Method and apparatus for predicting hydrocarbon production

Abstract

The application discloses a kind of oil and gas production prediction method and device.Therein, the method includes: obtaining the characteristic data of the core collected from target reservoir, obtaining the percolation data in target reservoir according to the core measured, obtaining the oil-bearing data in target reservoir according to the core measured, engineering data for oil and gas exploitation of target reservoir;The oil and gas production in target reservoir is predicted by spectral clustering model according to characteristic data, percolation data, oil-bearing data and engineering data, wherein the spectral clustering model is obtained by model training according to sample core corresponding sample data, and sample data at least includes: sample core corresponding sample characteristic data, sample percolation data, sample oil-bearing data, sample engineering data, sample oil and gas production data.The application solves the technical problems that the prediction of oil and gas production in related technologies is time-consuming and high-cost.

Description

Technical Field

[0001] This invention relates to the field of oil and gas extraction technology, and more specifically, to a method and apparatus for predicting oil and gas production. Background Technology

[0002] In the field of oil and gas extraction, when making development decisions for multiple reservoirs, it is necessary to combine basic reservoir information to determine how to conduct oil and gas extraction and to identify priority development areas. Oil and gas production directly reflects the oil reserves in the reservoir; therefore, oil and gas production plays a crucial role in oil and gas extraction decisions.

[0003] However, before oil and gas development, the amount of oil and gas stored in the reservoir, especially the oil and gas production capacity, is generally unknown. A complete core testing project is also time-consuming, which is not conducive to the rapid assessment by oil and gas developers. Well testing, which studies the various physical parameters, production capacity, and connectivity between oil, gas and water layers and the test well by testing the production dynamics of oil and gas wells, is a traditional method for assessing production capacity. However, it is costly and must be used in conjunction with other methods such as core analysis to comprehensively assess the current reservoir.

[0004] There is currently no effective solution to the problem that predicting oil and gas production in the aforementioned technologies is time-consuming and costly. Summary of the Invention

[0005] This invention provides a method and apparatus for predicting oil and gas production, which at least solves the technical problem that predicting oil and gas production is time-consuming and costly in related technologies.

[0006] According to one aspect of the present invention, a method for predicting oil and gas production is provided, comprising: acquiring characteristic data of core samples collected from a target reservoir; obtaining seepage data of the target reservoir measured from the core samples; obtaining oil-bearing data of the target reservoir measured from the core samples; and engineering data for oil and gas extraction of the target reservoir; and predicting the oil and gas production in the target reservoir using a spectral clustering model based on the characteristic data, the seepage data, the oil-bearing data, and the engineering data, wherein the spectral clustering model is obtained by training a model based on sample data corresponding to sample core samples, and the sample data includes at least: sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core samples.

[0007] Optionally, before acquiring the characteristic data of the core samples collected from the target reservoir, the seepage data of the target reservoir measured from the core samples, the oil-bearing data of the target reservoir measured from the core samples, and the engineering data to be used for oil and gas extraction from the target reservoir, the method further includes: acquiring training data corresponding to the sample core samples; predicting the predicted oil and gas production of the reservoir from which the sample core samples originate by using a pre-built spectral clustering model based on the sample seepage data, sample oil-bearing data, and sample engineering data in the training data; adjusting the parameters of the pre-built spectral clustering model based on the predicted oil and gas production and the actual oil and gas production calculated based on the sample oil and gas production data in the training data, so that the distance between the predicted oil and gas production and the actual oil and gas production meets a preset condition; and when the training termination condition is reached, using the pre-built spectral clustering model with adjusted parameters as the spectral clustering model.

[0008] Optionally, before acquiring the training data corresponding to the sample cores, the method further includes: acquiring the initial training data corresponding to each collected sample core; preprocessing the initial training data corresponding to each sample core to obtain the training data corresponding to each sample core.

[0009] Optionally, preprocessing the initial training data corresponding to each sample core to obtain the training data corresponding to each sample core includes: denoising the initial training data corresponding to each sample core; normalizing each parameter in the denoised initial training data to obtain normalized data; calculating the correlation coefficient between pairs of parameters in the normalized data corresponding to each sample core; and filtering the normalized data based on the correlation coefficient to obtain the training data corresponding to the sample core.

[0010] Optionally, filtering the normalized data based on the correlation coefficient to obtain the training data corresponding to the sample core includes: determining parameters in the reference operation data whose correlation coefficient with the parameters in the target data is lower than a first set threshold, and determining parameters in the reference operation data whose correlation coefficient between two parameters is higher than a second set threshold; filtering one of the parameters in the reference operation data whose correlation coefficient with the parameters in the target data is lower than the first set threshold, and filtering one of the parameters in the reference operation data whose correlation coefficient between two parameters is higher than the second set threshold, and using the filtered data as the training data corresponding to the sample core.

[0011] Optionally, the feature data includes at least feature parameters describing the core characteristics corresponding to the core, the seepage data includes at least seepage parameters describing the pore seepage performance of the target reservoir, and the oil-bearing data includes at least oil-bearing parameters describing the oil and gas situation in the target reservoir.

[0012] Optionally, after predicting the oil and gas production in the target reservoir using a spectral clustering model based on the feature data, the seepage data, the oil-bearing data, and the engineering data, the method further includes: determining the reservoir classification corresponding to the target reservoir based on the predicted oil and gas production.

