High-efficiency reservoir stimulation fracturing method based on proppant with graded particle size and its application

Abstract

The present invention provides an efficient reservoir production-increasing fracturing method based on microparticle diameter gradation proppants and its application, and belongs to the technical field of oilfield production-increasing. The method achieves efficient production-increasing of the reservoir by controlling the particle size and filling sequence of the proppant and optimizing the fracturing parameters in combination with an intelligent algorithm. The specific steps include: selecting proppants of different particle sizes according to formation characteristics, fluid flow laws and fracturing targets; designing the particle size gradation filling sequence of the proppant; optimizing the fracturing parameters through an intelligent algorithm; implementing a fracturing operation to fill the proppant into the crack; real-time monitoring of crack expansion, proppant filling and fluid flow, and adjusting the fracturing parameters and filling sequence; after the fracturing operation is completed, evaluating the fracturing effect, and further optimizing the proppant selection, filling sequence and fracturing parameters. The present invention adopts the above-mentioned efficient reservoir production-increasing fracturing method based on microparticle diameter gradation proppants and its application, which can significantly improve the production-increasing effect of the reservoir and has broad application prospects.

Description

Technical Field

[0001] The invention relates to the technical field of oilfield production increase, in particular to a high-efficiency reservoir production increase fracturing method based on a microparticle diameter graded proppant and an application thereof. Background Art

[0002] In the process of oil field development, fracturing technology is widely used in the development of low permeability oil and gas reservoirs as an important production increase measure. Traditional fracturing technology mainly relies on high-pressure fluid to form a fracture network in the formation, and fills the fractures with proppants (such as sand) to keep the fractures open, thereby improving the flow capacity of oil and gas. However, as oil field development enters the middle and late stages, the formation conditions become more complex, and traditional fracturing technology faces many challenges:

[0003] (1) Insufficient optimization of the fracture network: The expansion and conductivity of the fracture network are affected by many factors, including the geometry of the fractures, the distribution of the proppant, etc. Traditional methods lack comprehensive optimization of these factors, resulting in unsatisfactory fracturing results.

[0004] (2) Fracturing parameters are not adjusted in a timely manner: During the fracturing process, the expansion of the cracks and the filling of the proppant are changing dynamically, but traditional methods often lack real-time monitoring and adjustment mechanisms and cannot respond to formation changes in a timely manner.

[0005] (3) Inaccurate evaluation of production increase effect: After the fracturing operation is completed, there is a lack of accurate evaluation of the production increase effect, which cannot provide effective guidance for subsequent fracturing operations. Summary of the invention

[0006] The purpose of the present invention is to provide an efficient reservoir production-increasing fracturing method based on a microparticle-diameter graded proppant and its application, which can significantly improve the production-increasing effect of the oil reservoir and has a wide application prospect.

[0007] To achieve the above object, the present invention provides an efficient reservoir production enhancement fracturing method based on a microparticle diameter graded proppant, comprising the following steps:

[0008] Step S1, selecting proppants of different particle sizes according to formation characteristics, fluid flow rules and fracturing targets;

[0009] Step S2, designing the filling sequence of proppant particle size according to the formation fracture characteristics and the target production increase effect;

[0010] Step S3, optimizing fracturing parameters by intelligent algorithm;

[0011] Step S4, performing fracturing operations according to the designed proppant particle size gradient filling and optimized fracturing parameters, and filling the proppant into the fractures in sequence;

[0012] Step S5: During the fracturing process, real-time monitoring of the crack expansion, proppant filling and fluid flow is performed, and fracturing parameters and proppant filling sequence are adjusted according to the monitoring results;

[0013] Step S6: After the fracturing operation is completed, the fracturing effect is evaluated by comparing the production data before and after fracturing, and the proppant selection, filling sequence, and fracturing parameters are further optimized according to the evaluation results.

