Method for optimizing fracturing parameters of horizontal well of shale oil reservoir well factory

Abstract

The invention provides a shale oil reservoir well factory horizontal well fracturing parameter optimization method, and relates to the technical field of horizontal well fracturing parameter optimization, and the method specifically comprises the steps: obtaining a plurality of groups of horizontal well historical fracturing parameters, and obtaining an oil and gas production value corresponding to each group of historical fracturing parameters, performing one-to-one mapping on the historical fracturing parameters and the corresponding oil and gas yield values to generate a sample data set; and constructing an oil and gas yield prediction model, inputting the initial optimized fracturing parameter combination into the oil and gas yield prediction model to obtain an oil and gas yield prediction value, constructing a fitness function according to the oil and gas yield prediction value, and obtaining an optimal fracturing parameter combination through a genetic algorithm. Parameters in the optimal fracturing parameter combination are corrected according to the environment correction factors, specific values of the fracturing parameters are obtained, the horizontal well fracturing parameters are optimized according to the specific values of the fracturing parameters, the trial and error cost in the horizontal well fracturing process is reduced, and the resource utilization rate is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of horizontal well fracturing parameter optimization, and specifically provides an optimization method for fracturing parameters of horizontal wells in a well factory of a shale oil reservoir. Background Art

[0002] In the process of shale oil and gas development, fracturing technology, as an important means to increase oil and gas production, its optimization process has always relied on experience and tests. This method often lacks systematicness and scientific basis, resulting in unsatisfactory fracturing effects. Traditional fracturing parameter optimization mainly realizes by adjusting parameters such as injection flow rate, injection pressure, number of fracturing times, and fluid viscosity. However, due to the complexity and variability of the formation, single empirical judgment is difficult to meet the requirements of different reservoirs. In addition, changes in environmental factors during the fracturing process, such as temperature, porosity, and fluid properties, will also significantly affect the fracturing effect. However, these factors are often ignored in traditional methods, resulting in the difficulty of predicting the fracturing operation effect, which may cause waste of resources and environmental damage.

[0003] In addition, there are also significant deficiencies in the utilization of existing data. Many optimization studies of fracturing parameters only rely on a small amount of historical data, lacking in-depth analysis and utilization of large-scale data, resulting in insufficient scientific nature of the optimization process. Even with certain data support, advanced machine learning and data mining technologies are often not used for in-depth analysis, and the laws and potential correlations contained in historical data cannot be fully revealed. In addition, when dealing with the non-linear relationship between fracturing parameters and oil and gas production, existing methods often adopt linear models, which are difficult to capture complex dynamic changes, resulting in low prediction accuracy. These deficiencies make the process of fracturing parameter optimization full of uncertainties, increasing the risks of oil and gas exploitation and making it difficult to maximize economic benefits. Therefore, there is an urgent need for a more scientific, systematic, and data-driven fracturing parameter optimization method to overcome the limitations of existing technologies.

[0004] The above information disclosed in the background art section is only used to strengthen the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimization method for fracturing parameters of horizontal wells in a well factory of a shale oil reservoir to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] An optimization method for fracturing parameters of horizontal wells in a well factory of a shale oil reservoir, the specific steps include:

[0008] Step 1: Obtain several groups of historical fracturing parameters of horizontal wells. At the same time, obtain the corresponding oil and gas production values for each group of historical fracturing parameters. Map the historical fracturing parameters and the corresponding oil and gas production values one by one to generate a sample data set. The fracturing parameters include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity;

[0009] Step 2: Construct an oil and gas production prediction model based on a deep learning algorithm. Use the historical fracturing parameters in the sample data set as inputs and the corresponding oil and gas production values as labels to train the oil and gas production prediction model to obtain a trained oil and gas production prediction model;

[0010] Step 3: Obtain the value ranges of each fracturing parameter. Randomly generate several groups of initial optimized fracturing parameter combinations within the value ranges. Input the initial optimized fracturing parameter combinations into the oil and gas production prediction model to obtain the corresponding oil and gas production prediction values for the combinations, and construct a fitness function based on the oil and gas production prediction values. The initial optimized fracturing parameter combinations include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity;