[0013] According to another aspect of the present invention, an oil and gas production prediction device is also provided, comprising: a first acquisition module, configured to acquire characteristic data of core samples collected from a target reservoir, seepage data of the target reservoir measured from the core samples, oil-bearing data of the target reservoir measured from the core samples, and engineering data for oil and gas extraction from the target reservoir; and a first prediction module, configured to predict the oil and gas production in the target reservoir based on the characteristic data, the seepage data, the oil-bearing data, and the engineering data using a spectral clustering model, wherein the spectral clustering model is obtained by training a model based on sample data corresponding to sample core samples, and the sample data includes at least: sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core samples.

[0014] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the oil and gas production prediction method described in any one of the above.

[0015] According to another aspect of the present invention, a processor is also provided, the processor being configured to run a program, wherein the program, when running, executes the oil and gas production prediction method described in any of the foregoing embodiments.

[0016] In this embodiment of the invention, the method involves acquiring characteristic data from core samples collected from a target reservoir, obtaining seepage data from the core samples, obtaining oil-bearing data from the core samples, and engineering data for oil and gas extraction from the target reservoir. A spectral clustering model is then used to predict the oil and gas production in the target reservoir based on the characteristic data, seepage data, oil-bearing data, and engineering data. The spectral clustering model is trained using sample data corresponding to the sample core samples. This sample data includes at least: sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core samples. By using the spectral clustering model to predict the oil and gas production in the target reservoir based on the characteristic data, seepage data, oil-bearing data, and engineering data from the core samples, the method achieves the goal of rapidly and accurately predicting oil and gas production. This improves the speed of oil and gas production prediction and reduces the operational cost of oil and gas production prediction, thereby solving the technical problem of time-consuming and costly oil and gas production prediction in related technologies. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0018] Figure 1 This is a flowchart of an oil and gas production prediction method according to an embodiment of the present invention;

[0019] Figure 2 This is a schematic diagram of an oil and gas production prediction device according to an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0022] Example 1

[0023] According to an embodiment of the present invention, an embodiment of a method for predicting oil and gas production is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0024] Figure 1 This is a flowchart of an oil and gas production prediction method according to an embodiment of the present invention, such as... Figure 1 As shown, the method for predicting oil and gas production includes the following steps:

[0025] Step S102: Obtain characteristic data from the core collected from the target reservoir, seepage data from the target reservoir measured from the core, oil-bearing data from the target reservoir measured from the core, and engineering data to be used for oil and gas extraction from the target reservoir.

[0026] To understand the geological or mineral conditions of a reservoir, it is necessary to first collect rock cores from the reservoir for analysis. The analytical data from these cores serves as in-situ geological data for analyzing the reservoir's condition. In other words, a rock core is a sample of rock collected from the target reservoir.

[0027] As an optional embodiment, in order to estimate the oil and gas production in a target reservoir, a core sample is first collected from the target reservoir, and the oil and gas production in the target reservoir is predicted using the analysis data of the core sample. Here, "target reservoir" does not refer to a specific reservoir, but rather to any reservoir for which oil and gas production is to be predicted.

[0028] The characteristic data of the core can be basic information about the core, such as the depth of the target reservoir from which the core originated, the pressure gradient at the target reservoir, or information describing the pore structure in the core.

[0029] It should be noted that the aforementioned core includes pores and a framework. If the target reservoir from which the core originates contains oil and gas resources, these resources actually exist in the pores of the target reservoir. Therefore, the porosity of the target reservoir can be understood by analyzing the core.

[0030] In specific embodiments, the aforementioned characteristic parameters include, but are not limited to, depth, pressure gradient, core type, core density, cementation type, clay content, principal grain size, sorting coefficient, kurtosis, skewness, median pore throat radius, maximum pore throat radius, average capillary radius, pore throat volume ratio, median capillary pressure, displacement pressure, and mercury removal efficiency.

[0031] Here, depth refers to the geological depth of the target reservoir from which the core was derived. The pressure gradient corresponding to the core here refers to the pressure gradient at the depth of the target reservoir from which the core was derived. The pressure gradient is characterized by the pressure change per unit distance along the direction of fluid flow.

[0032] Core type refers to the classification to which the core belongs according to the defined type classification. Core types include cemented clastic rocks, carbonates, loose sandstones, karst carbonate rocks, clay-bearing rocks, low-permeability rocks, shale, oil shale, etc. Of course, the core types listed above are merely illustrative examples and should not be considered as limiting the scope of use of this disclosure.

[0033] Optionally, the classification criteria corresponding to the core type can be based on the composition of the core. Alternatively, the classification criteria corresponding to the core type can also be based on grain size.

[0034] The core type can be determined by analyzing the composition of the collected core or by further observing its microstructure.

[0035] Cementation type refers to the contact relationship between cement or interstitial material and clastic particles in a rock core. Specifically, cementation types include basal cementation, porous cementation, contact cementation, and mosaic cementation. Basal cementation is characterized by no contact between rock particles; porous cementation is characterized by point contact between particles, with the cement distributed within the pores; contact cementation is characterized by point-to-line contact between particles, with the cement distributed at the contact points; and mosaic cementation is characterized by line or concave contact between particles.

[0036] Cement refers to the minerals that form in the intergranular spaces of rock cores through chemical precipitation. Cement includes siliceous, carbonate, and some ferrous materials. Therefore, according to the type of cement, cement types can also include argillaceous cement, calcareous cement, siliceous cement, and ferromanganese cement.

