Gas extraction pipe network coupling characteristic global inversion real-time resolving method

Abstract

The invention discloses a gas extraction pipe network coupling characteristic global inversion real-time resolving method, which relates to the technical field of mine gas extraction safety monitoring and comprises the following steps of: constructing a closed-loop resolving system of a coupling characteristic penetrating through a whole process by taking coupling association of gas concentration and gas physical parameters as underlying logic; after coupling feature pre-extraction, coupling type global topology modeling, steady-state coupling control equation set construction and sparse measurement point data preprocessing, a multi-target weighted inversion target function is constructed, pipe network global parameter inversion is achieved through a coupling self-adaptive numerical iteration algorithm with double-threshold convergence judgment, then coupling type calculation reliability quantitative verification is conducted, and a multi-target weighted inversion target function is obtained. And performing working condition self-adaptive coupling correction on the low-precision section to form a closed-loop logic of coupling inversion-verification-correction. According to the method, high-precision global inversion under sparse monitoring can be realized, the investment of monitoring equipment is reduced, the sensing precision of the operation state of the pipe network and the safety management and control level are improved, and the method is suitable for state monitoring and regulation and control of various mine gas extraction pipe networks.

Description

Technical Field

[0001] This invention relates to the field of mine gas drainage safety monitoring technology, and in particular to a real-time solution method for full-domain inversion of coupling characteristics of gas drainage pipeline network. Background Technology

[0002] As the core carrier of gas transportation, the accurate perception of parameters such as pressure, flow rate, and gas concentration throughout the gas drainage pipeline network directly determines the optimization level of gas drainage efficiency and the mine's ability to prevent and control gas disasters. It is an important foundation for the intelligent and safe operation of the mine gas drainage system. Due to the limitations of complex underground geological and operational conditions, as well as equipment purchase and maintenance costs, it is impossible to deploy monitoring points throughout the entire gas drainage pipeline network. Monitoring points can only be deployed at a few key locations such as pump stations and main pipeline intersections. Therefore, achieving high-precision calculation of all parameters of the pipeline network based on sparse monitoring data has become a research focus of the perception of the overall operational status of the gas drainage pipeline network.

[0003] Currently, some technologies in the industry have attempted to calculate or invert parameters of gas drainage pipeline networks using monitoring data from limited measurement points, trying to overcome the technical and cost bottlenecks of full-area measurement point deployment. These technologies have provided some technical insights for the condition monitoring of mine gas drainage pipeline networks. However, existing gas drainage pipeline network parameter calculation technologies still have many shortcomings, falling far short of the accuracy requirements of actual underground engineering applications and failing to meet the actual needs of gas drainage pipeline network safety management. Firstly, existing technologies do not consider the coupled flow characteristics of the gas-air mixture within the gas drainage pipeline network, neglecting the coupling correlation between gas concentration and physical parameters such as mixed gas density, dynamic viscosity, friction coefficient, and Reynolds number. This leads to a disconnect between the calculation model and the actual underground flow field, resulting in significant deviations between the calculation results and actual operating conditions. Secondly, existing technologies mostly use general linear calculation frameworks, failing to construct a suitable coupled mathematical constraint system for the flow laws of the gas-air mixture, and the calculation process lacks... The strong constraints of the coupling characteristics result in low calculation accuracy, making it impossible to achieve coupled calculation of multiple parameters such as pressure, flow rate, concentration, and density. Third, existing technologies lack closed-loop calculation correction logic; the calculation process is merely a single parameter inversion output, and no quantitative reliability verification system for the calculation results has been established. This makes it impossible to quantitatively determine the calculation accuracy, and the calculation algorithm is prone to spurious convergence, resulting in a lack of effective guarantee for the reliability of the calculation results. Fourth, the existing technology's measurement point layout lacks the constraints of the gas-air coupling characteristics, and the selection of measurement points lacks specificity, failing to focus on key sections of the coupled flow regime. This easily introduces invalid monitoring data, further reducing the effectiveness of the calculation results and wasting monitoring resources.

[0004] Therefore, there is an urgent need to develop a real-time solution method that uses gas-air coupling characteristics as the core driver to achieve high-precision inversion of parameters across the entire gas extraction pipeline network under sparse monitoring, in order to make up for the shortcomings of existing technologies and improve the accuracy and reliability of pipeline network parameter solution. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a real-time solution method for the full-domain inversion of coupling characteristics of gas drainage pipeline networks. Based on the coupling relationship between gas concentration and gas physical parameters as the underlying logic, it constructs a gas-air coupling-driven full-domain inversion closed-loop solution system with coupling characteristics spanning the entire process. This achieves high-precision real-time solution of full-domain parameters of gas drainage pipeline networks under sparse monitoring conditions, reduces investment in monitoring equipment, improves the accuracy of gas drainage pipeline network operation status perception and safety management level, and meets the engineering needs of gas drainage pipeline network status monitoring and intelligent control in various mines.

