Thin reservoir horizontal well geosteering method based on three-dimensional model real-time adjustment

Abstract

The invention discloses a thin reservoir horizontal well geosteering method based on real-time adjustment of a three-dimensional model, which is used for predicting lithology of sand bodies between wells and geosteering the horizontal well by normalizing gamma curves of logging and logging, and comprises the following steps of: 1, normalizing natural gamma curve data of the horizontal well while drilling obtained from logging; the strain is named as GRL; 2, normalizing a logging natural gamma curve according to the mudstone baseline value of the natural gamma curve, and naming the logging natural gamma curve as GRC; 3, performing data inspection on the GRL curve and the GRC curve of the same horizontal well, and splicing the two inspected curves of the horizontal well into GRU; according to the method, in the aspect of mismatching of the data obtained by the logging measurement method and the logging measurement method, the accuracy of the logging measurement data is improved through normalization processing; the matching of the logging data increases the transverse information of the continuous drilling sand body of the horizontal section, and greatly improves the accuracy of the inter-well sand body prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas field development, in particular to a thin reservoir horizontal well geosteering method based on real-time adjustment of a three-dimensional model. BACKGROUND

[0002] The braided river delta facies is the main sedimentary facies in the dense sandstone gas reservoir, and in the actual production process, the horizontal well development method is adopted. In the implementation process of the horizontal well, it is inevitable to drill through the argillaceous interlayer. The complex interlayer distribution is an important factor affecting the deployment of the horizontal well. However, due to the high cost of the horizontal well drilling, when a relatively thick argillaceous interlayer is drilled, the contact range between the wellbore and the reservoir is reduced, the utilization rate of the horizontal well section is reduced, and the overall development effect of the horizontal well is restricted.

[0003] In order to realize the guidance of the bit travel, the data about the wellbore need to be collected, and the bit travel guidance system is established by analyzing and utilizing the data.

[0004] However, in the actual drilling process of the dense sandstone coal-bearing stratum and the coalbed methane reservoir, considering the stability of the drilling, the horizontal section is not measured or rarely measured, and only the logging data is recorded in the horizontal section. The logging and logging data are considered to be not corresponding because of different instruments and different basic environmental parameters. In the actual practical process, only the logging data is often used, and the measured data such as the natural gamma curve, lithology and total hydrocarbon of the logging are not considered.

[0005] Even though there is prior art in the market about bit travel direction, such as a Chinese patent with patent number CN114165160A discloses a fast geosteering method based on fine grid storage and one-dimensional function, the method comprises the following steps: obtaining real-time logging while drilling gamma logging drilling curve, weighting to obtain logging while drilling gamma logging simulation curve in one-dimensional grid, adjusting the geosteering model and the drilling direction of the bit in real time by comparing the logging while drilling gamma simulation curve and the logging while drilling gamma logging drilling curve, and realizing the fast geosteering of logging while drilling. This method only uses the logging while drilling gamma curve of logging, does not consider the logging while drilling gamma curve, logging curve and seismic lithology inversion body of the surrounding adjacent well, and adopts an equal proportion weighting method for interwell prediction, the algorithm is single, the interwell prediction accuracy is not enough, and it is difficult to meet the research needs of fine reservoir characterization. The non-patent document "Application of geosteering technology in horizontal well geosteering" provides a method for establishing a natural gamma attribute model based on natural gamma logging, but this method only uses the natural gamma data of the logging while drilling data and the adjacent well comparison to determine whether the reservoir top structure can enter the target, lacks quantitative research on the logging natural gamma and total hydrocarbon, and cannot guarantee that the bit advances along the best trajectory, and cannot achieve the purpose of improving the horizontal section gas layer drilling rate and the overall benefit of drilling. SUMMARY

[0006] The purpose of the present application is to provide a thin reservoir horizontal well geosteering method based on real-time adjustment of a three-dimensional model, aiming to improve the problem of lack of effective bit travel direction guiding method.

