Supercritical carbon dioxide pulsing fluidization induced crack permeability device and method thereof

Abstract

This invention discloses a supercritical carbon dioxide pulsating fluidized bed fracturing and permeability enhancement device and method, relating to the field of coalbed methane development and coal seam permeability enhancement. It includes: a liquid CO2 storage and supply unit, a supercritical state preparation and pressurization unit, an intelligent pulsating flow generation unit, a data monitoring and artificial intelligence control unit, and a downhole sealing fracturing unit, all connected in a sealed manner. The data monitoring and artificial intelligence control unit includes a sensor monitoring module and an artificial intelligence optimization control module. The artificial intelligence optimization control module incorporates a hybrid strategy improved snake optimization algorithm, outputting a stable near-optimal combination of fracturing parameters based on real-time data collected by the sensor monitoring module. The intelligent pulsating flow generation unit modulates steady-state supercritical carbon dioxide into a pulsating fluid with controllable pulsation frequency and pressure amplitude, and delivers this pulsating fluid to the downhole sealing fracturing unit. The downhole sealing fracturing unit injects the pulsating fluid into the target fracturing section of the coal seam, thereby ensuring the original permeability of the coal seam gas flow channel.

Description

Technical Field

[0001] The embodiments of the present invention relate to the fields of coalbed methane development and coal seam permeability enhancement technology, and particularly to a supercritical carbon dioxide pulsed fluidization fracturing and permeability enhancement device and method. Background Technology

[0002] Poor permeability of low-permeability coal seams is a core bottleneck restricting the efficient development of coalbed methane. Existing permeability enhancement technologies have significant drawbacks: traditional hydraulic fracturing technology requires the injection of large amounts of water, which can easily trigger water-locking effects in the coal body, expansion of clay minerals, and coal sliming, leading to irreversible damage to reservoir permeability; static gas fracturing technology relies on a single pressure impact, resulting in a limited range of fracture expansion and a simple network structure, leading to poor permeability enhancement; at the same time, key parameters such as pulsation frequency and pressure amplitude of existing fracturing technologies are mostly set by human experience and cannot be dynamically adjusted according to the real-time state of coal seam fracture development, resulting in low energy utilization efficiency and difficulty in adapting to complex and variable coal seam geological conditions.

[0003] Although some studies have attempted to introduce intelligent algorithms to optimize fracturing parameters, traditional algorithms suffer from slow convergence speed, susceptibility to local optima, and a lack of integrated synergy between fracturing, displacement, and storage, making it difficult to meet the dual demands of green and efficient development and carbon emission reduction in low-permeability coal seams. Therefore, developing a permeability enhancement technology that causes no reservoir damage, has a wide fracturing range, intelligently optimizes parameters, and also offers environmental benefits has become an urgent need in the current coalbed methane development field. Summary of the Invention

[0004] The core objective of this invention is to overcome the defects of existing low-permeability coal seam permeability enhancement technologies and to provide a supercritical carbon dioxide pulsed fluidized fracturing device and method for enhancing permeability. The entire process uses anhydrous Sc-CO2 operation, with no water in contact with the coal body, completely avoiding water lock, water-sensitive damage and coal sliming problems caused by hydraulic fracturing, and ensuring the original permeability of the coal seam gas flow channel.

[0005] In a first aspect, embodiments of the present invention provide a supercritical carbon dioxide pulsed fluidized bed cracking and permeation enhancement device, comprising:

[0006] The system consists of a liquid CO2 storage and supply unit, a supercritical state preparation and pressurization unit, an intelligent pulsating flow generation unit, a data monitoring and artificial intelligence control unit, and a downhole sealing and fracturing unit, all connected in a sealed manner, forming a closed-loop operation chain.

[0007] The data monitoring and artificial intelligence control unit includes a sensor monitoring module and an artificial intelligence optimization control module. The artificial intelligence optimization control module has a built-in hybrid strategy improved snake optimization algorithm, which is used to maximize the fracture range and maximize the unit energy consumption permeability as dual optimization objectives. Based on the real-time data collected by the sensor monitoring module, it dynamically outputs a stable near-optimal fracture parameter combination that makes the fitness function converge, and controls the intelligent pulsating flow generation unit to adjust the pulsation frequency and pressure amplitude.

[0008] The intelligent pulsating flow generating unit is used to modulate the steady-state supercritical carbon dioxide output by the supercritical state preparation and pressurization unit into a pulsating fluid with controllable pulsating frequency and pressure amplitude, and to deliver the pulsating fluid to the downhole sealing and fracturing unit.

[0009] The downhole sealing and fracturing unit is used to inject the pulsating fluid into the target fracturing section of the coal seam and to monitor micro-fracture events in the coal body in real time.

[0010] As a preferred embodiment, the hybrid strategy improved snake optimization algorithm employs at least one of the following improved strategies: good point set initialization strategy, nonlinear balance factor exploration-exploitation mechanism, differential evolution strategy, and Lévy flight strategy.

[0011] In a preferred embodiment, the artificial intelligence optimization control module is configured as follows:

[0012] Based on the pressure fluctuation curve collected in real time by the sensing and monitoring module and the spatiotemporal distribution data of micro-fracture events located in real time by the downhole sealing and fracturing unit, the current expansion status of the fracture network is dynamically evaluated.

[0013] When it is determined that the coal seam is in the period of rapid fracture expansion, the intelligent pulsating flow generating unit is controlled to increase the pulsating frequency to match the resonance conditions of the newly formed fractures.

[0014] When it is determined that crack propagation is hindered, the intelligent pulsating flow generating unit is controlled to adjust the pressure amplitude to maintain the stress intensity factor at the crack tip.

[0015] In a preferred embodiment, the intelligent pulsating flow generating unit includes a high-speed rotary valve-type pulsating generator, a bladder-type accumulator, and a high-frequency response regulating valve; wherein...

[0016] The high-speed rotary valve type pulsation generator is used to modulate steady-state supercritical carbon dioxide into pulsating fluid.

[0017] The bladder-type accumulator is connected in parallel to the outlet pipe of the high-speed rotary valve type pulsation generator to absorb the energy of the pulsating pressure peak and release it in the trough, so as to achieve smooth correction of the pulsating waveform.

[0018] The high-frequency response regulating valve is used to fine-tune the pressure amplitude and waveform of the pulsating fluid.

[0019] In a preferred embodiment, the downhole sealing and fracturing unit includes a high-strength, high-pressure-resistant fracturing tube and a capsule-type expanded cement dual-stage sealing device; wherein...

[0020] The high-strength, high-pressure resistant fracturing tube is equipped with a directional jet nozzle at its front end, which is used to precisely guide the pulsating fluid to the target fracturing section of the coal seam.

[0021] The capsule-shaped expanded cement dual-stage sealing device is used to provide a dual-stage seal for the borehole, preventing leakage of high-pressure fluid.