[0006] The technical solution adopted by this invention to solve its technical problem is: a real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network, comprising the following steps: S1: Pre-extraction of gas-air coupling characteristics: Based on the flow characteristics of mixed gas in the pipeline network, extract the coupling correlation characteristics between gas concentration and mixed gas density, dynamic viscosity, friction coefficient, and Reynolds number, and determine the quantitative correlation formulas of each coupling parameter; S2: Coupled global topology abstraction modeling: Based on graph theory and pipeline topology, coupling features are incorporated to abstract pipelines into coupled directed flow path units carrying coupling parameters. Nodes are divided into basic coupled nodes and feature-enhanced coupled nodes to complete the parameterized representation of coupling and construct a coupled global topology abstraction model. S3: Construction of the gas-air steady-state coupled control equation set: Based on the coupling features extracted in step S1, a gas-air steady-state coupled control equation set is established that satisfies the three assumptions of one-dimensional compressibility, isothermal steady state, and continuous and consistent nodal pressure. Gas concentration is used as the core coupling variable to realize multi-parameter coupling constraints of pressure-flow-concentration-density. S4: Measurement point data acquisition and preprocessing: Strongly constrained measurement points with coupling characteristics are set up in key sections of the pipeline network to collect core monitoring data. The coupled quantization formula in step S1 is used to preprocess the data to obtain the coupled characteristic monitoring dataset, which serves as the coupling input constraint for the global inversion. S5: Constructing the objective function and solving iteratively: The gas-air steady-state coupled control equations are transformed into standard nonlinear coupled function equations. A coupled multi-objective weighted global inversion objective function is constructed and solved using a coupled adaptive numerical iterative algorithm with dual threshold convergence criteria to achieve global parameter inversion and coupled state reconstruction of the pipeline network. S6: Quantitative verification of reliability of coupled solution: Based on the coupling characteristics, a quantitative index system is constructed that includes the solution accuracy of the extraction pipeline network, the deviation rate of coupling parameters, and the coupling error of pump station operation, and multi-dimensional coupling verification of the full-domain inversion results is carried out. S7: Adaptive Coupling Correction under Working Conditions: If the coupling verification result in step S6 meets the standard, the solution result is output; if it does not meet the standard, the low-precision section is located, the coupling parameters are dynamically corrected under working conditions, and the process returns to step S5 to perform coupling iteration again, forming a closed-loop solution system of coupling inversion-verification-correction.

[0007] Furthermore, the specific process of pre-extracting the gas-air coupling characteristics in step S1 is as follows: by combining downhole field tests and laboratory simulations, the variation laws of the density, dynamic viscosity, friction coefficient, and Reynolds number of the mixed gas under different gas concentrations are determined. A nonlinear fitting method is used to establish a quantitative correlation formula between each parameter and the gas concentration. Among them, the density of the mixed gas is linearly negatively correlated with the gas concentration, and the dynamic viscosity, friction coefficient, and Reynolds number are nonlinearly exponentially correlated with the gas concentration.

[0008] Furthermore, the coupling parameters carried by the coupled directional flow path unit in step S2 include density, dynamic viscosity, friction coefficient, local resistance equivalent length, and Reynolds number associated with gas concentration; the basic coupling node is the intersection of the gas source and sink positions of the coal seam drainage borehole, gas pump station inlet, and vent with the coupled directional flow path unit, integrating gas convergence / divergence coupling flow characteristics; the feature-enhanced coupling node is the location of the auxiliary components, integrating local resistance coupling loss characteristics; the basic coupling node and the feature-enhanced coupling node together constitute the global coupling solution node system of the gas drainage pipeline network.

[0009] Furthermore, the three assumptions mentioned in step S3 are as follows: One-dimensional compressible steady-state assumption: Ignore the pipe path velocity gradient and non-uniform parameter distribution, retain only the axial flow variable, and realize the coupled change of gas density with pressure and gas concentration through a coupled quantization formula; Isothermal steady-state assumption: Based on the environmental characteristics of the underground roadway being closed and constant in temperature with sufficient heat exchange, significant changes in gas temperature are ignored, the energy equation constraints are simplified, and only the basic influence of temperature on the coupling parameters is retained; Assumption of continuous and consistent node pressure: At any coupled solution node in the pipeline network, the terminal pressure of each branch flowing into the node remains equal to the initial pressure of each branch flowing out of the node, ensuring the physical consistency of the coupled flow field.