[0037] For core samples, different cementation types result in different mechanical properties, which in turn may lead to differences in the oil and gas storage conditions in the target reservoir from which core samples with different cementation types originate.

[0038] In a specific embodiment, the cementation type corresponding to the core can be determined by microscopic observation or composition analysis of the core.

[0039] For core samples originating from the target reservoir, these core samples are formed from rock and mineral fragments resulting from mechanical weathering. These fragments are then transported, deposited, compacted, and cemented together; therefore, the rock is essentially clastic rock. Consequently, the internal framework of this core sample is composed of numerous clastic particles.

[0040] To determine the principal grain size, kurtosis, skewness, and sorting factor of the core sample, grain size statistics were performed on the clastic particles within the core. Grain size refers to the size of the grain, typically expressed as diameter.

[0041] In specific embodiments, since the actual particles have diverse shapes, such as non-spherical debris particles, the particle size of debris particles can be characterized by equivalent particle size. Equivalent particle size refers to the diameter of an actual particle when a certain physical property of a particle is the same as or similar to that of a homogeneous spherical particle.

[0042] After measuring the particle size of each detrital grain in the core, a cumulative grain size curve for the core is constructed based on the particle size of each detrital grain. The principal grain size, kurtosis, skewness, and sorting coefficient are then determined based on this cumulative grain size curve.

[0043] The main grain size refers to the grain size corresponding to the cumulative grain size distribution number reaching a set proportion on the cumulative grain size curve of the core sample. This set proportion can be selected according to actual needs, for example, 50%, 60%, 75%, etc., without specific limitations here.

[0044] The sorting coefficient refers to the ratio of the diameter of the debris particles corresponding to 75% and 25% on the cumulative particle size distribution curve. This sorting coefficient serves as a reference for the sorting of the core. When the core is well sorted, the values ​​of P25 (i.e., the particle size corresponding to 25% of the cumulative particle size distribution) and P75 (i.e., the particle size corresponding to 75% of the cumulative particle size distribution) are very close, and the closer they are to 1, the greater the value will be.

[0045] Kurtosis, also known as the kurtosis coefficient, refers to the characteristic number of peak values ​​at the average grain size on the cumulative grain size curve. Kurtosis reflects the sharpness of the peak. The kurtosis of a core sample can be calculated using the following formula: Where X represents the particle size of the clastic particles in the rock core, μ is the average particle size, and σ is the standard deviation.

[0046] The skewness corresponding to the core is a characteristic number used to characterize the degree of asymmetry between the cumulative grain size curve and the average grain size. The skewness corresponding to the core can be calculated using the following formula:

[0047] In a rock core, besides solid phases such as clastic particles and cement, there are also numerous pores. The oil and gas in the target reservoir where the core is located actually exist within these pores. The distribution of pores in the core can approximately reflect the pore distribution in the target reservoir, and thus, the distribution of oil and gas in the target reservoir. Median pore throat radius, maximum pore throat radius, pore throat volume ratio, and average capillary radius can all serve as characteristic parameters reflecting the pore distribution in the core.

[0048] The pore throat radius, also known as the pore throat channel radius, is measured by the radius of the largest sphere that can pass through the pore throat. It is understood that a rock core contains several pore throats. To understand the overall distribution of pores in a rock core, the median and maximum pore throat radii are generally used to reflect the pore distribution.

[0049] The pore-throat volume ratio refers to the ratio of the pore volume to the throat volume in a rock core.

[0050] The median capillary pressure refers to the capillary pressure value corresponding to a mercury saturation of 50%. This median capillary pressure is obtained by measuring the capillary pressure curve of a core. For a core, the capillary pressure is determined by the curvature of the unwetted phase surface, which is related to the size of the pore throat. Furthermore, the capillary pressure increases as the wetted phase saturation decreases. The median capillary pressure corresponding to a core can be measured using methods such as mercury intrusion porosimetry and centrifuge analysis.

[0051] Displacement pressure refers to the minimum pressure required for the non-wetting phase (mercury) to begin entering the core. It is the starting pressure required for mercury to enter the largest connecting pore throat in the core and form a continuous flow. Under displacement pressure, the radius of the pore throat into which mercury can enter is the radius of the largest pore throat in the core.

[0052] If mercury intrusion porosimetry is used to measure the median capillary pressure and the displacement pressure, it involves the processes of mercury intrusion and removal into the pore throat of the core. An important characteristic parameter of the removal process is the removal efficiency. The removal efficiency refers to the ratio of the volume of mercury removed from the core to the volume of mercury injected during the intrusion process when the pressure decreases from the maximum to the minimum after the intrusion process is completed.

[0053] The above only lists characteristic parameters that reflect the porosity in the core, such as median pore throat radius, maximum pore throat radius, median capillary pressure, pore throat volume ratio, displacement pressure, and mercury removal efficiency. In other embodiments, parameters characterizing pore throat sorting characteristics (e.g., pore throat sorting coefficient, pore throat peak state, mean coefficient) and parameters characterizing pore throat connectivity and controlling fluid flow characteristics (e.g., pore throat coordination number, pore tortuosity, etc.) can be further selected as characteristic parameters.

[0054] The characteristic parameters listed above that characterize the pore structure in the rock core can be measured by methods such as mercury intrusion porosimetry, dynamic displacement, and centrifuge. Alternatively, they can be measured by constructing a digital rock core as a model of the pore structure in the rock core and conducting analysis and simulation experiments on the digital rock core. No specific limitations are imposed here.