[0022] In a preferred embodiment, the sensing and monitoring module includes a high-frequency pressure sensor, a high-frequency flow sensor, a temperature sensor, and a downhole microseismic monitoring system, and the data from each sensor are uploaded to the artificial intelligence optimization and control module in real time via a wireless transmission module.

[0023] Secondly, embodiments of the present invention also provide a method for supercritical carbon dioxide pulsed fluidization-induced fracturing and permeability enhancement using any of the above-described apparatuses, comprising:

[0024] In a low-permeability coal seam, a fracturing borehole is drilled, and the downhole sealing fracturing unit is placed into the borehole for pressurized sealing.

[0025] Steady-state supercritical carbon dioxide fluid is prepared using the liquid CO2 storage and supply unit and the supercritical state preparation and pressurization unit.

[0026] The coal seam physical property data is input into the artificial intelligence optimization control module to perform offline pre-initialization of the hybrid strategy improved snake optimization algorithm;

[0027] The steady-state supercritical carbon dioxide fluid is modulated into a pulsating fluid and injected into the coal seam through the intelligent pulsating flow generation unit, while the fracturing process data is collected in real time through the sensing and monitoring module.

[0028] The real-time collected fracturing process data is input into the hybrid strategy improved snake optimization algorithm for online optimization, and the stable near-optimal pulsation frequency and pressure amplitude that make the fitness function converge are dynamically output.

[0029] The intelligent pulsating flow generating unit adjusts the pulsating frequency and pressure amplitude in real time based on the output stable near-optimal pulsating frequency and pressure amplitude to induce pulsating fracturing in the coal seam.

[0030] In a preferred embodiment, after completing the pulsating fracturing, the method further includes a well-clogging process and a negative pressure extraction process, wherein...

[0031] The well-steaming process uses supercritical carbon dioxide to replace and displace coalbed methane.

[0032] During the negative pressure extraction process, the extracted gas and carbon dioxide mixture is separated by a ground separation device, and the separated carbon dioxide is returned to the liquid CO2 storage and supply unit for recycling.

[0033] In a preferred embodiment, the step of inputting the real-time collected fracturing process data into the hybrid strategy improved snake optimization algorithm for online optimization, and dynamically outputting a stable near-optimal pulsation frequency and pressure amplitude that makes the fitness function converge, includes:

[0034] Using real-time collected data on the fracturing process as input, and with the dual objectives of maximizing the fracturing range and maximizing the permeability increase per unit energy consumption, an initial snake population is generated through a good point set initialization strategy.

[0035] The global exploration and local development phases of the snake optimization algorithm are improved by using a hybrid strategy of dynamic switching of nonlinear balance factors.

[0036] The game incorporates Lévy flight strategy in combat mode and differential evolution strategy in mating mode to iteratively update the snake population position.

[0037] When the maximum number of iterations or the fitness value converges, a stable and near-optimal pulsation frequency and pressure amplitude are output and converted into control commands to be sent to the intelligent pulsation flow generation unit.

[0038] As a preferred embodiment, during the pulsating fracturing process, the artificial intelligence optimization control module monitors the fracturing pressure, micro-fracture events, and permeability enhancement rate per unit energy consumption in real time.

[0039] When the amplitude of the fracturing pressure fluctuation is less than the first preset threshold, the spatial density of micro-fracture events is lower than the second preset threshold, and the continuous iterative increment of the unit energy consumption permeability is less than the third preset threshold, the artificial intelligence optimization control module determines that fracturing is sufficient and automatically terminates the operation at this stage.

[0040] Compared with existing technologies, the present invention achieves the following beneficial effects:

[0041] (1) The present invention uses supercritical carbon dioxide operation throughout the process, with no water in contact with the coal body, thus eliminating the water-locking effect, clay mineral expansion and coal slime problems caused by traditional hydraulic fracturing from the root. It completely preserves the original gas flow channel of the coal seam, has stronger adaptability to soft and low-permeability coal seams, and ensures that the reservoir permeability is not irreversibly damaged.

[0042] (2) This invention utilizes the fatigue damage effect of pulsating load to cause repeated opening and closing of the coal body fracture tip, achieving long-distance extension of the fracture under working conditions lower than the static fracture pressure; combined with the physical characteristics of low viscosity and high diffusion of supercritical carbon dioxide, it effectively connects isolated pores and primary fractures, constructing a three-dimensional interwoven complex fracture network, increasing the fracture radius by more than 2 times compared to static carbon dioxide fracturing, significantly expanding the permeability range and improving the permeability of the coal seam.

[0043] (3) The hybrid strategy improved snake optimization algorithm built into the artificial intelligence control unit of this invention completely solves the technical pain points of slow convergence speed and easy getting trapped in local optima of traditional algorithms through four major improvement strategies such as good point set initialization and nonlinear balance factor exploration-development mechanism; relying on sensor monitoring data and downhole micro-seismic monitoring results, it dynamically optimizes 5-50Hz pulsation frequency and ±2-±4MPa pressure amplitude in real time to ensure that the fracturing parameters and coal seam fracture expansion state are accurately matched, the unit energy consumption permeability is increased by more than 30%, and the stable near-optimal solution is output through iterative convergence, ensuring the reliability and stability of engineering operations.

[0044] (4) This invention innovatively integrates three core functions: supercritical carbon dioxide physical fracturing and permeability enhancement, competitive adsorption of gas displacement, and carbon dioxide geological storage. While increasing the concentration and output of coalbed methane extraction, it permanently stores industrial waste gas carbon dioxide in the cracks and pores of the coal seam, simultaneously achieving the goals of green and efficient development of coalbed methane and carbon emission reduction, with significant economic value and environmental benefits.

[0045] (5) The present invention adopts a modular architecture, each functional unit is independently controllable, easy to disassemble and transport, and can be flexibly adapted to burial depths of 500-1500m and original air permeability <1×10 -18 The operation scenario is a low-permeability coal seam with a diameter of m². The operation method and steps are clear and standardized, and it is highly compatible with existing coalbed methane development processes. It does not require large-scale modification of existing equipment, lowers the threshold for engineering promotion, and has broad prospects for field application. Attached Figure Description

[0046] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings. The drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0047] Figure 1 This is a schematic diagram of the structure of the supercritical carbon dioxide pulsating fluidized bed cracking and permeation enhancement device provided in an embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram of the downhole sealing and fracturing unit structure provided in an embodiment of the present invention;

[0049] Figure 3 This is a flowchart of the supercritical carbon dioxide pulsating fluidization-induced cracking and permeability enhancement method provided in the embodiments of the present invention;

[0050] Figure 4 This is a logical block diagram of the Hybrid Strategy Improved Snake Optimization (HSO) algorithm provided in this embodiment of the invention;

[0051] Figure 5 This is a schematic diagram of the intelligent closed-loop control logic provided in an embodiment of the present invention;

[0052] Figure 6 This is a schematic diagram of the pulsating waveform correction principle provided in an embodiment of the present invention. Detailed Implementation

[0053] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0054] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe operations (or steps) as being processed sequentially, many of these operations (or steps) may be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations may be rearranged. The process may be terminated when its operation is completed, but may also have additional steps not included in the figures. The process may correspond to a method, function, procedure, subroutine, subroutine, etc.