[0007] The present application is implemented as follows: a thin reservoir horizontal well geosteering method based on real-time adjustment of a three-dimensional model, which predicts the lithology of interwell sand bodies and horizontal well geosteering by normalizing the gamma curves of logging and logging, comprising the following steps

[0008] Step one: normalize the horizontal well logging while drilling natural gamma curve data obtained from logging, and name it GRL;

[0009] Step two: normalize the logging natural gamma curve according to the natural gamma curve mudstone baseline value, and name it GRC;

[0010] Step three: data check is performed on the GRL and GRC curves of the same horizontal well, and the two curves of the checked horizontal well are spliced into GRU;

[0011] Step four: use neural network algorithm to explore the correlation between normalized natural gamma curve and lithology and total hydrocarbon, and obtain the natural gamma range corresponding to medium-coarse sandstone, fine siltstone and mudstone, and the total hydrocarbon range;

[0012] Step five: using GRU curve and total hydrocarbon as well point hard data, using seismic inversion data as soft data to constrain between wells, carrying out sequential Gaussian indicator simulation, establishing three-dimensional natural gamma body, three-dimensional lithology data body and three-dimensional total hydrocarbon data body;

[0013] Step six: using natural gamma body, lithology body and total hydrocarbon data body in new well drilling tracking adjustment process, drilling head penetrates in the best advancing track.

[0014] Preferably, in step one, the mudstone baseline is found one by one, and the mudstone baseline value is used for normalization calculation of mud logging while drilling GR curve.

[0015] Preferably, when the mud logging while drilling GR curve is normalized and calculated, the minimum value and the maximum value are in the interval [0, 1], and the curve result conforms to the normal distribution.

[0016] Preferably, in step two, different mudstone baseline values are selected for different production layer positions of horizontal wells for classification and normalization, wherein the minimum value and the maximum value are in the interval [0, 1], and the curve result conforms to the normal distribution.

[0017] Preferably, in step three, the coal line is specially marked in lithofacies interpretation by referring to curves such as acoustic wave and neutron.

[0018] Preferably, the starting points of well logging and mud logging are different, the data at the splicing position should be checked, according to the lithofacies interpretation, the abnormal value and invalid value should be removed, and the ambiguous value should be modified.

[0019] Preferably, in step four, correlation analysis is carried out step by step in classification, including the following steps

[0020] S1: correlation analysis is carried out on sandstone, coal line, mudstone and GRU and total hydrocarbon, to obtain the natural gamma range corresponding to sandstone and mudstone and the total hydrocarbon range;

[0021] S2: correlation analysis is carried out on medium-coarse sandstone and fine siltstone in the sandstone of the horizontal section and GRU and total hydrocarbon, to find the boundary line of medium-coarse sandstone and fine siltstone in the natural gamma range corresponding to sandstone and the total hydrocarbon range, and to obtain the natural gamma range corresponding to medium-coarse sandstone, fine siltstone and mudstone and the total hydrocarbon range.

[0022] Preferably, when correlation analysis is carried out, the correlation degree greater than 0.7 is considered reliable.

[0023] Preferably, the total hydrocarbon of mudstone is less than 1%, and the natural gamma value is greater than 80 API, the total hydrocarbon of coal seam and sandstone is greater than 1%, and the natural gamma value is less than 80 API; the total hydrocarbon of medium-coarse sandstone is between 0 and 100, the natural gamma value is less than 60 API, and is often greater than 20 API, and the total hydrocarbon of fine siltstone is less than 2%, and the natural gamma value is between 40 and 80 API.

[0024] The three-dimensional natural gamma ray body, the total hydrocarbon body and the lithology body are re-calculated by adding the measured curve of the horizontal section into the three-dimensional natural gamma ray body, the total hydrocarbon body and the lithology body through the sequential Gaussian indicator simulation calculation.

