; performing residual operations on the early synthetic seismic data and the early measured seismic data, and performing residual operations on the late synthetic seismic data and the late measured seismic data, and taking the residual satisfying the cutoff value as the iterative operation termination condition, repeatedly iterate to obtain the early final wave impedance model and the late final wave impedance model.

[0042] The specific method of step 7 is: using the early final wave impedance model and the late final wave impedance model obtained in step 6 to perform difference calculation to obtain a wave impedance difference model. Using the wave impedance difference model, a wave impedance difference profile and a wave impedance difference plane can be extracted. The part with smaller difference is the part with residual oil development. Example

[0043] Step 1: In the SN block of Daqing Oilfield, two sets of measured seismic data were obtained through field seismic acquisition in 2001 and 2015, as shown in Figure 2.

[0044] Step 2: In the SN block of Daqing Oilfield, the lithological interpretation conclusions of the drilling were obtained using drilling and core data, including two lithofacies: sandstone and mudstone.

[0045] Step 3: In the SN block of Daqing Oilfield, through two well loggings in the early and late periods, the early acoustic velocity curve, the early density curve, the late acoustic velocity curve, and the late density curve are obtained. The early acoustic velocity curve and the early density curve are multiplied, and the late acoustic velocity curve and the late density curve are multiplied to obtain the wave impedance curves of the early and late periods. The drilling lithofacies interpretation conclusion obtained in step 2 is used to establish the relationship between the drilling lithofacies interpretation conclusion and the early wave impedance curve, and the drilling lithofacies interpretation conclusion and the late wave impedance curve, as shown in Figure 2. Figure 3 shown.

[0046] Step 4: The early wave impedance curve obtained in step 3 is calibrated with the early seismic data to obtain the early seismic wavelet, and the late wave impedance curve is calibrated with the late seismic data to obtain the late seismic wavelet. The early seismic wavelets of multiple wells are used to obtain the early average seismic wavelet, and the late seismic wavelets of multiple wells are used to obtain the late average seismic wavelet, as shown in Figure 4.

[0047] Step 5: Use the drilling lithofacies interpretation data obtained in step 2 to establish a spatial training image. Using the spatial training image as a constraint, a multi-point geostatistical simulation algorithm is used to establish a lithofacies model, including two lithofacies, one is sandstone and the other is mudstone, such as Figure 5 shown.

[0048] Step 6: Using the lithofacies model obtained in step 5 as a constraint, based on the lithofacies and wave impedance curve relationship established in step 3, the truncated Gaussian simulation algorithm is used to establish the early initial wave impedance model and the late initial wave impedance model, and the early initial wave impedance model is convolved with the early average seismic wavelet obtained in step 4 to generate early synthetic seismic data, and the late initial wave impedance model is convolved with the late average seismic wavelet obtained in step 4 to generate late synthetic seismic data. The early synthetic seismic data and the early measured seismic data are subjected to residual calculation, and the late synthetic seismic data and the late measured seismic data are subjected to residual calculation, and the residuals satisfying the cutoff value are used as the termination conditions of the iteration calculation, and the iteration is repeated to obtain the early final wave impedance model and the late final wave impedance model. As shown in Figure 6, the early wave impedance is larger, the late wave impedance is smaller, and the impedance value decreases with time.

[0049] Step 7: Use the early final wave impedance model and the late final wave impedance model obtained in step 6 to perform difference calculation to obtain the wave impedance difference model. Use the wave impedance difference model to extract the wave impedance difference profile and wave impedance difference plane. The part with the smaller difference is the part with the development of residual oil, such as Figure 7 As shown, Figure 8 A plane diagram of the wave impedance difference model is given. It can be seen that the area with smaller difference is the area where the remaining oil is developed, which is mainly distributed at the edge of the composite channel sand body and inside the narrow channel, which is consistent with the result of the water saturation curve, indicating that the method of the present invention has high accuracy.

Claims

1. A method for predicting remaining oil by time-lapse seismic, characterized in that: The remaining oil distribution is predicted by using the seismic data collected twice, combined with the drilling acoustic velocity and density logging curves of the same period as the two sets of measured seismic data, and using the phase-controlled high-resolution time-lapse seismic inversion method; the following steps are included: Step 1: Obtain early measured seismic data and late measured seismic data through two field seismic acquisitions; Step 2: Obtain the drilling lithofacies interpretation conclusion through drilling and core data; Step 3: Obtain the early wave impedance curve and the late wave impedance curve through two well loggings, and establish the relationship between the drilling lithofacies interpretation conclusion and the early wave impedance curve, and between the drilling lithofacies interpretation conclusion and the late wave impedance curve respectively; Step 4: Perform well-seismic calibration on the wave impedance curves and seismic data of the same period to obtain the early average seismic wavelet and the late average seismic wavelet; Step 5: Based on the drilling lithofacies interpretation data, a lithofacies model is established using a multi-point geostatistical simulation method; Step 6: Based on the relationship between lithofacies and wave impedance curves, the truncated Gaussian simulation algorithm is used to establish an early initial wave impedance model and a late initial wave impedance model respectively, and the early initial wave impedance model is convolved with the early average seismic wavelet to generate early synthetic seismic data, and the late initial wave impedance model is convolved with the late average seismic wavelet to generate late synthetic seismic data; residual calculations are performed on the early synthetic seismic data and the early measured seismic data, and residual calculations are performed on the late synthetic seismic data and the late measured seismic data, and the iterative calculation termination conditions are respectively taken as the residuals satisfying the cutoff value, and the iterations are repeated to obtain the early final wave impedance model and the late final wave impedance model; Step 7: Calculate the difference between the early final wave impedance model and the late final wave impedance model to obtain a wave impedance difference model.

2. A method for predicting remaining oil by time-lapse seismic prediction according to claim 1, characterized in that: The specific method of step 3 is: through two loggings in the early and late periods, an early acoustic wave velocity curve, an early density curve, a late acoustic wave velocity curve, and a late density curve are obtained, and the early acoustic wave velocity curve and the early density curve are multiplied, and the late acoustic wave velocity curve and the late density curve are multiplied to obtain the wave impedance curves of the early and late periods, and the drilling lithofacies interpretation conclusion obtained in step 2 is used to establish the relationship between the drilling lithofacies interpretation conclusion and the early wave impedance curve, and the relationship between the drilling lithofacies interpretation conclusion and the late wave impedance curve.

3. The method for predicting remaining oil by time-lapse seismic prediction according to claim 1, characterized in that: The specific method of step 5 is: using the drilling lithofacies interpretation data obtained in step 2 to establish a spatial training image, using the spatial training image as a constraint, and using a multi-point geostatistical simulation algorithm to establish a lithofacies model, including two lithofacies: sandstone and mudstone.

4. The method for predicting remaining oil by time-lapse seismic prediction according to claim 1, characterized in that: The specific method of step 7 is: using the early final wave impedance model and the late final wave impedance model obtained in step 6 to perform difference calculation to obtain a wave impedance difference model, and using the wave impedance difference model to extract a wave impedance difference profile and a wave impedance difference plane diagram, and the part with smaller difference is the part with residual oil development.

Images