[0055] Example 1

[0056] like Figure 1 The diagram shown is a structural schematic of a supercritical carbon dioxide pulsed fluidization fracturing and permeability enhancement device provided in Embodiment 1 of the present invention. The device specifically includes:

[0057] The system consists of a liquid CO2 storage and supply unit, a supercritical state preparation and pressurization unit, an intelligent pulsating flow generation unit, a data monitoring and artificial intelligence control unit, and a downhole sealing and fracturing unit, all of which are sequentially sealed and connected, and together they form a closed-loop operation link.

[0058] Preferably, the liquid CO2 storage and supply unit consists of a liquid CO2 storage tank and a cryogenic delivery pump, providing a stable supply of liquid CO2 raw material for the entire system; the flow rate of the cryogenic delivery pump is adjustable from 5 to 50 L / min, and the raw material supply can be adjusted in real time according to the downhole fracturing requirements; the storage tank is equipped with a cryogenic pressure maintaining device to ensure the storage stability of liquid CO2.

[0059] Preferably, the supercritical state preparation and pressurization unit includes a high-pressure plunger pump and a heat exchange heating system, which is the core of supercritical carbon dioxide (Sc-CO2) preparation. The high-pressure plunger pump can pressurize liquid CO2 to 15-30 MPa (meeting the coal seam fracturing pressure requirements and exceeding the CO2 critical pressure of 7.38 MPa). The heat exchange heating system adopts electric heating to raise the temperature of the pressurized CO2 to 35-60℃ (exceeding the CO2 critical temperature of 31.1℃), so that CO2 is stably in the supercritical state and has the physical characteristics of low viscosity, high diffusivity and high solubility.

[0060] Preferably, the intelligent pulsating flow generating unit is used to modulate the steady-state supercritical carbon dioxide Sc-CO2 output from the supercritical state preparation and pressurization unit into a controllable pulsating fluid with a specific frequency and pressure amplitude, and to transport the pulsating fluid to the downhole sealing and fracturing unit. Its specific structure and corresponding functions include:

[0061] (1) High-speed rotary valve type pulsation generator: The valve core with special flow channel groove is driven by a variable frequency motor to rotate at high speed. The pulsation flow is prepared by periodically opening / closing the Sc-CO2 flow path. The speed of the variable frequency motor is adjustable from 60 to 4800 r / min, and the corresponding pulsation frequency is adjustable from 1 to 80 Hz (preferably 5 to 50 Hz). The basic adjustable range of pressure amplitude is ±1 to ±5 MPa (preferably ±2 to ±4 MPa).

[0062] (2) Bag accumulator: connected in parallel to the outlet of the above-mentioned high-speed rotary valve type pulsation generator, used to absorb excess energy at the pressure peak and release energy at the trough, realize smooth correction of pulsation waveform, prevent waveform distortion, and at the same time improve energy utilization efficiency and protect downstream pipelines.

[0063] (3) High-frequency response regulating valve: As an auxiliary regulating component, the regulating cycle can be 5s, which is used to finely adjust the pressure amplitude and waveform of the pulsating fluid, and to achieve precise control of the cracking parameters in conjunction with the pulsation generator.

[0064] Preferably, the data monitoring and artificial intelligence control unit includes a sensor monitoring module and an artificial intelligence optimization control module, used to dynamically adjust the speed of the variable frequency motor (controlling the pulsation frequency) and the opening degree (controlling the pressure vibration frequency response adjustment valve amplitude), so as to achieve real-time matching between fracturing parameters and coal seam fracture propagation state. The composition and function of each module are as follows:

[0065] The sensing and monitoring module consists of high-frequency pressure sensors (sampling frequency 100Hz, measurement accuracy ±0.01MPa), high-frequency flow sensors (sampling frequency 50Hz, measurement accuracy ±0.1L / min), and temperature sensors (measurement range -20~100℃, accuracy ±0.5℃) installed at each pipeline node, as well as a downhole microseismic monitoring system. The sensors are seamlessly connected to the fracturing pipe of the downhole sealing and fracturing unit and the surface pipeline, and collect dynamic parameters such as pressure, flow rate (preferably 10-30L / min), temperature, pulsation frequency, and pressure amplitude of Sc-CO2 in real time. The microseismic monitoring system realizes the real-time location of coal seam micro-fracture events, and all data is uploaded to the artificial intelligence optimization control module of the ground control center in real time through a wireless transmission module.

[0066] The artificial intelligence optimization control module, deployed in an industrial computer at the ground control center, incorporates a hybrid strategy improved snake optimization algorithm (HSO algorithm). With the dual optimization objectives of maximizing the crack range and maximizing the unit energy consumption penetration rate, it analyzes and processes the real-time data collected by the sensing and monitoring module, and outputs stable near-optimal control commands that converge the fitness function to the intelligent pulsating flow generation unit.

[0067] The HSO algorithm described above is based on the simulation of snake foraging, fighting, and mating behaviors. It integrates four improved strategies: good point set initialization, nonlinear balance factor exploration-development mechanism, differential evolution strategy, and Lévy flight strategy. The core of the algorithm consists of five stages: population initialization, dynamic switching between exploration-development stages, fighting mode (integrated with Lévy flight), mating mode (integrated with differential evolution), and population update. It simulates the search process for rupture parameters by iteratively updating the positions of individual snakes. The snake's position corresponds to the combination of rupture parameters (pulsation frequency, pressure amplitude), and the fitness value corresponds to the comprehensive evaluation result of the dual optimization objectives (rupture range, unit energy consumption permeability). The computational complexity of the algorithm is O(n). ,in The maximum number of iterations, For population size, To optimize the parameter dimension (Dim in this invention) That is, the pulsation frequency and the pressure amplitude.

[0068] Preferably, this embodiment will combine the CEC2022 benchmark function test with the actual needs of coal seam fracturing engineering to determine the optimal parameter combination of the HSO algorithm (verified by parameter sensitivity experiments). The specific key parameters are as follows:

[0069] (1) Population size Parameter dimensions Maximum number of iterations ;

[0070] (2) Nonlinear equilibrium factor parameters , ( , (This allows for manual parameter setting to adjust the shape of the balance factor curve).