[0025] Compared with the prior art, the method has the advantages of: ① the normalization processing improves the accuracy of the logging data; ② the matching of the logging data increases the lateral information of the sand body drilled in the horizontal section, and greatly improves the accuracy of the interwell sand body prediction; ③ the spatial position of the horizontal section is judged in real time, the drilling direction of the drill bit is optimized in real time, the gas layer drilling rate is ensured to be high, and the overall development effect of the horizontal well is improved. The method effectively provides a real-time correction of the geosteering idea based on the three-dimensional geological model, and provides a basis for oil and gas resource development and comprehensive adjustment and enhanced oil recovery. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 Technical flow chart for establishing the natural gamma ray body and the total hydrocarbon body to guide the geosteering

[0027] Figure 2 Contrast diagram of the logging GRL, the logging GRC and the spliced GRU

[0028] Figure 3 Schematic diagram of the correlation between the natural gamma ray and the total hydrocarbon of different lithologies

[0029] Figure 4 Schematic diagram of the contrast between the predicted three-dimensional natural gamma ray body and the measured curve. DETAILED DESCRIPTION

[0030] In the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "linking", "fixing" and the like should be understood in a broad sense, for example, can be fixed connection, can be detachable connection, or can be integrated; can be mechanical connection, or can be electrical connection; can be directly connected, or can be indirectly connected through an intermediate medium; can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0031] The present application will be further described below in combination with the drawings and specific embodiments:

[0032] The GRU is spliced by normalizing the GR of the mud logging while drilling in the horizontal section and the GR of the well logging in the build-up section for 22 horizontal wells. The correlation between the GRU of different lithology and the total hydrocarbon is established, and the value range of the GRU and the total hydrocarbon of the medium-coarse sandstone and the fine siltstone is determined. Under the constraint of the seismic inversion data body, the three-dimensional GRU, total hydrocarbon and lithology body are established by using the sequential Gaussian indicator simulation. If the drilled data is normalized and imported for recalculation when drilling to the mudstone or no display (total hydrocarbon <1m) sandstone for 50m, the next adjustment direction is determined.

[0033] The specific steps are as follows:

[0034] 1) The mud logging while drilling natural gamma curve data obtained by mud logging is normalized and named as GRL.

[0035] The mud logging while drilling gamma curve of the horizontal section of the preferred 22 horizontal wells is normalized based on the mudstone baseline value, wherein the minimum value and the maximum value are in the [0, 1] interval, and the curve result conforms to the normal distribution. The GRL result of the mud logging natural gamma normalization is shown in FIG. Figure 2 a.

[0036] 2) The well logging natural gamma curve data obtained by well logging is normalized and named as GRC.

[0037] According to the literature and the local empirical formula, the formation natural gamma mudstone baseline should be selected at 140-150 API. There are 5 small layers including Shan 2-2, Shan 2-1, Shan 1-3, Shan 1-2 and Shan 1-1, and 10 sets of single sand bodies. Considering the different production horizons of each small layer, different mudstone baseline values are selected for the different production horizons of the horizontal wells for classification and normalization. The natural gamma curve of the guide eye well or the A target point of the 22 horizontal wells is normalized according to the different horizons, the mudstone baseline value of Shan 2-2 and Shan 2-1 small layer is selected as 140 API, and the mudstone baseline value of Shan 1-3, Shan 1-2 and Shan 1-1 is selected as 150 API. The minimum value and the maximum value are in the [0, 1] interval, and the curve result conforms to the normal distribution. The GRC result of the mud logging natural gamma normalization is shown in FIG. Figure 2 b.

[0038] 3) The GRL and GRC curves of the 22 horizontal wells are checked, and the two curves of the horizontal wells after checking are spliced as GRU.

[0039] In this embodiment, the GRL and GRC curves of the same horizontal well are checked, and the two curves of the horizontal well after checking are spliced as GRU (as shown in FIG. Figure 2c) should be marked in lithofacies interpretation. Because of the different starting points of well logging and well logging, the data at the splicing point should be checked. According to the lithofacies interpretation, abnormal values, invalid values should be removed, and ambiguous values should be modified.