[0055] For the paint stored in the target reservoir, it is constantly flowing within the pores. The seepage parameters can reflect the seepage performance of the pores in the target reservoir. In a specific embodiment, the seepage parameters can be at least one of porosity and permeability.

[0056] Porosity refers to the ratio of pore volume to solid volume in a rock core. Porosity is a parameter that measures the storage capacity of a rock core.

[0057] Permeability is the ability of a rock core to allow fluid to pass through it. Permeability characterizes the core's capacity to conduct liquids. In practice, permeability can be at least one of absolute permeability, effective permeability, or relative permeability; no specific limitation is made here.

[0058] Oil-bearing parameters are used to describe the oil content in the target reservoir. In practice, oil-bearing parameters include, but are not limited to, oil saturation, free fluid saturation, bound water saturation, crude oil viscosity, and parameters indicating whether wax is present.

[0059] Oil saturation refers to the ratio of the volume of pores occupied by crude oil in an oil reservoir to the total pore volume. Free fluid saturation refers to the ratio of the volume of pores containing free fluid in an oil reservoir to the total pore volume. Free fluid can be water, oil, or gas. If a reservoir contains water, oil, and gas simultaneously, then the ratio of the volume of pores occupied by water, oil, and gas to the total pore volume is the free fluid saturation of that reservoir.

[0060] Extensive analysis of core samples from the field reveals that even pure oil and gas reservoirs contain a certain amount of stagnant water, referred to as bound water. The presence of bound water is related to the reservoir's formation process. Different reservoirs exhibit varying degrees of bound water saturation due to differences in rock and fluid properties and varying hydrocarbon migration conditions. Bound water saturation refers to the ratio of the volume of pores occupied by bound water to the total pore volume in the target reservoir.

[0061] Among them, the oil saturation, free fluid saturation, and bound water saturation parameters mentioned above can be measured by atmospheric distillation, distillation drawer method, chromatography, and well logging method. The oil saturation and water saturation in the core can also be determined by relative permeability curves or capillary pressure curves. The oil and water saturation in the core are then used as the oil and water saturation in the target reservoir from which the core is derived.

[0062] Crude oil viscosity is a measure of the frictional resistance of a portion of crude oil flowing relative to another portion. Crude oil viscosity can be measured by collecting crude oil samples from a target reservoir (e.g., crude oil cores) and conducting physical experiments on those samples.

[0063] Engineering data may include one or more engineering parameters, such as total fracturing fluid volume, proppant type, proppant dosage, etc., which are not specifically limited here.

[0064] Fracturing fluid is a heterogeneous and unstable chemical system composed of various additives in a specific ratio. It is the working fluid used for fracturing oil and gas reservoirs. Its main function is to transfer the high pressure generated by surface equipment to the formation, causing the formation to fracture and transport proppant along the fractures.

[0065] The total fracturing fluid volume refers to the total amount of fracturing fluid intended for oil and gas extraction from the target reservoir.

[0066] Proppant, also known as fracturing proppant, is used in deep oil and gas well development. In high-closure-pressure, low-permeability deposits, fracturing treatment opens up oil and gas-bearing rock formations, allowing oil and gas to flow through the channels created by the fractures. At this point, fluid is injected into the rock base layer at pressures exceeding the formation's fracturing strength to create fractures in the surrounding rock layers, forming a channel with high-level flow capacity. To keep the fractures open and allow oil and gas products to flow smoothly, the fluid injected into the base layer is called proppant. Types of proppant include hard and brittle ceramic proppants and toughening proppants. The proppant dosage refers to the amount of proppant intended for oil and gas extraction from the target reservoir.

[0067] Step S104: The oil and gas production in the target reservoir is predicted by the spectral clustering model based on the feature data, seepage data, oil-bearing data and engineering data. The spectral clustering model is obtained by training the model based on the sample data corresponding to the sample core. The sample data includes at least the sample feature data, sample seepage data, sample oil-bearing data, sample engineering data and sample oil and gas production data corresponding to the sample core.

[0068] Sample characteristic data refers to the characteristic data corresponding to the sample core, which includes at least one characteristic parameter describing the characteristics of the corresponding core. See the examples above for details on these characteristic parameters.

[0069] Sample seepage data refers to the seepage data in the reservoir from which the core sample originates, measured based on the core sample. It includes at least one seepage parameter that describes the pore seepage performance in the reservoir from which the core sample originates, as detailed in the above list.

[0070] Oil-bearing data from a sample refers to the oil-bearing data of the reservoir from which the sample core originated, measured from the sample core. It includes at least one oil-bearing parameter describing the hydrocarbon content in the reservoir from which the sample core originated. See the examples above for details on oil-bearing parameters.

[0071] Sample engineering data refers to the engineering parameters used in the actual oil and gas extraction of the reservoir from which the sample core was obtained. The engineering parameters are detailed in the above list.

[0072] Sample oil and gas production data refers to the amount of oil and gas extracted from the reservoir sourced from the sample core, using the engineering parameters in the aforementioned sample engineering data. In a specific embodiment, the sample oil and gas production data may include daily oil production, reservoir thickness, and working hours (i.e., the length of time each day of extraction).