[0011] Step 4: Take a group of initial optimized fracturing parameter combinations as an individual, and the fracturing parameters within the combination as genes. Calculate the fitness function values of each individual, and obtain the optimal fracturing parameter combination through a genetic algorithm, where the optimal fracturing parameter combination is the initial optimized fracturing parameter combination with the highest fitness function value;

[0012] Step 5: Obtain the environmental parameters during the fracturing process. Calculate the environmental correction factor based on the obtained environmental parameters. Modify the parameters within the optimal fracturing parameter combination according to the environmental correction factor to obtain the specific values of the fracturing parameters, and optimize the horizontal well fracturing parameters according to the specific values of the fracturing parameters. The environmental parameters are average temperature and formation porosity.

[0013] Furthermore, the specific logic for constructing the oil and gas production prediction model is as follows:

[0014] Obtain several groups of historical fracturing parameters of horizontal wells, and at the same time obtain the oil and gas production values corresponding to each group of historical fracturing parameters. Map the historical fracturing parameters and the corresponding oil and gas production values one by one to generate a sample data set. Randomly divide the sample data set into a training set and a test set. The fracturing parameters include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity. Construct an oil and gas production prediction model based on a deep learning algorithm. Use the fracturing parameters in the training set as input and the corresponding oil and gas production values as labels to train the model to obtain a trained oil and gas production prediction model. Substitute the fracturing parameters in the test set into the trained model to obtain the corresponding prediction results. Calculate the error between the prediction results and the actual values in the test set. Determine whether the error meets the preset error threshold. If it meets, output the trained model, that is, the oil and gas production prediction model. If it does not meet, return for continued training.

[0015] The process of randomly dividing into a training set and a test set, the specific logic is: randomly sort the sample data set, and use 80% of the sorted sample data set as the training set, and the remaining 20% as the test set.

[0016] Further, the error is the mean absolute error, root mean square error, and coefficient of determination between the prediction result and the actual value in the test set, specifically as follows:

[0017] The mean absolute error is:

[0018]

[0019] The root mean square error is:

[0020]

[0021] The coefficient of determination is:

[0022]

[0023] Among them, and y i respectively represent the average value, predicted value, and actual value of the oil and gas production corresponding to the data of the i-th group of test samples. n is the number of test samples in the test sample set, and the actual value is the mean value of the actual oil and gas production values of n groups of test samples;

[0024] When MAE ≤ ∈ MAE 、RMSE ≤ ∈ RMSE and It shows that the error meets the preset error threshold and the model training is completed; among them, ∈ MAE is the preset error threshold of MAE, ∈ RMSE is the preset error threshold of RMSE, is R 2 preset error threshold.

[0025] Further, taking a set of initial optimized fracturing parameter combinations as an individual and the fracturing parameters within the combination as genes, all the initial optimized fracturing parameter combinations are encoded, and a number of individuals are obtained. An initial population is constructed based on all the individuals. Each individual contains 4 genes, namely injection flow rate, injection pressure, number of fracturing times, and fluid viscosity.

[0026] Further, the specific logic for obtaining the optimal fracturing parameter combination through the genetic algorithm is as follows:

[0027] Perform selection, crossover, and mutation operations on the initial population in a loop to generate an iterative population containing multiple new individuals. Determine whether the maximum number of iterations is reached. When the maximum number of iterations is reached, select the individual with the highest fitness function value in the current population as the optimal fracturing parameter combination. Otherwise, use the generated iterative population as the initial population for iterative operations until the iterative termination condition is satisfied. Select the initial optimized fracturing parameter combination with the maximum fitness value that appears during the iteration process as the optimal fracturing parameter combination, where the iterative termination condition is the set maximum number of iterations.

[0028] Further, the mathematical expression of the fitness function is as follows:

[0029]

[0030] In the formula, QS is the fitness function value, O pre is the predicted oil and gas production value output by the oil and gas production prediction model, μ is a preset proportionality coefficient, and μ > 0.