[0040] 4) The correlation between normalized GR curve and lithology, total hydrocarbon is explored by using neural network algorithm, and the GR range and total hydrocarbon range corresponding to medium-coarse sandstone, fine siltstone and mudstone are obtained.

[0041] In this embodiment, the correlation between GRU and lithology, total hydrocarbon is analyzed, and the correlation between the classified and graded part of the straight well without medium-coarse sandstone and fine siltstone in the lithology curve and GRU and total hydrocarbon is analyzed. The specific steps include: ①The part of sandstone, coal line and mudstone is correlated with GRU and total hydrocarbon, and the correlation degree greater than 0.7 is considered reliable, and the GR range and total hydrocarbon range corresponding to sandstone and mudstone are obtained. For this example, the total hydrocarbon of mudstone is less than 1%, and the GR value is greater than 80 (the normalized value is 0.57), and the total hydrocarbon of coal seam and sandstone is greater than 1%, and the GR value is less than 80. ②The correlation between medium-coarse sandstone and fine siltstone in the sandstone of the horizontal section and GRU and total hydrocarbon is analyzed, and the boundary line of medium-coarse sandstone and fine siltstone is found in the GR range and total hydrocarbon range corresponding to sandstone, and the GR range and total hydrocarbon range corresponding to medium-coarse sandstone, fine siltstone and mudstone are obtained. For this example, the total hydrocarbon of medium-coarse sandstone is between 0 and 100, the GR value is less than 60 API (the normalized value is 0.43), and is often greater than 20 API (the normalized value is 0.14), and the total hydrocarbon of fine siltstone is less than 2%, and the GR value is between 40 and 80 API (for example Figure 3 )。

[0042] 5) The GRU curve and total hydrocarbon are used as well point hard data, and the seismic inversion data body is used as soft data to constrain the well-to-well, and the sequential Gaussian indicator simulation is carried out to establish three-dimensional GR body, lithology body and total hydrocarbon body Figure 4 a-c)。

[0043] In this embodiment, the lithology body is obtained by using the GR curve of well logging under the constraint of wave impedance, and the GRU curve and total hydrocarbon are used as well point hard data, and the lithology inversion body is used as soft data to constrain the prediction of GR value between wells. The correlation between the inversion body and single well should be greater than 60%, the sequential Gaussian indicator simulation algorithm is used to establish three-dimensional GR body. The root mean square of total hydrocarbon of the horizontal section is calculated as the grid attribute, and the lithofacies body and GR body are used as soft data to constrain the well-to-well, and the sequential Gaussian indicator simulation algorithm is used to obtain three-dimensional total hydrocarbon body.

[0044] 6) Natural gamma bodies and lithological bodies are used in the new well drilling tracking and adjustment process. The main method is to add the drilling results into the model, update the data in real time, and predict the lithology between wells.

[0045] In this embodiment, during the actual drilling tracking process, the degree of agreement between the predicted results of the three-dimensional natural gamma ray body and the measured curve is compared. If they do not agree, it indicates that the data measurement accuracy is insufficient and there is uncertainty in the inter-well sand bodies. The process returns to step 1, adding the measured curve of the horizontal segment to the three-dimensional natural gamma ray body, total hydrocarbon body, and lithological body (such as...). Figure 4 In the ac), the sequential Gaussian indicator co-simulation calculation is re-performed. Through continuous iterative analysis, the directional formation model and the drilling direction of the drill bit are adjusted in real time, realizing rapid geological guidance for logging while drilling.

[0046] This invention establishes a geological steering method for thin reservoir horizontal wells based on real-time adjustment using a three-dimensional model, which has the following characteristics: ① In cases of data mismatch between logging and well logging methods, normalization processing improves the accuracy of logging data; ② For wells lacking logging data in the horizontal section, matching logging data adds lateral information about continuously encountered sand bodies in the horizontal section, significantly improving the accuracy of inter-well sand body prediction; ③ Real-time judgment of the spatial position of the horizontal section allows for continuous optimization of the drill bit's drilling direction, ensuring a high gas layer encounter rate and improving the overall development effect of horizontal wells. This method effectively provides a strategy for real-time correction of geological steering based on a three-dimensional geological model, providing a basis for oil and gas resource development and comprehensive adjustment, and improving recovery rates.