[0073] During training, the initial model uses the oil and gas production data from each sample as the target. It then clusters oil and gas production data based on sample feature data, seepage data, oil-bearing data, and engineering data to obtain a predicted oil and gas production. The model parameters are adjusted based on this predicted production and the actual production indicated by the sample oil and gas production data until the predicted production is sufficiently close to the actual production, for example, the predicted production is the same as the actual production. Then, the model training continues using the sample data corresponding to the next core sample until the training termination condition is met.

[0074] After obtaining the spectral clustering model through the above training process, the spectral clustering model has the ability to predict the oil and gas production in the reservoir based on feature data, seepage data, oil-bearing data and engineering data, and thus outputs the predicted oil and gas production.

[0075] Through the above process, the spectral clustering model obtained through training is used to estimate the oil and gas production in the target reservoir. The estimated oil and gas production is used as a decision-making reference for oil and gas exploitation. For example, priority development areas are determined based on the predicted oil and gas production.

[0076] In the specific implementation process, based on existing exploration and development well core data, formation test data, and historical oil and gas production data, big data clustering can be used to establish a correlation model to select the most critical physical property parameters that determine production performance. After the drilling of a new well is completed, these key physical property parameters are prioritized for testing through core testing. Using the correlation model established in the early stage, oil and gas production can be quickly predicted, providing a reference for the well's production designers to design reasonable supporting measures and reduce operating costs.

[0077] It should be noted that the above-mentioned big data oil and gas reserve prediction method based on reservoir experimental parameters can be applied to sandstone and conglomerate areas. Within a few working days after the completion of new well drilling, the reservoir type of the current well can be quickly determined using 8-10 key core test data, and the expected production capacity can be given.

[0078] Through the above steps, the oil and gas production in the target reservoir can be predicted based on the characteristic data, seepage data, oil-bearing data, and engineering data of the core in the target reservoir using a spectral clustering model. This achieves the goal of rapid and accurate prediction of oil and gas production, thereby improving the speed of oil and gas production prediction and reducing the operating cost of oil and gas production prediction. This solves the technical problem that the prediction of oil and gas production is time-consuming and costly in related technologies.

[0079] Optionally, before acquiring the characteristic data of the core samples collected from the target reservoir, the seepage data of the target reservoir measured from the core samples, the oil-bearing data of the target reservoir measured from the core samples, and the engineering data to be used for oil and gas extraction from the target reservoir, the method further includes: acquiring training data corresponding to the sample core samples; predicting the predicted oil and gas production of the reservoir from which the sample core samples originate by using a pre-built spectral clustering model based on the sample seepage data, sample oil-bearing data, and sample engineering data in the training data; adjusting the parameters of the pre-built spectral clustering model based on the predicted oil and gas production and the actual oil and gas production calculated based on the sample oil and gas production data in the training data, so that the distance between the predicted oil and gas production and the actual oil and gas production meets the preset conditions; and when the training termination condition is met, using the pre-built spectral clustering model with adjusted parameters as the spectral clustering model.

[0080] The preset condition could be that the predicted oil and gas production is equal to the actual oil and gas production, or that the distance between the predicted and actual oil and gas production meets a certain numerical range; no specific limitation is made here. In summary, during the training process, the parameters of the pre-built spectral clustering model are adjusted with the goal of making the predicted oil and gas production as close as possible to the actual production. After each parameter adjustment, the oil and gas production is predicted again using the pre-built spectral clustering model with the adjusted parameters, and it is calculated whether the predicted oil and gas production obtained this time meets the preset condition. If it does, the next sample data is used for training; if not, the above process is repeated.

[0081] The pre-constructed spectral clustering model can be built based on sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to several sample cores. It is worth noting that the sample cores used for the pre-constructed spectral clustering model are different from the sample cores used for training the pre-constructed spectral clustering model.

[0082] The training termination condition can be either the number of iterations or the prediction accuracy; no specific limitation is made here. Specifically, when the training termination condition is the number of iterations, training stops when the set number of iterations is reached, and the pre-built spectral clustering model with adjusted parameters is used as the final spectral clustering model. When the training termination condition is the prediction accuracy, after training the pre-built spectral clustering model for a period of time, the prediction accuracy of the model is calculated using test samples. If the prediction accuracy meets the set prediction accuracy requirements, training stops; otherwise, training continues. The test samples include sample feature data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to several sample cores.

[0083] Since the spectral clustering model is trained based on sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data obtained from actual oil and gas extraction, the accuracy of the oil and gas production predicted by the spectral clustering model can be guaranteed.

[0084] Optionally, before obtaining the training data corresponding to the sample cores, the method further includes: obtaining the initial training data corresponding to each sample core; preprocessing the initial training data corresponding to each sample core to obtain the training data corresponding to each sample core.

[0085] The preprocessing performed can include noise reduction, numerical mapping, normalization, filtering, etc., without specific limitations. By performing preprocessing, the training data can be directly identified by the pre-built spectral clustering model for direct training without further data processing.

[0086] Optionally, preprocessing the initial training data corresponding to each sample core to obtain the training data corresponding to each sample core includes: denoising the initial training data corresponding to each sample core; normalizing each parameter in the denoised initial training data to obtain normalized data; calculating the correlation coefficient between pairs of parameters in the normalized data corresponding to each sample core; and filtering the normalized data based on the correlation coefficient to obtain the training data corresponding to the sample core.