[0014] Preferably, in step S2, the filling sequence of proppant particle size is designed according to the formation fracture characteristics and the target production increase effect, and the specific operation is as follows:

[0015] Step S21, according to the proppants of different particle sizes determined in step S1, first fill the proppants with large particle sizes to form a stable fracture framework, and then fill in the order of decreasing particle sizes;

[0016] Step S22, using numerical simulation software to simulate the distribution of proppants with different particle sizes in the fracture, and verifying the designed particle size gradient filling sequence;

[0017] Step S23: adjusting the proppant particle size and injection parameters according to the simulation results to optimize the step filling design.

[0018] Preferably, when filling the proppant in step S21, the process further includes:

[0019] Step S211, pre-treating the selected proppants of different particle sizes;

[0020] Step S212: After filling the large-size proppant to form a stable fracture framework, fill the remaining proppant in a layer-by-layer manner in the order of decreasing particle size, and perform a pressure holding operation after each layer of filling;

[0021] Step S213: During the filling process, the filling speed and filling amount of the proppant are dynamically adjusted according to the real-time monitored fracture expansion and fluid flow conditions.

[0022] Preferably, in step S22, numerical simulation software is used to simulate the distribution of proppants with different particle sizes in the fracture, and the specific operation is as follows:

[0023] Step S221, establishing a fracture network model based on formation characteristics and fracture characteristics;

[0024] ;

[0025] in, Indicates the crack width; Indicates the amount of proppant; represents the normal stress; represents the radius of proppant particle size; represents the crack normal stiffness; Represents the base of the logarithm of natural numbers;

[0026] Step S222: Establishing a fracture conductivity model:

[0027] ;

[0028] in, Indicates the fracture conductivity; Indicates fracturing fluid viscosity; represents the proppant particle size; represents porosity; represents the permeability;

[0029] Step S223, simulating the distribution of proppants of different particle sizes in the fracture:

[0030] ;

[0031] in, represents the distribution function; Indicates the injection flow; Indicates the length of the crack;

[0032] Step S224, analyzing the fracture extension mode and proppant distribution shown in the simulation results, and evaluating the fracture conductivity;

[0033] ;

[0034] in, Represents the crack surface area.

[0035] Preferably, in step S23, the proppant particle size and injection parameters are adjusted according to the simulation results to optimize the step filling design. The specific operations are as follows:

[0036] Step S231, when the fracture conductivity does not reach the target conductivity, adjusting the particle size distribution of the proppant;

[0037] ;

[0038] in, represents the new proppant particle size; Indicates the target flow capacity; represents the adjustment factor;

[0039] Step S232, adjusting the injection pressure and injection flow rate to optimize the fracture network expansion and support effect;

[0040] ;

[0041] in, Indicates injection pressure;

[0042] Step S233: re-perform numerical simulation using the adjusted parameters to verify the optimization effect.

[0043] Preferably, in step S3, the fracturing parameters are optimized by an intelligent algorithm, and the specific operations are as follows:

[0044] Step S31, according to the formation characteristics, fluid flow law and fracturing target of step S1 and the fracture characteristics and target production increase effect of step S2, historical fracturing data are collected, and the collected historical fracturing data are normalized to scale all features to the range of [0,1];

[0045] Historical fracturing data includes proppant particle size, injection pressure, injection rate, injection volume, fracture length, width, and production;

[0046] Step S32, the number of input layer nodes of the neural network is set to 6, two hidden layers are set, each hidden layer has 12 neurons, and the number of output layer nodes is set to 1;

[0047] Step S33: The hidden layer uses the ReLU activation function, the output layer uses the linear activation function, and the root mean square error is used as the loss function;

[0048] Step S34: Use the learning rate decay strategy, halve the learning rate every 10 epochs, use L2 regularization to prevent overfitting, and the regularization parameter Set to 0.01;

[0049] Step S35, using 5-fold cross validation to evaluate the generalization ability of the model;

[0050] Step S36: Use the trained neural network model to predict the production under different fracturing parameter combinations, and optimize the fracturing parameters by combining the particle swarm algorithm based on dynamic neighbors and generalized learning to quantify the production;

[0051] The particle swarm size is set to 30, the dimension is set to 6, and the inertia weight is set to Decrease from 0.9 to 0.4, the learning factor Set to 2. Set to 2;

[0052] Step S37: Apply the optimized fracturing parameters to actual fracturing operations, and compare them with historical data to evaluate the optimization effect.