[0047] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for real-time adjustment of horizontal well geosteering in thin reservoirs based on three-dimensional models, characterized in that, The lithology of interwell sand body is predicted and horizontal well geosteering by normalizing the gamma curves of well logging and logging, including the following steps Step one: the natural gamma curve data of horizontal well drilling is normalized from the logging, named GRL; Step two: the logging natural gamma curve is normalized according to the natural gamma curve mudstone baseline value, named GRC; Step three: the GRL and GRC curves of the same horizontal well are checked, and the two curves of the horizontal well after checking are spliced into GRU; Step four: the correlation between normalized natural gamma curve and lithology and total hydrocarbon is explored by using neural network algorithm, and the corresponding natural gamma range of medium-coarse sandstone, fine siltstone and mudstone and the total hydrocarbon range are obtained; Step five: the GRU curve and total hydrocarbon are used as well point hard data, and the seismic inversion data body is used as soft data to constrain interwell, and sequential Gaussian indicator simulation is carried out to establish three-dimensional natural gamma body, three-dimensional lithology data body and three-dimensional total hydrocarbon data body; Step six: the natural gamma body, lithology body and total hydrocarbon data body are used in the real drilling tracking process of new well, and the drill bit penetrates with the best advancing trajectory.

2. The method of claim 1, wherein, In step one, the mudstone baseline is found one by one, and the mudstone baseline value is used for normalization calculation of the GR curve of logging while drilling.

3. The method of claim 2, wherein, When the GR curve of logging while drilling is normalized, the minimum and maximum values are in the interval [0, 1], and the curve result conforms to the normal distribution.

4. The method of claim 1, wherein, In step two, different mudstone baseline values are selected for different production layer horizontal wells for classification and normalization, and the minimum and maximum values are in the interval [0, 1], and the curve result conforms to the normal distribution.

5. The method of claim 1, wherein, In step three, the coal line is specially marked in lithofacies interpretation by referring to curves such as acoustic wave and neutron.

6. The method of claim 5, wherein, The data at the splicing position should be checked according to the lithofacies interpretation, and the abnormal value and invalid value should be removed, and the ambiguous value should be modified.

7. The method of claim 1, wherein, In step four, correlation analysis is carried out step by step, including the following steps S1: correlation analysis is carried out between sandstone, coal line and mudstone and GRU and total hydrocarbon, and the corresponding natural gamma range of sandstone and mudstone and the total hydrocarbon range are obtained; S2: correlation analysis is carried out between medium-coarse sandstone and fine siltstone in the horizontal section sandstone and GRU and total hydrocarbon, and the boundary line of medium-coarse sandstone and fine siltstone is found in the corresponding natural gamma range of sandstone and the total hydrocarbon range, and the corresponding natural gamma range of medium-coarse sandstone, fine siltstone and mudstone and the total hydrocarbon range are obtained.

8. The method of claim 7, wherein, In the correlation analysis, the correlation degree greater than 0.7 is considered reliable.

9. The method of claim 8, wherein, The total hydrocarbon of mudstone is less than 1%, the natural gamma value is greater than 80 API, the total hydrocarbon of coal seam and sandstone is greater than 1%, and the natural gamma value is less than 80 API; the total hydrocarbon of medium-coarse sandstone is between 0-100, the natural gamma value is less than 60 API, and often greater than 20 API, and the total hydrocarbon of fine siltstone is less than 2%, and the natural gamma value is between 40-80 API.

10. The method of claim 1, wherein, In the real drilling tracking process, the coincidence degree of the predicted results of three-dimensional natural gamma body and the measured curve is compared; if it does not coincide, the measured curve of the horizontal section is added to the three-dimensional natural gamma body, total hydrocarbon body and lithology body, and the sequential Gaussian indicator simulation calculation is carried out again.

Images