[0087] The normalization process involves calculating the average and standard deviation of each parameter in the denoised data and applying the formula... Transform the values ​​corresponding to each parameter, where, Let Y represent the average value of parameter Y, and σ represent the standard deviation of parameter Y. Thus, through the above transformation, parameter Y is transformed into Y', thereby achieving the normalization of parameter Y.

[0088] The correlation coefficient calculation involves calculating the correlation coefficients between two different parameters among the characteristic parameters, oil-bearing parameters, seepage parameters, engineering parameters, and oil and gas production parameters in the normalized data.

[0089] The formula for calculating the correlation coefficient is:

[0090] Where Corr(M,N) is the covariance of parameters M and N, and σ m Let σ be the variance of parameter M. n Let M be the variance of parameter N, and then calculate the correlation coefficient between parameters M and N according to the above formula. The calculated correlation coefficient ranges from [0, 1], where 0 indicates no correlation and 1 indicates a perfect linear correlation. The magnitude of the correlation between the two parameters is determined based on the calculated correlation coefficient.

[0091] Optionally, filtering the normalized data based on the correlation coefficient to obtain the training data corresponding to the sample core includes: determining parameters in the reference operation data whose correlation coefficient with the parameters in the target data is lower than a first set threshold, and determining parameters in the reference operation data whose correlation coefficient between two parameters is higher than a second set threshold; filtering one of the parameters in the reference operation data whose correlation coefficient with the parameters in the target data is lower than the first set threshold, and filtering one of the parameters in the reference operation data whose correlation coefficient between two parameters is higher than the second set threshold, and using the filtered data as the training data corresponding to the sample core.

[0092] As an optional embodiment, the normalized data includes actual oil and gas production and reference operating data, which are the normalized data other than actual oil and gas production.

[0093] The first and second threshold values ​​are set to determine the correlation between two parameters. Specifically, if the correlation coefficient between the two parameters is lower than the first threshold value, the correlation between the two parameters is determined to be low; if the correlation coefficient between the two parameters is higher than the second threshold value, the correlation between the two parameters is determined to be high.

[0094] If the correlation coefficient between a parameter and the actual oil and gas production in the reference operation data is lower than the first set threshold, it is determined that the parameter has a low correlation with the actual oil and gas production, which also indicates that the parameter does not contribute much to the prediction of oil and gas production. Therefore, the parameter is filtered out.

[0095] If the correlation coefficient of two parameters in the reference work data is higher than the second set threshold, it indicates that the correlation coefficient of the two parameters is high. One of the two parameters can be represented by the other parameter. Therefore, one of the two parameters can be removed.

[0096] Optionally, the feature data includes at least feature parameters describing the characteristics of the core corresponding to the core, the seepage data includes at least seepage parameters describing the pore seepage performance of the target reservoir, and the oil-bearing data includes at least oil-bearing parameters describing the oil and gas situation in the target reservoir.

[0097] Optionally, after predicting the oil and gas production in the target reservoir using a spectral clustering model based on feature data, seepage data, oil-bearing data, and engineering data, the method further includes: determining the reservoir classification corresponding to the target reservoir based on the predicted oil and gas production.

[0098] The reservoir classification is based on oil and gas production.

[0099] As an optional embodiment, reservoirs are divided into three categories based on oil and gas production. Oil and gas production is characterized by the daily production per meter. The reservoir classification includes Class I, Class II, and Class III. Class I reservoirs have a daily oil and gas production per meter of at least 2.2 tons; Class II reservoirs have a daily oil and gas production per meter (in tons) ranging from [0.6, 2.2); and Class III reservoirs have a daily oil and gas production per meter (in tons) ranging from [0, 0.6). Of course, in other embodiments, reservoir types can be classified according to actual needs. The above are merely illustrative examples and should not be considered as limiting the scope of this disclosure.

[0100] After determining the reservoir classification corresponding to the target reservoir, oil and gas extraction can be planned accordingly. For example, for the Class I reservoirs mentioned above, which have a high daily oil and gas production per meter and are economically viable areas, they can be prioritized for development. Therefore, the determined reservoir classification can serve as a reference for planned oil and gas extraction.

[0101] Example 2

[0102] According to another aspect of the present invention, an oil and gas production prediction device is also provided. Figure 2 This is a schematic diagram of an oil and gas production prediction device according to an embodiment of the present invention, such as... Figure 2As shown, the oil and gas production prediction device includes a first acquisition module 22 and a first prediction module 24. The oil and gas production prediction device will be described in detail below.

[0103] The first acquisition module 22 is used to acquire characteristic data of core samples collected from the target reservoir, seepage data of the target reservoir measured from the core samples, oil-bearing data of the target reservoir measured from the core samples, and engineering data to be used for oil and gas extraction from the target reservoir. The first prediction module 24 is connected to the first acquisition module 22 and is used to predict the oil and gas production in the target reservoir based on the characteristic data, seepage data, oil-bearing data, and engineering data through a spectral clustering model. The spectral clustering model is obtained by training the model based on the sample data corresponding to the sample core samples. The sample data includes at least: sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core samples.

[0104] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0105] It should be noted that the first acquisition module 22 and the first prediction module 24 mentioned above correspond to steps S102 to S104 in Embodiment 1. The examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules, as part of the device, can be executed in a computer system such as a set of computer-executable instructions.