[0010] Furthermore, the gas-air steady-state coupled control equation set is a four-parameter coupled equation set of pressure-flow-velocity-concentration, specifically including: the coupled equation of the one-dimensional compressible motion state of the mixed gas in the pipeline, the coupled equation of mass conservation and concentration transport of gas components, the coupled equation of the relationship between the total flow rate and velocity of the mixed gas in the pipeline cross section, the coupled equation of the flow convergence balance at the nodes, and the coupled equation of friction resistance and local resistance. The gas-air steady-state coupled control equations are designed for the series convergence conditions from node n to node n-1 and from node n+1 to node n-1 in the pipeline network. They define pressure drop, flow rate, and concentration correlation formulas that include gas concentration coupling terms. The integrated parameters include friction coefficient, local resistance equivalent length, and absolute pipe wall roughness, thereby achieving accurate characterization of the gas-air mixed gas flow process. Specifically defined as: In the formula, , and These are the mixed gas volume flow rates at the nth node, the (n+1)th node, and the (n-1)th node, respectively; , and These are the gas mixture densities at the nth node, the (n+1)th node, and the (n-1)th node, respectively. , and These represent the mixed gas concentrations at the nth node, the (n+1)th node, and the (n-1)th node, respectively. Let be the gas flow rate in the pipe at node i; Let be the flow velocity within the pipe at node i; Let be the inner diameter of the extraction pipe where the i-th node is located; For the first Gas density in the pipe at point; Let be the gas pressure inside the pipe at node i; Let be the molar mass of the mixed gas in the pipe at node i; Let be the gas state constant; The temperature of the gas inside the pipe; The mechanical energy loss per unit weight of the mixed gas flow within the i-th pipe segment; Let be the friction coefficient of the i-th pipeline segment; Let be the Reynolds number of the mixed gas flow in the i-th pipeline segment; Let be the dynamic viscosity of the mixed gas in the i-th pipeline segment; The actual length of the i-th pipe segment; The equivalent length of the local resistance within the i-th pipeline segment; denoted as the absolute roughness of the i-th pipe segment; g is the acceleration due to gravity. , and These are the absolute pressures of the mixed gas at nodes n-1, n, and n+1, respectively. , and These are the elevations of the (n-1)th node, the nth node, and the (n+1)th node, respectively. , and These represent the mixed gas flow velocities in the (n-1)th, nth, and (n+1)th pipe segments, respectively. and These are the unit weight mechanical energy losses of the mixed gas flow in the nth and (n+1)th pipe segments, respectively. The system's pumping station extracts negative pressure; This refers to the absolute atmospheric pressure during pump station operation. The absolute pressure of the mixed gas in the pipeline at the pump station interface; Z1 represents the absolute pressure of the mixed gas in the adjacent pipeline section downstream of the pump station; Z2 and Z1 represent the elevations of the pump station interface node and the adjacent downstream node, respectively; V1 and V2 represent the flow velocities of the mixed gas in the pump station interface pipeline section and the adjacent downstream pipeline section, respectively. This refers to the mechanical energy loss per unit weight of the mixed gas flow within the pipeline section from the pump station interface to the downstream adjacent node.

[0011] Furthermore, the principle for deploying the strong constraint measurement points of coupling characteristics in step S4 is to fully cover the key sections of the coupled flow state, with the number of measurement points being 5%-15% of the total number of nodes in the pipeline network. Each measurement point synchronously collects core parameters reflecting the gas-air coupling characteristics. Through the coupling characteristic data of the measurement points, a strong constraint on the coupled flow state of the entire pipeline network is achieved, preventing interference from invalid data from non-coupled measurement points.

[0012] Furthermore, the unknown vector of the coupled multi-objective weighted global inversion objective function described in step S5 includes the pressure and gas velocity of all nodes in the pipeline network to be solved. The objective function is constructed based on the optimal synergy between physical coupling constraints and numerical solution accuracy. The components and their weights are as follows: The core coupling error term is the mean square error between the coupled characteristic monitoring data and the model's coupled calculation value, accounting for 60%-70% of the weight; the pipeline pressure drop coupling constraint term is the deviation threshold between the actual pipeline pressure drop and the theoretical pressure drop of the coupled model, accounting for 20%-25% of the weight; the flow conservation coupling penalty term is the coupling imbalance of the node's confluence flow, accounting for 5%-10% of the weight; each weight coefficient is adaptively adjusted according to the downhole extraction conditions; The specific definition of the coupled multi-objective weighted global inversion objective function is as follows: , where: unknown vector It contains the pressure and gas velocity of all the nodes to be solved, i.e. , This represents the total number of pipeline nodes. It consists of n-1 pressure drop equations and n-1 nodal flow equations. Each equation corresponds to a residual function, which represents the error change between the measured values ​​at finite measuring points and the theoretical calculation values ​​of the model. If the residual function is 0, the equation satisfies the set physical conservation conditions.