[0053] Preferably, in step S5, during the fracturing process, the expansion of the cracks, the filling of the proppant and the flow of the fluid are monitored in real time, and the fracturing parameters and the filling sequence of the proppant are adjusted according to the monitoring results. The specific operations are as follows:

[0054] Step S51, deploying a microseismic monitoring system, a pressure sensor, and a flow meter to collect data on crack expansion, pressure change, and fluid flow in real time;

[0055] Step S52: Use microseismic monitoring technology to track the expansion of the crack in real time and determine the length and width of the crack;

[0056] Step S53, monitoring the pressure change in the fracture through a pressure sensor to evaluate the filling condition of the proppant;

[0057] ;

[0058] in, represents the proppant concentration; represents the proppant volume; represents the crack volume;

[0059] Step S54, reading flow meter data to monitor the injection speed and total amount of fracturing fluid;

[0060] ;

[0061] in, Indicates the fracturing fluid velocity;

[0062] Step S55, calculating the pressure change rate;

[0063] ;

[0064] in, express The pressure of the moment; express The pressure of the moment; Indicates a time interval;

[0065] Step S56, adjusting the injection pressure and injection speed according to the pressure change rate and the crack expansion;

[0066] Step S57, adjusting the particle size of the proppant according to the fracture width and the proppant concentration;

[0067] Step S58: Establish a monitoring-adjustment cycle to monitor the fracturing process in real time and continuously adjust the fracturing parameters and proppant filling sequence according to the monitoring results.

[0068] Preferably, in step S6, after the fracturing operation is completed, the fracturing effect is evaluated by comparing the production data before and after fracturing, and the proppant selection, filling sequence, and fracturing parameters are further optimized according to the evaluation results. The specific operations are as follows:

[0069] Step S61, collecting production data before and after fracturing, and calculating the production increase, where the production data is oil production;

[0070] ;

[0071] in, Indicates the change in output; Indicates the production after fracturing; represents the production before fracturing;

[0072] Step S62, comprehensively comparing production data, evaluating fracture effects, and checking proppant distribution to determine the effect of the fracturing operation;

[0073] Step S63, feeding back the evaluation result to the neural network model of step S3, adjusting the model parameters, and further training and optimizing the model;

[0074] ;

[0075] in, represents the updated model parameters; Represents the model parameters before updating; Indicates the update amount of the parameter.

[0076] Preferably, the evaluation results include the yield change , fracture length and width, proppant concentration .

[0077] The present invention also provides an application of the above-mentioned high-efficiency reservoir production-increasing fracturing method based on graded proppants with different particle diameters in oil field development.

[0078] Therefore, the present invention adopts the above-mentioned high-efficiency reservoir production-increasing fracturing method based on the graded proppant of microparticle diameter and its application, and the beneficial technical effects are as follows:

[0079] (1) Significantly improve the effect of increasing oil reservoir production: By controlling the particle size and filling sequence of the proppant, the method of the present invention can more effectively support the fractures and improve the conductivity of the fractures, thereby significantly increasing the production of the oil reservoir. This particle size tiered filling method can better adapt to the formation characteristics and fluid flow laws, making the fracturing effect more significant.

[0080] (2) Optimizing fracturing parameters and improving operation efficiency: The present invention optimizes fracturing parameters by combining intelligent algorithms, and can more accurately control various parameters in the fracturing process, such as injection pressure, injection speed, injection volume, etc., thereby improving the efficiency and success rate of fracturing operations. This intelligent optimization method can greatly reduce the reliance on human experience and improve the reliability and repeatability of fracturing operations.