[0106] As can be seen from the above embodiments of this application, the oil and gas production in the target reservoir is predicted by the spectral clustering model based on the characteristic data, seepage data, oil-bearing data and engineering data of the core in the target reservoir. This achieves the purpose of predicting oil and gas production quickly and accurately, thereby realizing the technical effect of improving the speed of predicting oil and gas production and reducing the operating cost of predicting oil and gas production. This solves the technical problem in related technologies that the prediction of oil and gas production is time-consuming and costly.

[0107] Optionally, before acquiring the characteristic data of the core samples collected from the target reservoir, the seepage data of the target reservoir measured from the core samples, the oil-bearing data of the target reservoir measured from the core samples, and the engineering data to be used for oil and gas extraction from the target reservoir, the above-mentioned device further includes: a second acquisition module for acquiring training data corresponding to the sample core samples; a second prediction module for predicting the predicted oil and gas production of the reservoir from which the sample core samples originate based on the sample seepage data, sample oil-bearing data, and sample engineering data in the training data using a pre-built spectral clustering model; an adjustment module for adjusting the parameters of the pre-built spectral clustering model based on the predicted oil and gas production and the actual oil and gas production calculated based on the sample oil and gas production data in the training data, so that the distance between the predicted oil and gas production and the actual oil and gas production meets a preset condition; and a processing module for using the pre-built spectral clustering model with adjusted parameters as the spectral clustering model when the training end condition is met.

[0108] Optionally, before acquiring the training data corresponding to the sample cores, the above device further includes: a third acquisition module, used to acquire the initial training data corresponding to each sample core; and a preprocessing module, used to preprocess the initial training data corresponding to each sample core to obtain the training data corresponding to each sample core.

[0109] Optionally, the preprocessing module includes: a noise reduction unit for denoising the initial training data corresponding to each sample core; a normalization unit for normalizing the parameters in the denoised initial training data to obtain normalized data; a calculation unit for calculating the correlation coefficient between pairs of parameters in the normalized data corresponding to each sample core; and a filtering unit for filtering the normalized data based on the correlation coefficient to obtain the training data corresponding to the sample core.

[0110] Optionally, the filtering unit includes: a determining subunit, used to determine, based on the correlation coefficient, parameters in the reference operation data whose correlation coefficient with the parameters in the target data is lower than a first set threshold, and parameters in the reference operation data whose correlation coefficient between two parameters is higher than a second set threshold; and a filtering subunit, used to filter one of the parameters in the determined reference data whose correlation coefficient with the parameters in the target data is lower than the first set threshold, and parameters in the reference operation data whose correlation coefficient between two parameters is higher than the second set threshold, and to use the filtered data as training data corresponding to the sample core.

[0111] Optionally, the feature data includes at least feature parameters describing the characteristics of the core corresponding to the core, the seepage data includes at least seepage parameters describing the pore seepage performance of the target reservoir, and the oil-bearing data includes at least oil-bearing parameters describing the oil and gas situation in the target reservoir.

[0112] Optionally, after predicting the oil and gas production in the target reservoir using a spectral clustering model based on feature data, seepage data, oil-bearing data, and engineering data, the above-mentioned device further includes: a determination module, used to determine the reservoir classification corresponding to the target reservoir based on the predicted oil and gas production.

[0113] Example 3

[0114] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the oil and gas production prediction method of any one of the above.

[0115] Optionally, in this embodiment, the computer-readable storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals, and the computer-readable storage medium includes a stored program.

[0116] Optionally, during program execution, the device containing the computer-readable storage medium may perform the following functions: acquire characteristic data of core samples collected from the target reservoir; obtain seepage data of the target reservoir based on the core samples; obtain oil-bearing data of the target reservoir based on the core samples; and obtain engineering data for oil and gas extraction from the target reservoir. The program may also predict the oil and gas production in the target reservoir using a spectral clustering model based on the characteristic data, seepage data, oil-bearing data, and engineering data. The spectral clustering model is trained using sample data corresponding to the sample core samples, and the sample data includes at least: sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core samples.

[0117] Example 4

[0118] According to another aspect of the present invention, a processor is also provided for running a program, wherein the program executes the oil and gas production prediction method of any of the above-described methods during runtime.

[0119] This application provides an apparatus including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring characteristic data of core samples collected from a target reservoir; obtaining seepage data of the target reservoir based on the core samples; obtaining oil-bearing data of the target reservoir based on the core samples; and obtaining engineering data for oil and gas extraction from the target reservoir. The application also uses a spectral clustering model to predict the oil and gas production in the target reservoir based on the characteristic data, seepage data, oil-bearing data, and engineering data. The spectral clustering model is trained using sample data corresponding to the sample core samples. The sample data includes at least: sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core samples.

[0120] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having the following method steps: acquiring characteristic data of core samples collected from a target reservoir; obtaining seepage data of the target reservoir based on the core samples; obtaining oil-bearing data of the target reservoir based on the core samples; and engineering data to be used for oil and gas extraction from the target reservoir; and predicting the oil and gas production in the target reservoir based on the characteristic data, seepage data, oil-bearing data, and engineering data using a spectral clustering model. The spectral clustering model is obtained by training the model based on sample data corresponding to the sample core samples, and the sample data includes at least: sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core samples.