[0013] Furthermore, the coupled adaptive numerical iterative algorithm described in step S5 is either the Newton iteration method based on topological coupling constraints or the adaptive damped least squares method. The initial values ​​of the algorithm are engineering-based initial values ​​that conform to the physical laws of downhole negative pressure extraction: the initial pressure value is estimated based on the rated negative pressure of the extraction pump, along the pipeline towards the borehole direction according to the linear decay law of friction resistance; the initial flow velocity value is selected from the economic flow velocity range of 4-10 m / s, and is calculated in combination with the design flow rate of the extraction system and the cross-sectional area of ​​the pipeline. The coupled adaptive numerical iterative algorithm is specifically defined as follows: In the formula, This is the initial iteration vector; The initial iterative pressure value for the i-th node; For the first The length of the pipeline section; This is the total length of the pipeline; Let be the initial iterative value of the flow velocity in the i-th pipe segment; Design the flow rate for the extraction system; Let be the cross-sectional area of ​​the i-th pipe segment; This represents the current iteration number; Let k be the pressure balance residual function; The pressure at node i-1 after the k-th iteration; The pressure at node i after the k-th iteration; Let be the friction coefficient of the i-th pipeline segment; Let be the inner diameter of the i-th pipe segment; Let be the density of the mixed gas in the i-th pipeline segment after k iterations; denoted as , where is the flow velocity in the i-th pipe segment after the k-th iteration; g is the acceleration due to gravity. Let be the elevation difference between the two nodes of the i-th pipeline segment; Let this be the (k+n)th flow balance residual function; It is the sum of the volumetric flow rates of all inflow branches to the i-th node after the k-th iteration; It is the sum of the volumetric flow rates of all outflow branches of the i-th node after the k-th iteration; This is the global residual vector; It is a topologically constrained sparse Jacobian matrix; This is the correction vector for the k-th iteration; It is a relaxation factor; Let be the iteration vector for the (k+1)th iteration.

[0014] Furthermore, the core feature of the coupled adaptive numerical iterative algorithm is the synchronous iteration of coupling parameters. During the iteration process, the node pressure, branch flow, gas concentration, and various coupled parameters are simultaneously corrected. The iteration convergence determination adopts a dual threshold determination of residual accuracy coupling threshold and parameter accuracy coupling threshold, with the residual accuracy coupling threshold set to 10. -5 ~10 - 6 Pa, parameter precision coupling threshold is set to 10. -4 ~10 -5 Pa, and set the maximum number of iterations to 50; The specific definition of iterative convergence criterion is: in, The Euclidean norm of the updated residual vector; This is the m-th component of the global residual vector; The residual accuracy threshold is set to 10. -5 ~10 -6 If Pa satisfies the above conditions, then the residual is considered to be small enough, and the system of equations satisfies the equilibrium conditions of the laws of conservation of energy and conservation of mass. This represents the relative change of the updated unknown vector; Let be the relative change of the t-th unknown vector; The parameter precision threshold is set to 10. -4 ~10 -5 Pa.

[0015] Furthermore, the coupled adaptive numerical iterative algorithm introduces a topologically constrained sparse Jacobian matrix, where the matrix elements are the first-order partial derivatives of the residual function with respect to each unknown. The partial derivatives are non-zero only when there is a physical coupling relationship between the unknown and the equation, and all other partial derivatives are zero. The efficiency of coupled iteration is improved through sparsification.

[0016] Furthermore, the termination verification process of the coupled adaptive numerical iterative algorithm consists of three levels of determination: First, if both the residual accuracy coupling threshold and the parameter accuracy coupling threshold are satisfied simultaneously, the iteration is terminated, and the final solution for the pressure and gas flow rate at each node is output. Secondly, if the double threshold is not met, let the iteration number k = k + 1, and return to recalculate the residual vector and the topologically constrained sparse Jacobian matrix; Third, if the number of iterations exceeds the maximum number of iterations (50) and still fails to converge, a backtracking check is triggered to verify in sequence whether the initial value conforms to the physical law of underground negative pressure extraction, whether the sign of the residual function is correct, and whether the calculation of the partial derivative of the Jacobian matrix is ​​correct.