[0031] Further, the specific logic for obtaining the average temperature during the fracturing process is as follows: Collect the temperature at different stages of the fracturing operation, and take the average value of the temperatures at all stages as the average temperature during the fracturing process, denoted as T;

[0032] The specific logic for obtaining the formation porosity during the fracturing process is as follows: Take three equal - volume samples from the shale reservoir, calculate their porosities respectively, and take the average value of the three results as the formation porosity during the fracturing process, denoted as

[0033] The formula for calculating the porosity is:

[0034]

[0035] In the formula, φ is the porosity, V p is the pore volume, V t is the total sample volume;

[0036] According to the average temperature and formation porosity after dimensionless processing, calculate the environmental correction factor, and the formula is as follows:

[0037]

[0038] In the formula, COR is the environmental correction factor, T is the average temperature, T ref is the temperature reference threshold, and φ is the formation porosity.

[0039] Furthermore, according to the environmental correction factor, correct the parameters in the optimal fracturing parameter combination to obtain the specific values of each fracturing parameter, and the formula is as follows:

[0040] Q′ = Q * (1 + k1 * COR 2 )

[0041] P′ = P * (1 + k2 * COR 2 )

[0042]

[0043] In the formula, Q′ is the specific value of the injection flow rate, Q is the optimal injection flow rate value in the optimal fracturing parameter combination, P′ is the specific value of the injection pressure, P is the optimal injection pressure value in the optimal fracturing parameter combination, N′ is the specific value of the fracturing times, N is the optimal fracturing times value in the optimal fracturing parameter combination, S′ is the specific value of the fluid viscosity, S is the optimal fluid viscosity value in the optimal fracturing parameter combination, COR is the environmental correction factor, and k1, k2, k3, and k4 are preset proportionality coefficients, and k1 > 0, k2 > 0, k3 > 0, k4 > 0;

[0044] Optimize the horizontal well fracturing parameters according to the specific values of the injection flow rate, injection pressure, fracturing times, and fluid viscosity.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] The present invention combines deep learning and genetic algorithms. By mapping historical fracturing parameters and oil and gas production, an oil and gas production prediction model is constructed, which can accurately predict the oil and gas production under different parameter combinations. By randomly generating an initial optimized fracturing parameter combination and using the environmental correction factor to correct the fracturing parameters, the scientificity and adaptability of the fracturing parameters are ensured, thereby effectively improving the accuracy of oil and gas production prediction. In addition, environmental factors are fully considered in the optimization process, making the fracturing operation more in line with the actual formation conditions, reducing the trial-and-error cost, and improving the resource utilization rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the overall method flow of the present invention. Detailed implementation manners

[0048] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0049] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. The terms such as "comprising" or "including" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative position relationships, and when the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0050] Embodiment:

[0051] Please refer to Figure 1 , the present invention provides a technical solution:

[0052] An optimization method for fracturing parameters of horizontal wells in a shale reservoir well factory, the specific steps include:

[0053] Step 1: Obtain several groups of historical fracturing parameters of horizontal wells, and at the same time obtain the oil and gas production values corresponding to each group of historical fracturing parameters, map the historical fracturing parameters and the corresponding oil and gas production values one by one to generate a sample data set, and the fracturing parameters include injection flow rate, injection pressure, number of fracturing times and fluid viscosity;

[0054] In this embodiment, the advantage of Step 1 is that by obtaining multiple groups of historical fracturing parameters and their corresponding oil and gas production values, a comprehensive sample data set is constructed. This process enables the correlation between fracturing parameters and production to be clearly mapped, providing sufficient data support. Compared with the prior art, this data-driven method can more scientifically reflect the effects of historical fracturing operations, rather than relying solely on experience or uncertain test results. This systematic data collection and processing method helps to improve the prediction accuracy of the model and provides a solid foundation for subsequent optimization.

[0055] In this solution, adopting Step 1 can effectively promote the implementation of the overall solution because the established sample data set provides a reliable training basis for the subsequent deep learning model. By making full use of historical data, the model can discover potential patterns and trends, thereby improving the prediction ability of oil and gas production. This process makes the subsequent parameter optimization more targeted and effective, and ultimately enables efficient and economical fracturing operations, maximizing the development potential of oil and gas resources. Therefore, Step 1 is not only the starting point of the entire optimization method but also an important guarantee for achieving accurate decision-making.

[0056] Step 2: Construct an oil and gas production prediction model based on a deep learning algorithm. Use the historical fracturing parameters in the sample data set as inputs and the corresponding oil and gas production values as labels to train the oil and gas production prediction model, and obtain a trained oil and gas production prediction model.