[0071] (3) Lévy flight parameters (Control the flight step size to avoid algorithm oscillation or local convergence);

[0072] (4) The threshold Threshold1 for the exploration phase is 0.25, and the temperature threshold Threshold2 for the development phase is 0.6;

[0073] (5) Scaling factor for differential evolution strategy (The scaling factor of the control deviation variable) The periodic coefficient of the cosine operator is coupled with the number of iterations;

[0074] (6) The ratio of males to females in a snake group ( The number of male snakes, (Number of female snakes).

[0075] Preferably, the core improved formula of the above HSO algorithm is as follows, and the remaining basic formulas refer to the original SO algorithm. The improved formula ensures a balance between the algorithm's global exploration and local exploitation capabilities, avoiding local optima:

[0076] (1) Initialization formula for good point set: to achieve uniform distribution of population and improve the comprehensiveness of search.

[0077] ;

[0078] In the formula: For the first Initial position of each individual snake (combination of splitting parameters); , Search for upper and lower limits (frequency) for parameters , ,amplitude , ); The uniformly distributed points generated for the good point set.

[0079] (2) Nonlinear balance factor formula: realizes dynamic switching between the exploration and development phases, and balances the global search and local optimization capabilities of the algorithm.

[0080] ;

[0081] In the formula: As a balance factor; This represents the current iteration number; This represents the maximum number of iterations. , This is the balance factor parameter; when random numbers The algorithm performs a global search. Partial development is performed at certain times.

[0082] (3) Differential evolution strategy position update formula (female / male snake): Improves population diversity and solves the local convergence problem in the later stage of the algorithm:

[0083] ;

[0084] In the formula: Male snake Individuals The location at a given moment corresponds to the combination of fracturing parameters (pulsation frequency, pressure amplitude). Female snake Individuals The location at a given moment corresponds to the combination of fracturing parameters (pulsation frequency, pressure amplitude). 1: Mutually exclusive random integers, with a value range of [value range missing]. (Male snake group) or (Female snake group), used to select the individual sequence number to participate in differential evolution; : Mutually exclusive random integers, with a value range of 1000. (Male snake group) or (Female snake group), used to select the individual sequence number to participate in differential evolution; : Mutually exclusive random integers, with a value range of 1000. (Male snake group) or (Female snake group), used to select the individual sequence number to participate in differential evolution; : Current iteration number, the current loop count of the algorithm; : Maximum number of iterations, with a value of 2000, is the maximum number of loops the algorithm must complete to optimize parameters; : time The corresponding position of the male snake individual corresponds to the combination of rupture parameters at that moment; : time The corresponding position of the male snake individual corresponds to the combination of rupture parameters at that moment; : time The corresponding position of the male snake individual corresponds to the combination of rupture parameters at that moment; : time The corresponding position of the female snake individual corresponds to the combination of rupture-causing parameters at that moment; : time The corresponding position of the female snake individual corresponds to the combination of rupture-causing parameters at that moment; : time The corresponding position of the female snake individual corresponds to the combination of rupture-causing parameters at that moment; The cosine operator, coupled with the number of iterations, is used to adjust the periodicity of position updates in differential evolution strategies, thereby improving population diversity.

[0085] (4) Lévy flight strategy formula (integrated into combat mode): expands the search range and enables the algorithm to escape from local optima:

[0086]

[0087] ;

[0088] In the formula, For Lévy's flight stride; , ( , Normally distributed random numbers, , (where the standard deviation corresponds to the normal distribution). For Lévy's flight parameters; The fighting ability of male snakes; This is the optimal position for the female snake; It is a constant; The food quantity factor is represented by `rand`, which is a random number between 0 and 1. For male snakes Individuals Location at any given moment For male snakes Individuals The location at any given moment.

[0089] Preferably, in response to the actual needs of coal seam fracturing in this invention, the input and output of the HSO algorithm are directly related to engineering parameters to achieve seamless conversion of algorithm optimization results into control commands:

[0090] HSO algorithm input: The parameter set collected in real time by the sensing and monitoring module ,in The fracturing pressure (unit: MPa). Sc-CO2 injection flow rate (unit: ), The current pulse frequency (unit: Hz). This represents the current pressure amplitude (unit: MPa). Sc-CO2 temperature (unit: ), Energy per unit energy consumption per unit of light penetration (unit: Simultaneously input downhole microseismic monitoring data and parameter search upper and lower limits (frequency). ,amplitude ).

[0091] HSO algorithm output: a stable near-optimal combination of fracture-inducing parameters that makes the fitness function converge. ,in The near-optimal pulsation frequency (unit: Hz). The near-optimal pressure amplitude (unit: MPa) is then converted into control commands: Near-optimal speed of the variable frequency motor (unit: ), Near optimal opening degree of the corresponding high-frequency response control valve (unit: The command is transmitted in real time to the intelligent pulse flow generator unit via the wireless transmission module.

[0092] Preferably, the training of the HSO algorithm is divided into two stages: offline pre-training and online real-time optimization. First, the parameters of the algorithm model are calibrated through offline pre-training, and then the crack-causing parameters are dynamically adjusted through online real-time optimization. The specific steps are as follows:

[0093] (1) Offline pre-training stage

[0094] ①Based on historical data of coal seam physical properties, laboratory coal and rock sample test data, and underground microseismic monitoring data, a sample set of fracturing parameters and fracture propagation effects was constructed;

[0095] ② Input the sample set into the HSO algorithm, with the dual objectives of maximizing the fracture range and maximizing the penetration rate per unit energy consumption, and perform iterative training of the algorithm to calibrate the key parameters of the algorithm. wait);

[0096] ③ The algorithm training results were statistically verified using the Friedman test and the Wilcoxon signed-rank test to ensure the effectiveness and stability of the algorithm in finding the best candidate.

[0097] (2) Online real-time optimization stage

[0098] The sensing and monitoring module and the downhole microseismic monitoring system collect parameters such as pressure, flow rate, frequency, amplitude, and location of micro-fractures in the coal seam in real time during the fracturing process, and upload them to the artificial intelligence optimization and control module;

[0099] The HSO algorithm takes real-time collected parameters as input, initializes the snake population (combination of splitting parameters), and calculates the fitness value of each individual (comprehensive evaluation of dual optimization objectives).

[0100] The exploration-development phase is dynamically switched through a nonlinear balance factor, the combat mode is integrated with Lévy flight, and the mating mode is integrated with differential evolution, and the snake population position is iteratively updated.

[0101] When the algorithm reaches the maximum number of iterations or the fitness value converges, it outputs a stable near-optimal fracture-causing parameter combination that makes the fitness function converge, and converts it into control commands and sends them to the intelligent pulsating flow generation unit.

[0102] Repeat the above steps to achieve real-time, cyclic optimization of the fracturing parameters and form a closed-loop control.