[0017] Furthermore, the gas-air coupled solution reliability quantification index system described in step S6 includes three quantitative indicators, namely the solution accuracy of the extraction pipeline network, the deviation rate of coupling parameters, and the coupling error of pump station operation; Among them, the accuracy rate of the extraction pipeline network calculation is the proportion of the number of coupled nodes that meet the engineering qualification standards to the total number of coupled monitoring nodes. The engineering qualification standards are that the relative error between the calculated negative pressure of the node and the actual value measured by the sensor is ≤5%, and the relative error between the calculated value of gas concentration and the actual value is ≤3%. The coupling parameter deviation rate is the deviation rate between the calculated coupling parameters and the values ​​calculated by the quantization formula of the pre-extracted coupling features of S1. A deviation rate of ≤2% is considered acceptable. The pump station operation coupling error is the deviation between the average negative pressure under stable operating conditions and the operating negative pressure under coupled solution conditions. A deviation ≤ 1 kPa is considered acceptable. The coupling verification result is deemed to conform to engineering application standards only when all three quantitative indicators meet the qualification criteria; the specific definitions of the reliability quantitative indicators are as follows: In the formula, This indicates the accuracy rate of the extraction pipeline network calculation; The total number of all nodes in the pipeline network; This represents the number of nodes with a relative error of less than 5%. Let be the solution error for the i-th node; The measured negative pressure at node i; The theoretical negative pressure for solving the i-th node; This refers to the relative error in pump station operation. The average negative pressure of the pumping station under stable operating conditions for a period of time; To calculate the operating negative pressure of the pumping station under the given conditions; To account for the relative error, the maximum value of the single-node solution error and the pump station operation error is taken as the upper limit of the overall solution error.

[0018] Furthermore, the specific process of the adaptive coupling correction described in step S7 is as follows: Based on the non-conforming items of the reliability quantification index of the coupled solution, the core problem of the low-precision solution section is located. The coupling coefficient, resistance coefficient, and viscosity coefficient in the gas-air steady-state coupled control equation set are adaptively adjusted using a coupling parameter dynamic optimization algorithm. The adjustment range is ±5% to ±10% of the original parameters. After adjustment, the process returns to S5 to perform coupling iteration solution again until the coupling verification result meets the engineering standard. When the accuracy of the extraction pipeline solution is ≥90%, the solution result is directly output for pipeline resistance analysis, extraction efficiency optimization, and gas disaster early warning.

[0019] The beneficial effects of this invention are: 1. Coupled features drive the entire process, ensuring that the solution model fits the actual flow field downhole from the physical mechanism level: Based on the coupling relationship between gas concentration and physical parameters such as mixed gas density, dynamic viscosity, and friction coefficient, the coupling features are integrated into the entire process of modeling, solution, verification, and correction. This breaks through the limitation of traditional technology that ignores the gas-air coupled flow characteristics, enabling the solution model to accurately match the real flow field downhole. This fundamentally avoids the problem of the model being out of touch with the actual working conditions and greatly improves the physical consistency of the solution results.

[0020] 2. High-precision global inversion is achieved under sparse monitoring mode, significantly reducing the investment in monitoring equipment: Only 5%-15% of the total number of nodes in the pipeline network with strong constraints on coupling characteristics need to be deployed to complete the high-precision inversion of the parameters of the entire pipeline network. While ensuring the accuracy of the solution, the purchase, installation and operation and maintenance costs of monitoring equipment are greatly reduced. At the same time, the effectiveness of input constraints is ensured by coupling characteristic data preprocessing, which solves the industry pain point that the deployment of global monitoring points is limited by the working conditions and cost in the well.

[0021] 3. By using closed-loop solution and dual-threshold convergence judgment, the risk of false convergence is avoided, ensuring stable and controllable solution: A closed-loop solution logic of coupled inversion, quantization verification, and operating condition correction is constructed. Combined with dual-threshold convergence judgment of residual accuracy and parameter accuracy, the false convergence problem of small residuals but large fluctuations in coupled parameters in traditional algorithms is effectively avoided. This achieves stable and controllable solution process and dynamic optimization of results, improving the robustness of global inversion.

[0022] 4. A quantitative reliability verification system provides quantifiable accuracy assurance for the solution results: A multi-dimensional quantitative index system is established, including the accuracy rate of the extraction pipeline network solution, the deviation rate of coupling parameters, and the coupling error of pump station operation. The engineering qualification standards are clarified, and the quantitative judgment of the solution accuracy is realized. This solves the defect of traditional technology without quantitative accuracy assurance and provides a reliable and traceable accuracy basis for the engineering application of the solution results.

[0023] 5. Adaptive correction and engineering adaptation to meet the complex and ever-changing operational needs underground: For scenarios where verification fails, the core parameters are adaptively adjusted through a coupled parameter dynamic optimization algorithm to form a closed-loop optimization; at the same time, the iterative initial values ​​are aligned with the physical laws of underground negative pressure extraction, and a fast output rule with a solution accuracy of ≥90% is set, which can be directly adapted to engineering scenarios such as pipeline resistance analysis, extraction efficiency optimization, and gas disaster early warning. It is applicable to various mine gas extraction pipeline networks and has extremely strong engineering adaptability. Detailed Implementation

[0024] This invention discloses a real-time solution method for global inversion of coupling characteristics of gas extraction pipeline networks.