[0121] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0122] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0123] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0124] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0125] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0126] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0127] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of predicting hydrocarbon production, characterized by, The method comprises: obtaining characteristic data of a core collected from a target reservoir, obtaining seepage data of the target reservoir measured based on the core, obtaining oil-bearing data of the target reservoir measured based on the core, and obtaining engineering data of the target reservoir for oil and gas exploitation; predicting oil and gas production in the target reservoir based on the characteristic data, the seepage data, the oil-bearing data, and the engineering data by using a spectral clustering model, wherein the spectral clustering model is obtained by training a sample data corresponding to a sample core, and the sample data at least includes sample characteristic data, sample seepage data, sample oil-bearing data, sample engineering data, and sample oil and gas production data corresponding to the sample core; Before the obtaining of the characteristic data of the core collected from the target reservoir, the obtaining of the seepage data of the target reservoir measured based on the core, the obtaining of the oil-bearing data of the target reservoir measured based on the core, and the obtaining of the engineering data of the target reservoir for oil and gas exploitation, the method further comprises: obtaining training data corresponding to the sample core; predicting predicted oil and gas production of a reservoir from which the sample core is derived based on the sample seepage data, the sample oil-bearing data, and the sample engineering data in the training data by using a pre-constructed spectral clustering model; adjusting parameters of the pre-constructed spectral clustering model based on the predicted oil and gas production and actual oil and gas production calculated based on the sample oil and gas production data in the training data, so that a distance between the predicted oil and gas production and the actual oil and gas production satisfies a preset condition; when a training end condition is reached, using the pre-constructed spectral clustering model with the adjusted parameters as the spectral clustering model; After the predicting of the oil and gas production in the target reservoir based on the characteristic data, the seepage data, the oil-bearing data, and the engineering data by using the spectral clustering model, the method further comprises determining a reservoir classification corresponding to the target reservoir based on the predicted oil and gas production, wherein the reservoir classification is divided according to the oil and gas production.

2. The method of claim 1, wherein, Before the obtaining of the training data corresponding to the sample core, the method further comprises: obtaining initial training data corresponding to each sample core; preprocessing the initial training data corresponding to each sample core to obtain the training data corresponding to each sample core.

3. The method of claim 2, wherein, The preprocessing of the initial training data corresponding to each sample core to obtain the training data corresponding to each sample core comprises: performing noise reduction processing on the initial training data corresponding to each sample core; normalizing each parameter in the initial training data after the noise reduction processing to obtain normalized data; calculating a correlation coefficient of each pair of parameters in the normalized data corresponding to each sample core; performing filtering processing on the normalized data based on the correlation coefficient to obtain the training data corresponding to the sample core.

4. The method of claim 3, wherein, The performing of the filtering processing on the normalized data based on the correlation coefficient to obtain the training data corresponding to the sample core comprises: determining, according to the correlation coefficients, parameters in the reference operation data whose correlation coefficients with parameters in the target data are lower than a first set threshold, and parameters in the reference operation data whose correlation coefficients between two parameters are higher than a second set threshold; filtering one of the parameters in the reference data whose correlation coefficients with parameters in the target data are lower than the first set threshold, and the parameters in the reference operation data whose correlation coefficients between two parameters are higher than the second set threshold, and taking the filtered data as training data corresponding to the sample core.

5. The method according to any one of claims 1 to 4, characterized in that, The feature data at least includes feature parameters describing core features corresponding to the core, the percolation data at least includes percolation parameters describing pore percolation performance of the target reservoir, and the oil-bearing data at least includes oil-bearing parameters describing oil and gas conditions in the target reservoir.

6. An apparatus for predicting hydrocarbon production, characterized by, The method comprises: a first obtaining module, configured to obtain feature data of a core collected from a target reservoir, percolation data of the target reservoir measured according to the core, oil-bearing data of the target reservoir measured according to the core, and engineering data of the target reservoir for oil and gas exploitation; a first prediction module, configured to predict oil and gas production in the target reservoir according to the feature data, the percolation data, the oil-bearing data and the engineering data by a spectral clustering model, wherein the spectral clustering model is obtained by model training according to sample data corresponding to a sample core, and the sample data at least includes sample feature data, sample percolation data, sample oil-bearing data, sample engineering data and sample oil and gas production data corresponding to the sample core. The device further comprises: a second obtaining module, configured to obtain training data corresponding to a sample core; a second prediction module, configured to predict predicted oil and gas production of a reservoir from which the sample core is derived according to sample percolation data, sample oil-bearing data and sample engineering data in the training data by a pre-constructed spectral clustering model; an adjusting module, configured to adjust parameters of the pre-constructed spectral clustering model according to the predicted oil and gas production and actual oil and gas production calculated according to sample oil and gas production data in the training data, so that a distance between the predicted oil and gas production and the actual oil and gas production satisfies a preset condition; a processing module, configured to take the pre-constructed spectral clustering model with the adjusted parameters as the spectral clustering model when a training end condition is reached; a determining module, configured to determine a reservoir classification corresponding to the target reservoir according to the predicted oil and gas production after the oil and gas production in the target reservoir is predicted according to the feature data, the percolation data, the oil-bearing data and the engineering data by the spectral clustering model, wherein the reservoir classification is divided according to the oil and gas production.

7. A computer-readable storage medium, characterized in that, The computer readable storage medium comprises a stored program, wherein the program controls a device in which the computer readable storage medium is located to perform the method for predicting oil and gas production according to any one of claims 1 to 5 when the program is running.

8. A processor, comprising: The processor is configured to run a program, wherein the program, when running, performs the method for predicting oil and gas production according to any one of claims 1 to 5.

Images