[0103] The training data for the HSO algorithm is a fusion of multiple data sources to ensure the comprehensiveness and authenticity of the data. Specific data sources include:

[0104] Historical physical property data of coal seams: geological parameters of the target coal seam, including tensile strength (MPa), elastic modulus (GPa), geostress (MPa), original permeability (m²), and porosity (%).

[0105] Laboratory coal and rock sample test data: natural frequency (unit: Hz), static fracture pressure (unit: MPa), fatigue damage characteristics and other indoor experimental data of coal and rock samples;

[0106] On-site fracturing monitoring data: Real-time dynamic operational data such as pressure, flow rate, frequency, and amplitude collected by the sensor monitoring module, as well as coal micro-fracture event data collected by the underground micro-vibration monitoring system;

[0107] Historical operational data for similar coal seam fracturing projects: Data from domestic and international engineering cases on parameter settings and permeability enhancement effects of CO2-induced fracturing in low-permeability coal seams. Specifically, fracturing range data was obtained by real-time location of micro-fracture events using a downhole microseismic monitoring system, and the permeability enhancement rate per unit energy consumption was calculated by comprehensively considering the injection flow rate, pressure, and extraction flow rate increments.

[0108] Preferably, the downhole sealing and fracturing unit is used to inject the pulsating fluid into the target fracturing section of the coal seam and to monitor micro-fracture events in the coal body in real time. It includes a high-strength, high-pressure-resistant fracturing tube and a capsule-expanding cement dual-stage sealing device, and is suitable for deep hole (≥50m) pressurized (≥20MPa) fracturing operations. The specific structure and function are as follows:

[0109] High-strength, high-pressure resistant fracturing tube: Made of alloy steel, with a pressure resistance of ≥40MPa. The front end of the tube is equipped with directional jet nozzles (4-8 nozzles, 5-10mm in diameter) to ensure that the Sc-CO2 pulsating fluid acts precisely on the target fracturing segment and avoids irregular fluid diffusion.

[0110] Capsule-expanded cement dual-stage sealing device: It consists of an expanding capsule and an expanding cement grouting section to achieve dual-stage sealing of the borehole. The sealing pressure resistance is ≥20MPa, which can effectively prevent high-pressure Sc-CO2 fluid from leaking from the borehole opening and ensure that the fracturing pressure is effectively applied to the coal seam.

[0111] Preferably, the working process of the above-mentioned supercritical carbon dioxide pulsed fluidized bed cracking and permeation enhancement device is as follows:

[0112] The liquid CO2 storage and supply unit is started, and the cryogenic transfer pump extracts liquid CO2 from the storage tank and delivers it to the supercritical state preparation and pressurization unit;

[0113] The supercritical state preparation and pressurization unit is started. The high-pressure plunger pump pressurizes the liquid CO2 to 15-30MPa, and the heat exchange heating system heats the pressurized CO2 to 35-60℃ to prepare a steady-state Sc-CO2 fluid. The supercritical state stability is maintained by real-time monitoring through temperature and pressure sensors.

[0114] Start the data monitoring and artificial intelligence control unit, input the historical physical property data of the target coal seam and the test data of laboratory coal and rock samples into the artificial intelligence optimization control module, complete the offline pre-initialization of the HSO algorithm, and set the initial value of the pulsation frequency to 5-50Hz and the initial value of the pressure amplitude to ±2-±4MPa;

[0115] The intelligent pulsating flow generation unit is activated, and the high-speed rotary valve type pulsating generator modulates the steady-state Sc-CO2 into pulsating fluid. After the waveform is smoothed by the bladder accumulator and the amplitude is finely adjusted by the high-frequency response regulating valve, it is injected into the target fracturing section of the coal seam through the directional jet nozzle of the high-strength and high-pressure resistant fracturing tube.

[0116] During the fracturing process, the sensing and monitoring module and the downhole microseismic monitoring system collect and upload various parameters in real time. The HSO algorithm performs online real-time optimization, dynamically outputs near-optimal fracturing parameters and converts them into control commands, realizing intelligent closed-loop control that monitors, optimizes and fracturing simultaneously.

[0117] After the fracturing operation is completed, the relevant valves are closed, and subsequent operations such as well shut-in, gas replacement and displacement, negative pressure extraction and CO2 geological sealing are carried out in sequence.

[0118] Based on the above embodiments, the core beneficial effects of the present invention are as follows:

[0119] (1) The present invention uses supercritical carbon dioxide operation throughout the process, with no water in contact with the coal body, thus eliminating the water-locking effect, clay mineral expansion and coal slime problems caused by traditional hydraulic fracturing from the root. It completely preserves the original gas flow channel of the coal seam, has stronger adaptability to soft and low-permeability coal seams, and ensures that the reservoir permeability is not irreversibly damaged.

[0120] (2) This invention utilizes the fatigue damage effect of pulsating load to cause repeated opening and closing of the coal body fracture tip, achieving long-distance extension of the fracture under working conditions lower than the static fracture pressure; combined with the physical characteristics of low viscosity and high diffusion of supercritical carbon dioxide, it effectively connects isolated pores and primary fractures, constructing a three-dimensional interwoven complex fracture network, increasing the fracture radius by more than 2 times compared to static carbon dioxide fracturing, significantly expanding the permeability range and improving the permeability of the coal seam.

[0121] (3) The hybrid strategy improved snake optimization algorithm built into the artificial intelligence control unit of this invention completely solves the technical pain points of slow convergence speed and easy getting trapped in local optima of traditional algorithms through four major improvement strategies such as good point set initialization and nonlinear balance factor exploration-development mechanism; relying on sensor monitoring data and downhole micro-seismic monitoring results, it dynamically optimizes 5-50Hz pulsation frequency and ±2-±4MPa pressure amplitude in real time to ensure that the fracturing parameters and coal seam fracture expansion state are accurately matched, the unit energy consumption permeability is increased by more than 30%, and the stable near-optimal solution is output through iterative convergence, ensuring the reliability and stability of engineering operations.

[0122] (4) This invention innovatively integrates three core functions: supercritical carbon dioxide physical fracturing and permeability enhancement, competitive adsorption of gas displacement, and carbon dioxide geological storage. While increasing the concentration and output of coalbed methane extraction, it permanently stores industrial waste gas carbon dioxide in the cracks and pores of the coal seam, simultaneously achieving the goals of green and efficient development of coalbed methane and carbon emission reduction, with significant economic value and environmental benefits.

[0123] (5) The present invention adopts a modular architecture, each functional unit is independently controllable, easy to disassemble and transport, and can be flexibly adapted to burial depths of 500-1500m and original air permeability <1×10 -18 The operation scenario is a low-permeability coal seam with a diameter of m². The operation method and steps are clear and standardized, and it is highly compatible with existing coalbed methane development processes. It does not require large-scale modification of existing equipment, lowers the threshold for engineering promotion, and has broad prospects for field application.