[0025] A real-time solution method for the full-domain inversion of coupled characteristics of gas drainage pipeline networks is proposed. Based on the coupling relationship between gas concentration and gas physical parameters, it constructs a gas-air coupled-driven full-domain inversion closed-loop solution system with coupled characteristics throughout the entire process. Through the progressive implementation of the following steps, it achieves high-precision full-domain inversion of gas-air mixed gas pipeline networks under sparse monitoring conditions, effectively reducing investment in monitoring equipment, improving the accuracy of gas drainage pipeline network operation status perception and safety management level, and is widely applicable to status monitoring and intelligent control of gas drainage pipeline networks in various mines. Specifically, it includes the following steps: S1: Pre-extraction of gas-air coupling characteristics: Based on the flow characteristics of gas-air mixture in the mine gas drainage pipeline network, the coupling correlation characteristics between gas concentration and mixed gas density, dynamic viscosity, friction coefficient and Reynolds number are extracted, and the quantitative correlation formula of each coupling parameter is determined.

[0026] The specific implementation method is as follows: by combining in-situ tests in the well with laboratory numerical simulation, the dynamic variation law of the density, dynamic viscosity, friction coefficient and Reynolds number of the mixed gas under different gas concentrations is determined. A nonlinear fitting method is used to establish a quantitative correlation formula between each physical parameter and the gas concentration, and the coupling correlation characteristics between the gas concentration and each physical parameter are accurately extracted.

[0027] From the perspective of coupling principle, gas concentration, as the core coupling variable, has a clear quantitative correlation with various physical parameters. Among them, the density of the mixed gas is linearly negatively correlated with the gas concentration, while the dynamic viscosity, friction coefficient, and Reynolds number are nonlinearly exponentially correlated with the gas concentration.

[0028] This step, by accurately extracting the gas-air coupling characteristics, overcomes the technical deficiency of traditional solution methods that ignore the coupled flow characteristics of mixed gases. This ensures that the basic parameters for subsequent modeling and solution closely match the actual flow field characteristics downhole, guaranteeing the consistency and accuracy of the whole-domain inversion from the perspective of physical mechanisms.

[0029] S2: Coupled Global Topology Abstract Modeling: Based on graph theory and pipeline topology, coupling features are integrated into pipeline topology modeling. Pipelines are abstracted into coupled directed flow path units carrying gas-air coupling parameters. Nodes are divided into basic coupled nodes with integrated coupled flow characteristics and feature-enhanced coupled nodes. Coupled parameterized representation of pipeline nodes, main branches and auxiliary components is completed, and a coupled global topology abstract model is constructed.

[0030] Among them, the coupling parameters carried by the coupled directional flow path unit include density, dynamic viscosity, friction coefficient, local resistance equivalent length, and Reynolds number associated with gas concentration; the basic coupling node is the intersection of the gas source and sink positions of the coal seam extraction borehole, gas pump station inlet, and vent with the coupled directional flow path unit, integrating the gas convergence / diversion coupling flow characteristics.

[0031] The feature-enhanced coupling nodes are the locations of auxiliary components such as valves, reducers, tees, and bends, integrating local resistance coupling loss characteristics; the basic coupling nodes and the feature-enhanced coupling nodes together constitute the whole-domain coupling solution node system of the gas extraction pipeline network, completing the coupling parameterization characterization of pipeline network nodes, main branches and auxiliary components.

[0032] This step achieves a deep integration of pipeline network topology and coupling characteristics, providing standardized topological constraints and parameter carriers for the subsequent construction of steady-state coupled control equations.

[0033] S3: Construction of the gas-air steady-state coupled control equation set: Based on the coupling features extracted in step S1, a gas-air steady-state coupled control equation set is established that satisfies the three assumptions of one-dimensional compressibility, isothermal steady state, and continuous and consistent nodal pressure. The gas-air steady-state coupled control equation set takes gas concentration as the core coupling variable to realize multi-parameter coupling constraints of pressure-flow-concentration-density, forming a coupled mathematical constraint system applicable to the inversion of the entire pipeline network.