[0057] In this embodiment, the specific logic for constructing the oil and gas production prediction model is as follows:

[0058] Obtain several groups of historical fracturing parameters of horizontal wells, and at the same time obtain the corresponding oil and gas production values for each group of historical fracturing parameters. Map the historical fracturing parameters and the corresponding oil and gas production values one by one to generate a sample data set. Randomly divide the sample data set into a training set and a test set. The fracturing parameters include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity. Construct an oil and gas production prediction model based on a deep learning algorithm. Use the fracturing parameters in the training set as inputs and the corresponding oil and gas production values as labels to train the model, and obtain a trained oil and gas production prediction model. Substitute the fracturing parameters in the test set into the trained model to obtain the corresponding prediction results. Calculate the error between the prediction results and the actual values in the test set. Determine whether the error meets a preset error threshold. If it meets, output the trained model, that is, the oil and gas production prediction model. If it does not meet, return for continued training.

[0059] The specific logic of the process of randomly dividing into a training set and a test set is as follows: Randomly sort the sample data set, and use 80% of the sorted sample data set as the training set and the remaining 20% as the test set.

[0060] The error is the mean absolute error, root mean square error, and coefficient of determination between the prediction result and the actual value in the test set, as follows:

[0061] The mean absolute error is:

[0062]

[0063] The root mean square error is:

[0064]

[0065] The coefficient of determination is:

[0066]

[0067] Wherein, and y i respectively represent the average value, predicted value and actual value of the oil and gas production corresponding to the data of the i-th group of test samples. n is the number of groups of test samples in the test sample set, and the actual value is the mean value of the actual oil and gas production values of n groups of test samples;

[0068] When MAE ≤ ∈ MAE , RMSE ≤ ∈ RMSE and , it indicates that the error meets the preset error threshold and the model completes training; wherein, ∈ MAE is the preset error threshold of MAE, ∈ RMSE is the preset error threshold of RMSE, is the R 2 preset error threshold.

[0069] The advantage of Step 2 is that by constructing an oil and gas production prediction model through a deep learning algorithm, it realizes the accurate capture of the complex non-linear relationship between historical fracturing parameters and oil and gas production. Compared with the existing technology, traditional methods often use linear models or empirical formulas for prediction, lacking the excavation of deep patterns in the data, resulting in low prediction accuracy. The deep learning model can effectively process large-scale data sets and learn more complex feature representations through multi-level neural networks, thus significantly improving the prediction accuracy of oil and gas production.

[0070] In this solution, adopting Step 2 can effectively promote the implementation of the overall solution because this step provides a scientific basis and a reliable prediction tool for the subsequent optimization of fracturing parameters. By establishing an accurate oil and gas production prediction model, the optimization process can more accurately identify the best combination of fracturing parameters, reducing resource waste and test costs. In addition, accurate prediction results can help engineers make more reasonable decisions in actual operations, thus improving the efficiency and economic benefits of the overall fracturing operation. Therefore, Step 2 is not only a key link to achieve the optimization goal, but also the foundation to drive the development of the entire solution.

[0071] Step 3: Obtain the value ranges of each fracturing parameter, randomly generate several groups of initial optimized fracturing parameter combinations within the value ranges, input the initial optimized fracturing parameter combinations into the oil and gas production prediction model to obtain the predicted oil and gas production values corresponding to the combinations, and construct a fitness function based on the predicted oil and gas production values. The initial optimized fracturing parameter combinations include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity;

[0072] In this embodiment, a set of initial optimized fracturing parameter combinations is regarded as an individual, and the fracturing parameters within the combination are regarded as genes. All the initial optimized fracturing parameter combinations are encoded, and several individuals are obtained in total. An initial population is constructed based on all the individuals. Each individual contains 4 genes, namely injection flow rate, injection pressure, number of fracturing times, and fluid viscosity.

[0073] The advantage of step 3 is that by obtaining the value range of the fracturing parameters and randomly generating the initial optimized fracturing parameter combinations, a comprehensive exploration of the parameter space is realized. This process enables the optimization algorithm to search for the optimal solution among a wider range of parameter combinations, reducing the waste of resources caused by local optimal solutions. Compared with the prior art, traditional methods often rely on fixed or limited fracturing parameter combinations, which limits the effectiveness of the optimization results. By randomly generating the initial combinations, step 3 more effectively considers various possible fracturing strategies, thus improving the flexibility and adaptability of the optimization.