[0124] Example 2

[0125] Figure 3 This is a flowchart of a method 30 for supercritical carbon dioxide pulsed fluidization-induced cracking and permeability enhancement provided in Embodiment 2 of the present invention, as shown below. Figure 3 As shown, the method includes:

[0126] Step S31: Construct a fracturing borehole in the low-permeability coal seam, and place the downhole sealing fracturing unit into the borehole for pressurized sealing.

[0127] Preferably, in this embodiment, based on the geological conditions of the target coal seam, including burial depth, in-situ stress, and coal seam thickness, a fracturing borehole of a preset depth (≥50m) is constructed in the coal seam, with the borehole diameter adapted to the downhole sealing fracturing unit; a high-strength, high-pressure resistant fracturing pipe is lowered into the borehole to the target fracturing section, and a downhole microseismic monitoring system is simultaneously deployed; a capsule-expanded cement dual-stage sealing device is used to perform pressurized grouting sealing of the borehole; after sealing, a pressure holding test is performed, with a pressure holding pressure ≥20MPa and a pressure holding time ≥30min. No pressure decay indicates that the sealing is qualified, ensuring that the sealing section is tight and reliable and preventing high-pressure Sc-CO2 leakage.

[0128] Step S32: Prepare steady-state supercritical carbon dioxide fluid through the liquid CO2 storage and supply unit and the supercritical state preparation and pressurization unit.

[0129] Preferably, the liquid CO2 storage and supply unit is started first, and the liquid CO2 is extracted from the storage tank and transported to the supercritical state preparation and pressurization unit by a cryogenic transfer pump;

[0130] Secondly, the supercritical state preparation and pressurization unit is started. The liquid CO2 is pressurized to 15-30MPa by a high-pressure plunger pump, and the heat exchange heating system heats the pressurized CO2 to 35-60℃, so that the CO2 is stabilized in the supercritical state and prepared into a steady-state Sc-CO2 fluid.

[0131] Meanwhile, during the preparation process, temperature and pressure sensors are used to monitor the process in real time to ensure that the temperature and pressure of Sc-CO2 are maintained within the set range.

[0132] Step S33: Input the coal seam physical property data into the artificial intelligence optimization control module to perform offline pre-initialization of the hybrid strategy improved snake optimization algorithm.

[0133] Preferably, the data monitoring and artificial intelligence control unit is activated, and all historical physical property data of the target coal seam and laboratory coal and rock sample test data are input into the artificial intelligence optimization control module within the unit. This module has a built-in hybrid strategy improved snake optimization algorithm (HSO algorithm) of the present invention, which provides a comprehensive and real data source for the offline pre-training of the algorithm.

[0134] The offline pre-training and initialization of the hybrid strategy improved snake optimization algorithm includes the following steps:

[0135] ① Based on the estimated natural frequency of the coal seam, the initial value of the pulsation frequency is set to 5-50Hz (preferred frequency range);

[0136] ② Set the initial pressure amplitude to ±2-±4MPa (medium to low level, to avoid brittle fracture of coal seam caused by excessive initial pressure);

[0137] ③ Initialize the good point set of the snake population and determine the initial parameter combination for algorithm iteration, laying the foundation for subsequent online real-time optimization.

[0138] Step S34: The steady-state supercritical carbon dioxide fluid is modulated into a pulsating fluid by the intelligent pulsating flow generating unit and injected into the coal seam, while the fracturing process data is collected in real time by the sensing and monitoring module.

[0139] Preferred, combined Figure 6 As shown, the intelligent pulsating flow generation unit is activated. The high-speed rotary valve pulsating generator in the unit modulates the steady-state Sc-CO2 into pulsating fluid under the drive of the variable frequency motor. The bladder accumulator is connected in parallel to the outlet of the pulsating generator to smooth the waveform of the pulsating fluid and prevent waveform distortion. The high-frequency response regulating valve simultaneously and precisely fine-tunes the pressure amplitude and waveform of the fluid. Finally, the parameter-controllable pulsating supercritical carbon dioxide fluid is injected into the target fracturing section of the coal seam through the directional jet nozzle at the front end of the high-strength high-pressure resistant fracturing tube. The injection flow rate is controlled at 10-30 L / min.

[0140] After the injection of pulsating supercritical carbon dioxide fluid, the pulsating pressure wave acts directly on the pores and fracture ends of the coal body, causing cyclic tensile and compressive stress in the coal body. Combined with the low viscosity and high diffusivity of supercritical carbon dioxide, the fluid can penetrate deep into the micropores of the coal body, causing the pores to expand and contract in a breathing manner, continuously reducing the fracture toughness of the coal body, and inducing microcracks in the coal body to initiate and stably expand under conditions lower than the static fracture pressure. As the pulsating injection operation continues, the original fractures of the coal seam continue to extend and interconnect through the newly formed microcracks, eventually forming a complex and interwoven fracture network, which greatly improves the permeability of the coal seam.

[0141] Step S35: Input the real-time collected fracturing process data into the hybrid strategy improved snake optimization algorithm for online optimization, and dynamically output the stable near-optimal pulsation frequency and pressure amplitude that make the fitness function converge.

[0142] As a preferred embodiment, during the fracturing process, the sensor monitoring module of the data monitoring and artificial intelligence control unit collects parameters such as pressure, flow rate, frequency, and amplitude of Sc-CO2 in real time at a sampling frequency of 100 Hz. The downhole micro-vibration monitoring system locates coal micro-fracture events in real time, and all data is uploaded to the artificial intelligence optimization control module simultaneously.

[0143] Combination Figure 4 As shown, the HSO algorithm in the artificial intelligence optimization control module takes the parameters collected in real time as input, performs online real-time optimization, and dynamically outputs a stable near-optimal pulsation frequency that makes the fitness function converge. (5-50 Hz) and pressure amplitude It is then converted into control commands and sent to the intelligent pulsating flow generator unit to adjust the speed of the variable frequency motor and the opening of the high-frequency response regulating valve;

[0144] For example, when the algorithm detects a rapid decay of fracturing pressure and a wide-ranging spread of micro-fracture events in the coal seam, it automatically determines that the coal seam fractures are in a high-speed propagation stage. It then slightly increases the pulsation frequency to match the resonance conditions of the newly formed fractures, or adjusts the pressure amplitude to maintain the stress intensity factor at the crack tip, ensuring that the fracturing parameters and the coal seam fracture propagation state are matched accurately in real time, thus achieving intelligent closed-loop control that monitors, optimizes, and induces fracturing simultaneously.

[0145] Combination Figure 5 As shown, the duration of single-hole fracturing operation is automatically determined by the artificial intelligence control unit based on the pressure response curve and microseismic monitoring data. When the fracturing pressure tends to stabilize, the micro-fracture events in the coal body do not spread significantly, and the permeability per unit energy consumption no longer increases, it is determined that the coal seam fracturing is sufficient, and the algorithm automatically issues a termination command to end the fracturing operation of the borehole. The single-hole fracturing time is precisely controlled to be 15-60 minutes.