[0034] The three assumptions are as follows: One-dimensional compressible steady-state assumption: Ignore the pipe path velocity gradient and non-uniform parameter distribution, retain only the axial flow variable, and realize the coupled change of gas density with pressure and gas concentration through a coupled quantization formula; Isothermal steady-state assumption: Based on the environmental characteristics of the underground roadway being closed and constant in temperature with sufficient heat exchange, significant changes in gas temperature are ignored, the energy equation constraints are simplified, and only the basic influence of temperature on the coupling parameters is retained; Assumption of continuous and consistent node pressure: At any coupled solution node in the pipeline network, the terminal pressure of each branch flowing into the node remains equal to the initial pressure of each branch flowing out of the node, ensuring the physical consistency of the coupled flow field.

[0035] Based on this, a set of gas-air steady-state coupled control equations is constructed. The set of gas-air steady-state coupled control equations is a four-parameter coupled equation set of pressure-flow rate-velocity-concentration, specifically including: the coupled equation of the one-dimensional compressible motion state of the mixed gas in the pipeline, the coupled equation of mass conservation and concentration transport of gas components, the coupled equation of the relationship between total flow rate and velocity of the mixed gas in the pipeline cross section, the coupled equation of flow convergence balance at the nodes, and the coupled equation of friction resistance and local resistance. The gas-air steady-state coupled control equations are designed for the series convergence conditions from node n to node n-1 and from node n+1 to node n-1 in the pipeline network. They define pressure drop, flow rate, and concentration correlation formulas that include gas concentration coupling terms. The integrated parameters include friction coefficient, local resistance equivalent length, and absolute pipe wall roughness, so as to achieve accurate characterization of the gas-air mixed gas flow process. Specifically defined as: In the formula, , and These represent the mixed gas volume flow rates at the nth node, the (n+1)th node, and the (n-1)th node, respectively, in cubic meters per second (m³). 3 / s; , and These represent the gas mixture densities at the nth, (n+1)th, and (n-1)th nodes, respectively, in kg / m³. 3 ; , and These represent the mixed gas concentrations at the nth node, the (n+1)th node, and the (n-1)th node, respectively. The gas flow rate at node i is expressed in m³ / s. 3 / s; Let be the flow velocity in the pipe at node i, in m / s; The inner diameter of the extraction pipeline where the i-th node is located is in meters. For the first The gas density in the pipeline at the specified point is expressed in kg / m³. 3 ; The pressure of the gas inside the pipe at node i is expressed in Pa. denoted as the molar mass of the mixed gas in the pipe at node i, in kg / mol. is the gas state constant, with units of J / (mol·K); This refers to the temperature of the gas inside the pipe, expressed in Kelvin (K). The unit weight mechanical energy loss of the mixed gas flow in the i-th pipeline segment is expressed in meters (m), which includes friction loss and friction loss corresponding to the equivalent length of local resistance. Let be the friction resistance coefficient of the i-th pipeline segment, which is a dimensionless parameter; Let be the Reynolds number of the mixed gas flow in the i-th pipeline segment, and be a dimensionless parameter. The dynamic viscosity of the gas mixture in the i-th pipeline segment is expressed in N·s / m³. 2 ; The actual length of the i-th pipeline segment is in meters. The equivalent length of the local resistance within the i-th pipeline segment is expressed in meters. denoted as the absolute roughness of the i-th pipe segment, in meters (m); g represents the acceleration due to gravity. , and These are the absolute pressures of the mixed gas at nodes n-1, n, and n+1, respectively, in Pa. , and These are the elevations of the (n-1)th node, the nth node, and the (n+1)th node, respectively, in meters. , and These represent the flow velocities of the mixed gas in the (n-1)th, nth, and (n+1)th pipe segments, respectively, in m / s. and These are the unit weight mechanical energy losses of the mixed gas flow in the nth and (n+1)th pipe segments, respectively, which are equivalent to the fluid head height in meters and are consistent with the dimensions of the head term in Bernoulli's equation. The negative pressure extracted by the system's pumping stations is expressed in Pa. This refers to the absolute atmospheric pressure during pump station operation, expressed in Pa. This is the absolute pressure of the mixed gas in the pipeline at the pump station interface, in Pa. Z1 represents the absolute pressure of the mixed gas in the adjacent pipeline section downstream of the pump station, in Pa; Z2 and Z1 represent the elevations of the pump station interface node and the adjacent downstream node, respectively, in m; V1 and V2 represent the flow velocities of the mixed gas in the pump station interface pipeline section and the adjacent downstream pipeline section, respectively, in m / s. The unit weight mechanical energy loss of the mixed gas flow in the pipeline section from the pump station interface to the downstream adjacent node is expressed in meters.

[0036] The coupled equations established in this step, with gas concentration as the core coupling variable, realize the coordinated constraint of multiple physical parameters and solve the technical problem that traditional linear solution models are difficult to adapt to the coupled flow of mixed gases.