[0074] In this solution, adopting step 3 can effectively promote the implementation of the overall solution because it provides a rich data basis for the subsequent fitness calculation. By inputting multiple initial optimized combinations into the oil and gas production prediction model, the impact of different parameter configurations on production can be comprehensively evaluated, and then a more accurate fitness function can be constructed. This process not only improves the prediction ability of the model for the fracturing effect but also provides a more reliable starting point for the subsequent genetic algorithm optimization, thus enhancing the effect and reliability of the entire fracturing parameter optimization and ensuring that the finally selected parameter combination can maximize the oil and gas production.

[0075] Step 4: Regard a set of initial optimized fracturing parameter combinations as an individual, and the fracturing parameters within the combination as genes. Calculate the fitness function value of each individual, and obtain the optimal fracturing parameter combination through the genetic algorithm, where the optimal fracturing parameter combination is the initial optimized fracturing parameter combination with the highest fitness function value;

[0076] In this embodiment, the mathematical expression of the fitness function is as follows:

[0077]

[0078] In the formula, QS is the fitness function value, O pre is the predicted oil and gas production value output by the oil and gas production prediction model, μ is a preset proportionality coefficient, and μ > 0; when O pre increases, it indicates that the effectiveness of this fracturing parameter combination is enhanced, and the optimization algorithm will be more inclined to select such a combination for further iteration and optimization, thus increasing the oil and gas production, and the fitness function value QS increases accordingly; that is to say, O pre has a positive correlation with QS.

[0079] The advantage of using the logarithmic function in this formula is that it can compress the variation in predicting oil and gas production. Especially when dealing with a large value range, the characteristics of the logarithmic function enable small incremental changes to be more reasonably reflected in the fitness value, which helps to ensure the stability and convergence of the optimization process. The introduction of the cube root is to avoid the situation where the fitness function returns negative infinity or is undefined when the predicted value is zero. By adding 1 before O pre it can ensure that even when the predicted production is zero, the fitness function can have a valid output value and still maintain a smooth change trend.

[0080] The advantage of step 4 is that by treating the initial optimized fracturing parameter combination as an individual and using the genetic algorithm for selection, crossover, and mutation operations, it can effectively find the optimal solution in a large-scale parameter space. Compared with the existing technology, traditional methods mostly rely on a single optimization strategy, which is prone to local optima. The population search characteristic of the genetic algorithm enables multiple solutions to be evaluated simultaneously, thus exploring possible fracturing parameter combinations more comprehensively and enhancing the flexibility and efficiency of the optimization process.

[0081] In this solution, adopting step 4 can effectively promote the implementation of the overall solution because it provides a systematic methodology for finding the optimal fracturing parameter combination. Through the iterative process of the genetic algorithm, the fitness function value can be gradually improved, thereby obtaining a more effective fracturing parameter configuration. This not only helps to accurately optimize the economic benefits of the fracturing operation but also reduces the test cost and resource waste, ensuring that the finally selected parameter combination can maximize the oil and gas production and enhance the success rate and sustainability of the overall operation. Therefore, step 4 is an important link in realizing the optimization of fracturing parameters, ensuring the scientific nature and practicality of the entire optimization process.

[0082] Step 5: Obtain the environmental parameters during the fracturing process, calculate the environmental correction factor based on the obtained environmental parameters, correct the parameters within the optimal fracturing parameter combination according to the environmental correction factor to obtain the specific values of the fracturing parameters, and optimize the horizontal well fracturing parameters according to the specific values of the fracturing parameters. The environmental parameters are the average temperature and formation porosity;

[0083] In this embodiment, the specific logic for obtaining the average temperature during the fracturing process is: collect the temperatures at different stages of the fracturing operation, and take the average value of the temperatures at all stages as the average temperature during the fracturing process, denoted as T;

[0084] The specific logic for obtaining the formation porosity during the fracturing process is: take three equal samples from the shale reservoir, calculate their porosities respectively, and take the average value of the three results as the formation porosity during the fracturing process, denoted as

[0085] The formula for calculating the porosity is:

[0086]

[0087] where φ is the porosity, V p is the pore volume, and V t is the total volume of the sample;

[0088] According to the average temperature and formation porosity after dimensionless processing, the environmental correction factor is calculated, and the formula is as follows:

[0089]

[0090] where COR is the environmental correction factor, T is the average temperature, and T ref is the temperature reference threshold, and φ is the formation porosity. When |T - T ref | increases, it means that the deviation between the average temperature and the temperature reference threshold increases, which has a certain negative impact on the effect of the fracturing operation. Therefore, the environmental correction factor COR decreases; a higher porosity means more space to accommodate fluids, and the formation can better store and flow oil and gas, improving the production potential. Therefore, when φ increases, the environmental correction factor COR increases; that is, it shows that |T - T ref | is negatively correlated with COR, and φ is positively correlated with COR.

[0091] According to the environmental correction factor, the parameters in the optimal fracturing parameter combination are corrected to obtain the specific values of each fracturing parameter. The formula is as follows:

[0092] Q′ = Q * (1 + k1 * COR 2 )

[0093] P′ = P * (1 + k2 * COR 2 )

[0094]

[0095] where Q′ is the specific value of the injection flow rate, Q is the optimal injection flow rate value in the optimal fracturing parameter combination, P′ is the specific value of the injection pressure, P is the optimal injection pressure value in the optimal fracturing parameter combination, N′ is the specific value of the fracturing times, N is the optimal fracturing times value in the optimal fracturing parameter combination, S′ is the specific value of the fluid viscosity, S is the optimal fluid viscosity value in the optimal fracturing parameter combination, COR is the environmental correction factor, and k1, k2, k3, and k4 are preset proportionality coefficients, and k1 > 0, k2 > 0, k3 > 0, k4 > 0.

[0096] The specific logic for determining each proportionality coefficient is as follows: By analyzing historical data, determine the variation trend of flow rate under different environmental conditions, and use the linear regression method to obtain an appropriate value of k1; Use historical fracturing engineering data to analyze the variation of injection pressure under different environmental conditions and identify an appropriate value of k2; By studying the correlation between the number of fracturing and environmental parameters, determine the sensitivity of this parameter under different conditions, and then set k3; Obtain the viscosity data of the fluid under different environmental conditions through experiments, analyze its variation, and select an appropriate k4.

[0097] For the correction formulas of injection flow rate and injection pressure, the introduction of the square term means that the correction effect is proportional to the square of the environmental correction factor. This design can make the correction amplitude increase significantly when COR is relatively large, reflecting the non-linear characteristics of the influence of environmental factors such as temperature and porosity on the fracturing operation. As the environmental deviation increases, the adjustment of fracturing requirements will also be more obvious. For the correction formula of the number of fracturing, using the square root form means that the adjustment of the number of fracturing is proportional to the increase of the correction factor, but the growth rate is slow. For the correction formula of fluid viscosity, using the reciprocal form means that as the environmental correction factor increases, the viscosity of the fluid needs to be appropriately reduced.

[0098] When COR increases, it means that the temperature condition is more suitable or the porosity is higher. In this case, the formation can more effectively accept more fluid, thereby increasing the injection flow rate, so Q′ increases; Similarly, in the above case, it is necessary to increase the injection pressure to ensure that the fluid can smoothly enter the formation, that is, P′ increases; When the fluidity of the formation improves, the fluid can be injected more smoothly and propagated in the formation, which means that under the same fracturing conditions, it is not necessary to perform multiple fracturing to achieve good flow effects, that is, N′ decreases; When COR increases, the viscosity of the fluid generally decreases, which is mainly because high porosity helps the fluid to flow more easily; That is to say, COR has a positive correlation with Q′ and P′, and COR has a negative correlation with N′ and S′.

[0099] Optimize the horizontal well fracturing parameters according to the specific values of injection flow rate, injection pressure, number of fracturing, and fluid viscosity.

[0100] The advantage of step 5 is that by introducing the influence of environmental parameters, the optimization result can be adjusted more precisely. This process not only ensures that the fracturing parameters can adapt to the actual operation environment, but also improves the final operation efficiency and safety. Compared with the existing technology, traditional methods usually ignore the influence of environmental factors on the fracturing effect, resulting in the optimization result may not be effective enough in actual operation. By calculating the environmental correction factor and correcting the optimal fracturing parameter combination, step 5 makes the optimization process more practical and can achieve higher oil and gas production.