[0146] Step S36: Based on the output stable near-optimal pulsation frequency and pressure amplitude, the intelligent pulsation flow generating unit is controlled to adjust the pulsation frequency and pressure amplitude in real time to induce pulsation fracturing in the coal seam.

[0147] Preferably, the artificial intelligence optimization control module converts the stable near-optimal pulsation frequency and pressure amplitude output by the hybrid strategy improved snake optimization algorithm into precise execution control commands, which are then sent to the intelligent pulsating flow generation unit in real time. The unit dynamically adjusts the speed of the variable frequency motor and the opening of the high-frequency response regulating valve according to the commands, thereby achieving real-time and precise control of the pulsation frequency and pressure amplitude of the pulsating fluid. This ensures that the pulsating supercritical carbon dioxide fluid always acts on the coal seam with parameters adapted to the coal seam fracture propagation state, completing the precise pulsating fluidization fracturing of the coal seam, maximizing the fracturing effect, and improving energy utilization efficiency.

[0148] Preferably, after the fracturing operation is completed, the process also includes well shut-in and gas extraction operations. Specifically, the well shut-in operation includes:

[0149] After the fracturing operation is completed, immediately shut down the intelligent pulsating flow generator and the downhole borehole valve, and carry out a well shut-in operation for 2-4 hours;

[0150] During the well-sealing process, the high adsorption capacity and extraction characteristics of supercritical carbon dioxide are utilized to achieve a dual permeability enhancement and displacement effect: on the one hand, supercritical carbon dioxide competes with methane adsorbed on the surface of the coal matrix for adsorption, rapidly achieving methane displacement and desorption, and improving gas desorption efficiency; on the other hand, supercritical carbon dioxide can extract some small molecule organic matter in the coal body, effectively clearing the blocked pores and throats in the coal body, further improving the permeability of the coal seam, and laying the foundation for subsequent gas extraction.

[0151] The gas extraction operation includes: after the well shut-in operation is completed, the downhole sealing and fracturing unit is connected to the gas extraction system for negative pressure extraction. The extraction negative pressure is controlled at -0.05~-0.1MPa, and the extraction time is ≥72 hours. The extracted gas is a mixture of CH4 and CO2. After purification by the surface gas separation device, the high-concentration CH4 is used as a clean energy resource. The separated CO2 can be transported back to the liquid CO2 storage and supply unit and recycled for coal seam fracturing operations to achieve resource recycling. A large amount of Sc-CO2 is sealed in the fractures and pores of the coal seam through physical adsorption and free state to achieve geological CO2 sequestration. During the extraction process, the coal seam permeability and CO2 sequestration rate are continuously monitored until the permeability and extraction concentration tend to stabilize.

[0152] Preferably, to verify the effectiveness and superiority of the hybrid strategy improved snake optimization algorithm in the optimization of fracture-causing parameters, this invention conducts a special verification experiment. The experiment is based on the CEC2022 benchmark function set and is carried out only in 2-dimensional space, corresponding to the two optimization dimensions of the fracture-causing parameter combination in this invention, namely pulsation frequency and pressure amplitude. The snake optimization algorithm (SO), whale optimization algorithm (WOA), and chameleon swarm algorithm (CSA) are selected as baseline comparison algorithms.

[0153] The experiment was conducted with uniform parameters: the population size of all algorithms was set to 50, the maximum number of iterations was 2000, and each algorithm was run independently 30 times. The experimental results were verified by Friedman test to confirm the overall performance difference between the algorithms, and by Wilcoxon signed-rank test to confirm the pairwise difference between HSO and each baseline algorithm (significance level = 0.05).

[0154] The Friedman test evaluates the overall performance of each algorithm by calculating the average rank of each algorithm across all test functions. A smaller average rank indicates a stronger comprehensive optimization capability. The test results are shown in Table 1. In the 2D fracturing parameter optimization scenario, the hybrid strategy improved snake optimization algorithm of this invention exhibits the smallest average rank across the three core indicators: average optimization result, result stability, and optimal solution search capability. This significantly outperforms the original snake optimization algorithm and two classic metaheuristic algorithms: the whale optimization algorithm and the chameleon swarm algorithm. Furthermore, the p-values ​​for all indicators are far less than 0.05, indicating that the performance differences among the four algorithms are statistically significant. Specifically, the average rank of the hybrid strategy improved snake optimization algorithm is only 1 / 4.5 that of the whale optimization algorithm. The smallest average rank in the result stability indicator reflects its smaller fluctuations in results, and the leading average rank in the optimal solution search capability indicator proves its higher global optimization accuracy. This fully verifies that the algorithm combines high efficiency and stability in 2D parameter search, perfectly meeting the engineering requirements for real-time optimization of coal seam fracturing parameters.

[0155] Table 1

[0156]

[0157] The Wilcoxon signed-rank test verifies the significance of pairwise differences between algorithms by calculating the rank sum and the number of difference functions. The rank sum includes R+ (the rank sum of functions indicating that the hybrid strategy improved snake optimization algorithm outperforms the comparison algorithms) and R- (the rank sum of functions indicating that the hybrid strategy improved snake optimization algorithm is inferior to the comparison algorithms). The number of difference functions includes + (the hybrid strategy improved snake optimization algorithm is superior), - (the hybrid strategy improved snake optimization algorithm is inferior), and = (no significant difference). The test results show that all p-values ​​are less than 0.05, indicating that the performance difference between the hybrid strategy improved snake optimization algorithm of this invention and the three baseline algorithms is statistically significant.

[0158] The experimental results are shown in Table 2 below. Compared with the original snake optimization algorithm, the hybrid strategy improved snake optimization algorithm outperforms in 10 out of 12 test functions, effectively overcoming the technical shortcomings of the original snake optimization algorithm, such as slow convergence speed and susceptibility to local optima. For the whale optimization algorithm and the chameleon swarm algorithm, the hybrid strategy improved snake optimization algorithm shows an absolute advantage in 11-12 test functions, with no performance degradation in any function, and the R+ value is significantly higher than the R- value, proving the significant synergistic effect of its four hybrid improvement strategies: good point set initialization, nonlinear balance factor exploration-expansion mechanism, differential evolution strategy, and Lévy flight strategy. These results indicate that the performance advantage of the hybrid strategy improved snake optimization algorithm in 2D fracture parameter optimization is not a random fluctuation, but an inevitable result of the algorithm structure improvement, providing reliable statistical support for its online real-time output of stable near-optimal fracture parameters.