[0081] (3) Real-time monitoring and adjustment to ensure fracturing effect: During the fracturing process, the present invention can timely adjust the fracturing parameters and filling sequence by real-time monitoring of crack expansion, proppant filling and fluid flow to ensure that the fracturing effect reaches the expected result. This real-time monitoring and adjustment method can timely discover and solve problems and avoid unnecessary waste and loss. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 This is a flow chart of a high-efficiency reservoir production-increasing fracturing method based on proppants with graded particle diameters according to the present invention;

[0083] Figure 2 Flow chart for numerical simulation software to simulate the distribution of proppants with different particle sizes in fractures;

[0084] Figure 3 Flowchart for optimizing fracturing parameters for intelligent algorithms;

[0085] Figure 4 To monitor and adjust the fracturing process diagram in real time. DETAILED DESCRIPTION

[0086] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0087] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0088] Embodiment 1

[0089] like Figure 1 As shown, it is a flow chart of the efficient reservoir production enhancement fracturing method based on the microparticle diameter graded proppant of the present invention, which specifically includes the following steps:

[0090] Step S1: Select proppants of different particle sizes according to formation characteristics, fluid flow rules and fracturing targets. This is the prior art and will not be elaborated in detail.

[0091] Step S2: Design the filling sequence of proppant particle size according to the formation fracture characteristics and the target production increase effect. The specific operation is as follows:

[0092] Step S21, according to the proppants of different particle sizes determined in step S1, first fill the proppants with large particle sizes to form a stable fracture framework, and then fill in the order of decreasing particle sizes;

[0093] When filling the proppant in step S21, the method further includes:

[0094] Step S211, pre-treating the selected proppants of different particle sizes;

[0095] Pretreatment methods include but are not limited to crushing, screening, cleaning, drying and surface modification;

[0096] Step S212: After filling the large-size proppant to form a stable fracture framework, the remaining proppant is filled in a layer-by-layer manner in the order of decreasing particle size, and a short pressure holding operation is performed after each layer of filling to ensure that the proppant can be densely stacked and effectively support the fracture;

[0097] Step S213: During the filling process, the filling speed and filling amount of the proppant are dynamically adjusted according to the real-time monitored fracture expansion and fluid flow conditions to maintain the stable expansion of the fracture and the effective distribution of the proppant.

[0098] Step S22: Figure 2 As shown in the figure, numerical simulation software is used to simulate the distribution of proppants with different particle sizes in the fracture and verify the designed particle size gradient filling sequence. The specific operations are as follows:

[0099] Step S221, establishing a fracture network model based on formation characteristics and fracture characteristics;

[0100] ;

[0101] in, Indicates the crack width; Indicates the amount of proppant; represents the normal stress; represents the radius of proppant particle size; represents the crack normal stiffness; Represents the base of the logarithm of natural numbers;

[0102] Step S222: Establish a fracture conductivity model:

[0103] ;

[0104] in, Indicates the fracture conductivity; Indicates fracturing fluid viscosity; represents the proppant particle size; represents porosity; represents the permeability;

[0105] Step S223, simulating the distribution of proppants of different particle sizes in the fracture:

[0106] ;

[0107] in, represents the distribution function; Indicates the injection flow; Indicates the length of the crack;

[0108] Step S224, analyzing the fracture extension mode and proppant distribution shown in the simulation results, and evaluating the fracture conductivity;

[0109] ;

[0110] in, Represents the crack surface area.

[0111] Step S23: adjust the proppant particle size and injection parameters according to the simulation results to optimize the step filling design. The specific operations are as follows:

[0112] Step S231, when the fracture conductivity does not reach the target conductivity, adjusting the particle size distribution of the proppant;

[0113] ;

[0114] in, represents the new proppant particle size; Indicates the target flow capacity; represents the adjustment factor;

[0115] Step S232, adjusting the injection pressure and injection flow rate to optimize the fracture network expansion and support effect;

[0116] ;

[0117] in, Indicates injection pressure;

[0118] Step S233: re-perform numerical simulation using the adjusted parameters to verify the optimization effect.