[0101] In this solution, adopting Step 5 can effectively promote the implementation of the overall solution because it ensures that the optimized parameters are not only the theoretically optimal choice but also can meet the actual conditions on-site. Through the specific values of the corrected fracturing parameters, the success rate and production of the fracturing operation can be increased, making the entire fracturing process more efficient and economical. In addition, this optimization method can significantly reduce the operation risks caused by environmental changes, ensuring the operation effects under different geological and climatic conditions, thereby providing more scientific and reliable technical support for the overall oil and gas development.

[0102] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0103] The above embodiments can be implemented in whole or in part through software, hardware, firmware, or any other arbitrary combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented through electronic hardware or the combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0104] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0105] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all of them should be covered within the protection scope of this application.

Claims

1. A method for optimizing fracturing parameters of horizontal wells in a shale oil reservoir well factory, characterized in that The specific steps include: Step 1: Obtain several groups of historical fracturing parameters of horizontal wells. At the same time, obtain the oil and gas production values corresponding to each group of historical fracturing parameters. Map the historical fracturing parameters and the corresponding oil and gas production values one by one to generate a sample data set. The fracturing parameters include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity. Step 2: Build an oil and gas production prediction model based on a deep learning algorithm. Use the historical fracturing parameters in the sample data set as inputs and the corresponding oil and gas production values as labels to train the oil and gas production prediction model to obtain a trained oil and gas production prediction model. Step 3: Obtain the value ranges of each fracturing parameter. Randomly generate several groups of initial optimized fracturing parameter combinations within the value ranges. Input the initial optimized fracturing parameter combinations into the oil and gas production prediction model to obtain the oil and gas production prediction values corresponding to the combinations, and construct a fitness function based on the oil and gas production prediction values. The initial optimized fracturing parameter combinations include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity. Step 4: Take a group of initial optimized fracturing parameter combinations as an individual, and the fracturing parameters within the combination as genes. Calculate the fitness function values of each individual, and obtain the optimal fracturing parameter combination through a genetic algorithm. The optimal fracturing parameter combination is the initial optimized fracturing parameter combination with the highest fitness function value. Step 5: Obtain the environmental parameters during the fracturing process. Calculate the environmental correction factor based on the obtained environmental parameters. Correct the parameters within the optimal fracturing parameter combination according to the environmental correction factor to obtain the specific values of the fracturing parameters, and optimize the horizontal well fracturing parameters according to the specific values of the fracturing parameters. The environmental parameters include average temperature and formation porosity.

2. The method for optimizing the fracturing parameters of horizontal wells in a shale oil reservoir well factory according to claim 1, wherein: The specific logic for building the oil and gas production prediction model is as follows: Obtain several groups of historical fracturing parameters of horizontal wells. At the same time, obtain the oil and gas production values corresponding to each group of historical fracturing parameters. Map the historical fracturing parameters and the corresponding oil and gas production values one by one to generate a sample data set. Randomly divide the sample data set into a training set and a test set. The fracturing parameters include injection flow rate, injection pressure, number of fracturing times, and fluid viscosity. Build an oil and gas production prediction model based on a deep learning algorithm. Use the fracturing parameters in the training set as inputs and the corresponding oil and gas production values as labels to train the model to obtain a trained oil and gas production prediction model. Substitute the fracturing parameters in the test set into the trained model to obtain the corresponding prediction results. Calculate the error between the prediction results and the actual values in the test set. Determine whether the error meets a preset error threshold. If it meets, output the trained model, that is, the oil and gas production prediction model. If it does not meet, return for continued training. The error is the mean absolute error, root mean square error, and coefficient of determination between the prediction results and the actual values in the test set. The specific logic for randomly dividing into a training set and a test set is as follows: Randomly sort the sample data set, and take 80% of the sorted sample data set as the training set and the remaining 20% as the test set.