[0159] Table 2

[0160]

[0161] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0162] The above embodiments are merely illustrative examples and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A supercritical carbon dioxide pulsed fluidized bed cracking and permeation enhancement device, characterized in that, It includes a liquid CO2 storage and supply unit, a supercritical state preparation and pressurization unit, an intelligent pulsating flow generation unit, a data monitoring and artificial intelligence control unit, and a downhole sealing and fracturing unit, all connected in a sealed manner, and these units together form a closed-loop operation chain; among them, The data monitoring and artificial intelligence control unit includes a sensor monitoring module and an artificial intelligence optimization control module. The artificial intelligence optimization control module has a built-in hybrid strategy improved snake optimization algorithm, which is used to maximize the fracture range and maximize the unit energy consumption permeability as dual optimization objectives. Based on the real-time data collected by the sensor monitoring module, it dynamically outputs a stable near-optimal fracture parameter combination that makes the fitness function converge, and controls the intelligent pulsating flow generation unit to adjust the pulsation frequency and pressure amplitude. The intelligent pulsating flow generating unit is used to modulate the steady-state supercritical carbon dioxide output by the supercritical state preparation and pressurization unit into a pulsating fluid with controllable pulsating frequency and pressure amplitude, and to deliver the pulsating fluid to the downhole sealing and fracturing unit. The downhole sealing and fracturing unit is used to inject the pulsating fluid into the target fracturing section of the coal seam and to monitor micro-fracture events in the coal body in real time.

2. The apparatus according to claim 1, characterized in that, The hybrid strategy improved snake optimization algorithm adopts at least one of the following improved strategies: good point set initialization strategy, nonlinear balance factor exploration-exploitation mechanism, differential evolution strategy, and Lévy flight strategy.

3. The apparatus according to claim 1, characterized in that, The artificial intelligence optimization control module is configured as follows: Based on the pressure fluctuation curve collected in real time by the sensing and monitoring module and the spatiotemporal distribution data of micro-fracture events located in real time by the downhole sealing and fracturing unit, the current expansion status of the fracture network is dynamically evaluated. When it is determined that the coal seam is in the period of rapid fracture expansion, the intelligent pulsating flow generating unit is controlled to increase the pulsating frequency to match the resonance conditions of the newly formed fractures. When it is determined that crack propagation is hindered, the intelligent pulsating flow generating unit is controlled to adjust the pressure amplitude to maintain the stress intensity factor at the crack tip.

4. The apparatus according to claim 1, characterized in that, The intelligent pulsating flow generating unit includes a high-speed rotary valve type pulsating generator, a bladder accumulator, and a high-frequency response regulating valve; wherein... The high-speed rotary valve type pulsation generator is used to modulate steady-state supercritical carbon dioxide into pulsating fluid. The bladder-type accumulator is connected in parallel to the outlet pipe of the high-speed rotary valve type pulsation generator to absorb the energy of the pulsating pressure peak and release it in the trough, so as to achieve smooth correction of the pulsating waveform. The high-frequency response regulating valve is used to fine-tune the pressure amplitude and waveform of the pulsating fluid.

5. The apparatus according to claim 1, characterized in that, The downhole sealing and fracturing unit includes a high-strength, high-pressure-resistant fracturing tube and a capsule-type expanded cement dual-stage sealing device; wherein... The high-strength, high-pressure resistant fracturing tube is equipped with a directional jet nozzle at its front end, which is used to precisely guide the pulsating fluid to the target fracturing section of the coal seam. The capsule-shaped expanded cement dual-stage sealing device is used to provide a dual-stage seal for the borehole, preventing leakage of high-pressure fluid.

6. The apparatus according to claim 3, characterized in that, The sensing and monitoring module includes a high-frequency pressure sensor, a high-frequency flow sensor, a temperature sensor, and a downhole micro-vibration monitoring system. Data from each sensor is uploaded to the artificial intelligence optimization and control module in real time via a wireless transmission module.

7. A method for supercritical carbon dioxide pulsed fluidization-induced fracturing and permeability enhancement, implemented using the apparatus according to any one of claims 1-6, characterized in that, include: In a low-permeability coal seam, a fracturing borehole is drilled, and the downhole sealing fracturing unit is placed into the borehole for pressurized sealing. Steady-state supercritical carbon dioxide fluid is prepared using the liquid CO2 storage and supply unit and the supercritical state preparation and pressurization unit. The coal seam physical property data is input into the artificial intelligence optimization control module to perform offline pre-initialization of the hybrid strategy improved snake optimization algorithm; The steady-state supercritical carbon dioxide fluid is modulated into a pulsating fluid and injected into the coal seam through the intelligent pulsating flow generation unit, while the fracturing process data is collected in real time through the sensing and monitoring module. The real-time collected fracturing process data is input into the hybrid strategy improved snake optimization algorithm for online optimization, and the stable near-optimal pulsation frequency and pressure amplitude that make the fitness function converge are dynamically output. The intelligent pulsating flow generating unit adjusts the pulsating frequency and pressure amplitude in real time based on the output stable near-optimal pulsating frequency and pressure amplitude to induce pulsating fracturing in the coal seam.

8. The method according to claim 7, characterized in that, After completing the pulsating fracturing, the method also includes a well-sealing process and a negative pressure extraction process, wherein... The well-steaming process uses supercritical carbon dioxide to replace and displace coalbed methane. During the negative pressure extraction process, the extracted gas and carbon dioxide mixture is separated by a ground separation device, and the separated carbon dioxide is returned to the liquid CO2 storage and supply unit for recycling.

9. The method according to claim 7, characterized in that, The process involves inputting real-time collected fracturing process data into the hybrid strategy improved snake optimization algorithm for online optimization, dynamically outputting a stable near-optimal pulsation frequency and pressure amplitude that converges the fitness function, including: Using real-time collected data on the fracturing process as input, and with the dual objectives of maximizing the fracturing range and maximizing the permeability increase per unit energy consumption, an initial snake population is generated through a good point set initialization strategy. The global exploration and local development phases of the snake optimization algorithm are improved by using a hybrid strategy of dynamic switching of nonlinear balance factors. The game incorporates Lévy flight strategy in combat mode and differential evolution strategy in mating mode to iteratively update the snake population position. When the maximum number of iterations or the fitness value converges, a stable and near-optimal pulsation frequency and pressure amplitude are output and converted into control commands to be sent to the intelligent pulsation flow generation unit.

10. The method according to claim 9, characterized in that, The method further includes: During the pulsating fracturing process, the artificial intelligence optimization control module monitors the fracturing pressure, micro-fracture events, and permeability enhancement rate per unit energy consumption in real time. When the amplitude of the fracturing pressure fluctuation is less than the first preset threshold, the spatial density of micro-fracture events is lower than the second preset threshold, and the continuous iterative increment of the unit energy consumption permeability is less than the third preset threshold, the artificial intelligence optimization control module determines that fracturing is sufficient and automatically terminates the operation at this stage.

Images