[0119] Step S3: Figure 3 As shown in the figure, the fracturing parameters are optimized through intelligent algorithms. The specific operations are as follows:

[0120] Step S31, according to the formation characteristics, fluid flow law and fracturing target of step S1 and the fracture characteristics and target production increase effect of step S2, historical fracturing data are collected, and the collected historical fracturing data are normalized to scale all features to the range of [0,1];

[0121] Historical fracturing data include proppant particle size, injection pressure, injection rate, injection volume, fracture length, width, and production;

[0122] Step S32, the number of input layer nodes of the neural network is set to 6, two hidden layers are set, each hidden layer has 12 neurons, and the number of output layer nodes is set to 1;

[0123] Step S33: The hidden layer uses the ReLU activation function, the output layer uses the linear activation function, and the root mean square error is used as the loss function;

[0124] Step S34: Use the learning rate decay strategy, halve the learning rate every 10 epochs, use L2 regularization to prevent overfitting, and the regularization parameter Set to 0.01;

[0125] Step S35, using 5-fold cross validation to evaluate the generalization ability of the model;

[0126] Step S36: Use the trained neural network model to predict the production under different fracturing parameter combinations, and optimize the fracturing parameters by combining the particle swarm algorithm based on dynamic neighbors and generalized learning to quantify the production;

[0127] The particle swarm size is set to 30, the dimension is set to 6, and the inertia weight is set to Decrease from 0.9 to 0.4, the learning factor Set to 2. Set to 2;

[0128] Step S37: Apply the optimized fracturing parameters to actual fracturing operations, and compare them with historical data to evaluate the optimization effect.

[0129] Step S4, performing fracturing operations according to the designed proppant particle size gradient filling and optimized fracturing parameters, and filling the proppant into the fractures in sequence;

[0130] Step S5: Figure 4 As shown, during the fracturing process, the expansion of the cracks, the filling of the proppant and the flow of the fluid are monitored in real time, and the fracturing parameters and the filling sequence of the proppant are adjusted according to the monitoring results. The specific operations are as follows:

[0131] Step S51, deploying a microseismic monitoring system, a pressure sensor, and a flow meter to collect data on crack expansion, pressure change, and fluid flow in real time;

[0132] Step S52: Use microseismic monitoring technology to track the expansion of the crack in real time and determine the length and width of the crack;

[0133] Step S53, monitoring the pressure change in the fracture through a pressure sensor to evaluate the filling condition of the proppant;

[0134] ;

[0135] in, represents the proppant concentration; represents the proppant volume; represents the crack volume;

[0136] Step S54, reading flow meter data to monitor the injection speed and total amount of fracturing fluid;

[0137] ;

[0138] in, Indicates the fracturing fluid velocity;

[0139] Step S55, calculating the pressure change rate;

[0140] ;

[0141] in, express The pressure of the moment; express The pressure of the moment; Indicates a time interval;

[0142] Step S56, adjusting the injection pressure and injection speed according to the pressure change rate and the crack expansion;

[0143] Step S57, adjusting the particle size of the proppant according to the fracture width and the proppant concentration;

[0144] Step S58: Establish a monitoring-adjustment cycle to monitor the fracturing process in real time and continuously adjust the fracturing parameters and proppant filling sequence according to the monitoring results.

[0145] Step S6: After the fracturing operation is completed, the fracturing effect is evaluated by comparing the production data before and after fracturing, and the proppant selection, filling sequence, and fracturing parameters are further optimized according to the evaluation results. The specific operations are as follows:

[0146] Step S61, collecting production data before and after fracturing, and calculating the production increase, where the production data is oil production;

[0147] ;

[0148] in, Indicates the change in output; Indicates the production after fracturing; represents the production before fracturing;

[0149] Step S62, comprehensively comparing production data, evaluating fracture effects, and checking proppant distribution to determine the effect of the fracturing operation;

[0150] Step S63, feeding back the evaluation result to the neural network model of step S3, adjusting the model parameters, and further training and optimizing the model;

[0151] ;

[0152] in, represents the updated model parameters; Represents the model parameters before updating; Indicates the update amount of the parameter.

[0153] The evaluation results include the change in production , fracture length and width, proppant concentration .

[0154] It is worth noting that the contents not elaborated in detail in the present invention are all prior art and are well known to those skilled in the art.