3. The optimization method of fracturing parameters for horizontal wells in a shale oil reservoir well factory according to claim 2, characterized in that: The error is the mean absolute error, root mean square error, and coefficient of determination between the prediction result and the actual value in the test set, as follows: The mean absolute error is: The root mean square error is: The coefficient of determination is: Among them, and y i respectively represent the average value, predicted value, and actual value of the oil and gas production corresponding to the data of the i-th group of test samples. n is the number of groups of test samples in the test sample set, and the actual value is the mean of the actual oil and gas production values of the n groups of test samples; When MAE ≤ ∈ MAE and RMSE ≤ ∈ RMSE and it indicates that the error meets the preset error threshold, and the model has completed training; where ∈ MAE is the preset error threshold for MAE, ∈ RMSE is the preset error threshold for RMSE, is the R 2 preset error threshold.

4. A method for optimizing fracturing parameters of horizontal wells in a shale oil reservoir well factory according to claim 1, characterized in that: Taking a set of initial optimized fracturing parameter combinations as an individual, and the fracturing parameters within the combination as genes, all initial optimized fracturing parameter combinations are encoded to obtain several individuals, and an initial population is constructed based on all individuals. Each individual contains 4 genes, namely injection flow rate, injection pressure, number of fracturing stages, and fluid viscosity.

5. The optimization method for fracturing parameters of horizontal wells in a shale oil reservoir well factory according to claim 1, characterized in that: The specific logic for obtaining the optimal fracturing parameter combination through the genetic algorithm is as follows: Perform selection, crossover, and mutation operations on the initial population in a loop to generate an iterative population containing multiple new individuals. Determine whether the maximum number of iterations has been reached. When the maximum number of iterations is reached, select the individual with the highest fitness function value in the current population as the optimal fracturing parameter combination. Otherwise, use the generated iterative population as the initial population for iterative operations until the iterative termination condition is met. Select the initial optimized fracturing parameter combination with the maximum fitness value that appears during the iteration process as the optimal fracturing parameter combination, where the iterative termination condition is the set maximum number of iterations.

6. The optimization method for fracturing parameters of horizontal wells in a shale oil reservoir well factory according to claim 5, wherein: The mathematical expression of the fitness function is as follows: where QS is the fitness function value, and O pre is the predicted oil and gas production value output by the oil and gas production prediction model, μ is a preset proportionality coefficient, and μ > 0.

7. A method for optimizing the fracturing parameters of horizontal wells in a shale oil reservoir well factory according to claim 1, characterized in that: The specific logic for obtaining the average temperature during the fracturing process is: Collect the temperature at different stages of the fracturing operation, and take the average value of the temperatures at all stages as the average temperature during the fracturing process, denoted as T; To obtain the formation porosity during the fracturing process, the specific logic is as follows: Take three equal samples from the shale reservoir, calculate their porosities respectively, and take the average of the three results as the formation porosity during the fracturing process, denoted as The formula for calculating porosity is: where φ is the porosity, V p is the pore volume, and V t is the total volume of the sample; Based on the dimensionless average temperature and formation porosity, calculate the environmental correction factor. The formula is as follows: Where COR is the environmental correction factor, T is the average temperature, T ref is the temperature reference threshold, and φ is the formation porosity.

8. The method for optimizing the fracturing parameters of horizontal wells in a shale oil reservoir well factory according to claim 1, characterized in that: Based on the environmental correction factor, correct the parameters within the optimal fracturing parameter combination to obtain the specific values of each fracturing parameter. The formula is as follows: Q′ = Q * (1 + k1 * COR 2 ) P′ = P * (1 + k2 * COR 2 ) In the formula, Q′ is the specific value of the injection flow rate, Q is the optimal injection flow rate value within the optimal fracturing parameter combination, P′ is the specific value of the injection pressure, P is the optimal injection pressure value within the optimal fracturing parameter combination, N′ is the specific value of the number of fracturing stages, N is the optimal number of fracturing stages value within the optimal fracturing parameter combination, S′ is the specific value of the fluid viscosity, S is the optimal fluid viscosity value within the optimal fracturing parameter combination, COR is the environmental correction factor, and k1, k2, k3, and k4 are preset proportionality coefficients, and k1>0, k2>0, k3>0, k4>0; Optimize the horizontal well fracturing parameters according to the specific values of the injection flow rate, injection pressure, number of fracturing stages, and fluid viscosity.

Images