[0155] Therefore, the present invention adopts the above-mentioned high-efficiency reservoir production-increasing fracturing method based on the graded proppant of microparticle diameter and its application, which can significantly improve the production-increasing effect of the oil reservoir and has broad application prospects.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. An efficient reservoir production enhancement fracturing method based on proppant with graded particle diameters, characterized in that: The following steps are involved: Step S1, selecting proppants of different particle sizes according to formation characteristics, fluid flow rules and fracturing targets; Step S2, designing the filling sequence of proppant particle size according to the formation fracture characteristics and the target production increase effect; Step S3, optimizing fracturing parameters by intelligent algorithm; Step S4, performing fracturing operations according to the designed proppant particle size gradient filling and optimized fracturing parameters, and filling the proppant into the fractures in sequence; Step S5: During the fracturing process, real-time monitoring of the crack expansion, proppant filling and fluid flow is performed, and fracturing parameters and proppant filling sequence are adjusted according to the monitoring results; Step S6: After the fracturing operation is completed, the fracturing effect is evaluated by comparing the production data before and after fracturing, and the proppant selection, filling sequence, and fracturing parameters are further optimized according to the evaluation results; In step S2, the filling sequence of proppant particle size is designed according to the formation fracture characteristics and the target production increase effect. The specific operation is as follows: Step S21, according to the proppants of different particle sizes determined in step S1, first fill the proppants with large particle sizes to form a stable fracture framework, and then fill in the order of decreasing particle sizes; Step S22, using numerical simulation software to simulate the distribution of proppants with different particle sizes in the fracture, and verifying the designed particle size gradient filling sequence; Step S23, adjusting the proppant particle size and injection parameters according to the simulation results to optimize the step filling design; In step S22, numerical simulation software is used to simulate the distribution of proppants with different particle sizes in the fracture. The specific operation is as follows: Step S221, establishing a fracture network model based on formation characteristics and fracture characteristics; ； in, Indicates the crack width; Indicates the amount of proppant; represents the normal stress; represents the radius of proppant particle size; represents the crack normal stiffness; Represents the base of the logarithm of natural numbers; Step S222: Establish a fracture conductivity model: ； in, Indicates the fracture conductivity; Indicates fracturing fluid viscosity; represents the proppant particle size; represents porosity; represents the permeability; Step S223, simulating the distribution of proppants of different particle sizes in the fracture: ； in, represents the distribution function; Indicates the injection flow; Indicates the length of the crack; Step S224, analyzing the fracture extension mode and proppant distribution shown in the simulation results, and evaluating the fracture conductivity; ； in, Represents the crack surface area.

2. The high-efficiency reservoir production-increasing fracturing method based on proppant with graded particle diameters according to claim 1, characterized in that: When filling the proppant in step S21, the method further includes: Step S211, pre-treating the selected proppants of different particle sizes; Step S212: After filling the large-size proppant to form a stable fracture framework, fill the remaining proppant in a layer-by-layer manner in the order of decreasing particle size, and perform a pressure holding operation after each layer of filling; Step S213: During the filling process, the filling speed and filling amount of the proppant are dynamically adjusted according to the real-time monitored fracture expansion and fluid flow conditions.

3. The high-efficiency reservoir production-increasing fracturing method based on proppant with graded particle diameters according to claim 2, characterized in that: In step S23, the proppant particle size and injection parameters are adjusted according to the simulation results to optimize the step filling design. The specific operations are as follows: Step S231, when the fracture conductivity does not reach the target conductivity, adjusting the particle size distribution of the proppant; ； in, represents the new proppant particle size; Indicates the target flow conductance; represents the adjustment factor; Step S232, adjusting the injection pressure and injection flow rate to optimize the fracture network expansion and support effect; ； in, Indicates injection pressure; Step S233: re-perform numerical simulation using the adjusted parameters to verify the optimization effect.

4. The high-efficiency reservoir production-increasing fracturing method based on proppant with graded particle diameters according to claim 3, characterized in that: In step S3, the fracturing parameters are optimized by an intelligent algorithm, and the specific operations are as follows: Step S31, according to the formation characteristics, fluid flow law and fracturing target of step S1 and the fracture characteristics and target production increase effect of step S2, historical fracturing data are collected, and the collected historical fracturing data are normalized to scale all features to the range of [0,1]; Historical fracturing data includes proppant particle size, injection pressure, injection rate, injection volume, fracture length, width, and production; Step S32, the number of input layer nodes of the neural network is set to 6, two hidden layers are set, each hidden layer has 12 neurons, and the number of output layer nodes is set to 1; Step S33: The hidden layer uses the ReLU activation function, the output layer uses the linear activation function, and the root mean square error is used as the loss function; Step S34: Use the learning rate decay strategy, halve the learning rate every 10 epochs, use L2 regularization to prevent overfitting, and the regularization parameter Set to 0.01; Step S35, using 5-fold cross validation to evaluate the generalization ability of the model; Step S36: Use the trained neural network model to predict the production under different fracturing parameter combinations, and optimize the fracturing parameters by combining the particle swarm algorithm based on dynamic neighbors and generalized learning to quantify the production; The particle swarm size is set to 30, the dimension is set to 6, and the inertia weight is set to Decrease from 0.9 to 0.4, the learning factor Set to 2. Set to 2; Step S37: Apply the optimized fracturing parameters to actual fracturing operations, and compare them with historical data to evaluate the optimization effect.

5. The high-efficiency reservoir production-increasing fracturing method based on proppant with graded particle diameters according to claim 4, characterized in that: In step S5, during the fracturing process, the expansion of the cracks, the filling of the proppant and the flow of the fluid are monitored in real time, and the fracturing parameters and the filling sequence of the proppant are adjusted according to the monitoring results. The specific operations are as follows: Step S51, deploying a microseismic monitoring system, a pressure sensor, and a flow meter to collect data on crack expansion, pressure change, and fluid flow in real time; Step S52: Use microseismic monitoring technology to track the expansion of the crack in real time and determine the length and width of the crack; Step S53, monitoring the pressure change in the fracture through a pressure sensor to evaluate the filling condition of the proppant; ； in, represents the proppant concentration; represents the proppant volume; represents the crack volume; Step S54, reading flow meter data to monitor the injection speed and total amount of fracturing fluid; ； in, Indicates the fracturing fluid velocity; Step S55, calculating the pressure change rate; ； in, express The pressure of the moment; express The pressure of the moment; Indicates a time interval; Step S56, adjusting the injection pressure and injection speed according to the pressure change rate and the crack expansion; Step S57, adjusting the particle size of the proppant according to the fracture width and the proppant concentration; Step S58: Establish a monitoring-adjustment cycle to monitor the fracturing process in real time and continuously adjust the fracturing parameters and proppant filling sequence according to the monitoring results.

6. The high-efficiency reservoir production-increasing fracturing method based on proppant with graded particle diameters according to claim 5, characterized in that: In step S6, after the fracturing operation is completed, the fracturing effect is evaluated by comparing the production data before and after fracturing, and the proppant selection, filling sequence, and fracturing parameters are further optimized according to the evaluation results. The specific operations are as follows: Step S61, collecting production data before and after fracturing, and calculating the production increase, where the production data is oil production; ； in, Indicates the change in output; Indicates the production after fracturing; represents the production before fracturing; Step S62, comprehensively comparing production data, evaluating fracture effects, and checking proppant distribution to determine the effect of the fracturing operation; Step S63, feeding back the evaluation result to the neural network model of step S3, adjusting the model parameters, and further training and optimizing the model; ； in, represents the updated model parameters; Represents the model parameters before updating; Indicates the update amount of the parameter.

7. The high-efficiency reservoir production-increasing fracturing method based on proppant with graded particle diameters according to claim 6, characterized in that: The evaluation results include the change in production , fracture length and width, proppant concentration .

8. Use of the efficient reservoir stimulation fracturing method based on graded proppant microparticle diameters as claimed in any one of claims 1 to 7 in oil field development.

Images