[0037] S4: Strongly Constrained Coupled Measurement Point Data Acquisition and Preprocessing: This step is based on the design concept of sparse monitoring and strongly constrained input. Targeted measurement points are set up and data preprocessing is completed to provide accurate input constraints for the global inversion and avoid invalid data from interfering with the solution accuracy.

[0038] The specific implementation process is as follows: Strongly constrained measurement points based on coupling characteristics are deployed in key sections of the gas extraction pipeline network, such as pump stations, main pipeline intersections, topological inflection points, and resistance concentration zones. The deployment principle for these measurement points is full coverage of key sections of the coupled flow pattern, with the number of measurement points ranging from 5% to 15% of the total number of nodes in the pipeline network. Each measurement point simultaneously collects core parameters reflecting the gas-air coupling characteristics. The coupling characteristic data from these measurement points achieves strong constraints on the coupled flow state across the entire pipeline network, preventing interference from invalid data from non-coupled measurement points. Then, the coupling quantization formula determined in S1 is used to preprocess the raw monitoring data, resulting in a coupled characteristic monitoring dataset, which serves as the coupling input constraint for the full-domain inversion.

[0039] This step, through the deployment of strongly constrained measurement points with coupling characteristics and data preprocessing, achieves strong constraints on the coupled flow state of the entire pipeline network with a small number of measurement points, significantly reducing the purchase, installation and maintenance costs of monitoring equipment, while ensuring the effectiveness of the solution input from the data source.

[0040] S5: Construction and Iterative Solution of Coupling Multi-Objective Weighted Global Inversion Objective Function: This step transforms the physical coupling constraints into a numerical optimization problem, constructs an objective function that adapts to the coupling characteristics, and uses a high-precision iterative algorithm to complete the global parameter inversion, thereby realizing the reconstruction of the coupling state of all nodes and branches of the pipeline network.

[0041] The specific implementation process is as follows: using the coupled characteristic monitoring dataset as constraints, the gas-air steady-state coupled control equations are transformed into standard nonlinear coupled function equations, and a coupled multi-objective weighted global inversion objective function is constructed, which includes core terms of coupling error, coupling constraints of pipeline pressure drop, and coupling penalties of flow conservation. The objective function is solved by a coupled adaptive numerical iterative algorithm with dual threshold convergence criteria. By iteratively correcting the node pressure, branch flow, and gas concentration, the parameters of all nodes and branches in the entire pipeline network are fully inverted and the coupled state is reconstructed.

[0042] Among them, the unknown vector of the coupled multi-objective weighted global inversion objective function includes the pressure and gas velocity of all nodes in the pipeline network to be solved. The objective function is constructed based on the synergistic optimization of physical coupling constraints and numerical solution accuracy. The components and their weights are as follows: The core coupling error term is the mean square error between the coupled characteristic monitoring data and the model's coupled calculation value, accounting for 60%-70% of the weight; the pipeline pressure drop coupling constraint term is the deviation threshold between the actual pipeline pressure drop and the theoretical pressure drop of the coupled model, accounting for 20%-25% of the weight; the flow conservation coupling penalty term is the coupling imbalance of the node's confluence flow, accounting for 5%-10% of the weight; each weight coefficient is adaptively adjusted according to the downhole extraction conditions; The specific definition of the coupled multi-objective weighted global inversion objective function is as follows: , where: unknown vector It contains the pressure and gas velocity of all the nodes to be solved, i.e. , This represents the total number of pipeline nodes. It consists of n-1 pressure drop equations and n-1 nodal flow equations. Each equation corresponds to a residual function, which represents the error change between the measured values ​​at finite measuring points and the theoretical calculation values ​​of the model. If the residual function is 0, the equation satisfies the set physical conservation conditions.

[0043] The coupled adaptive numerical iterative algorithm is either the Newton iteration method or the adaptive damped least squares method based on topological coupling constraints. The initial values ​​of the algorithm are engineering-based initial values ​​that conform to the physical laws of downhole negative pressure extraction: the pressure initial value is based on the rated negative pressure of the extraction pump and is estimated along the pipeline towards the borehole according to the linear decay law of friction resistance; the flow velocity initial value is selected from the economic flow velocity range of 4-10 m / s and is calculated in combination with the design flow rate of the extraction system and the cross-sectional area of ​​the pipeline. The coupled adaptive numerical iterative algorithm is specifically defined as follows: In the formula, This is the initial iteration vector; The initial iterative pressure value of the i-th node is estimated by taking the rated negative pressure of the extraction pump as the benchmark and estimating the pressure of each node along the pipeline towards the borehole direction according to the linear attenuation law of friction resistance. The unit is Pa. For the first The length of a pipeline section, in meters (m). This refers to the total length of the pipeline, in meters (m). The initial iterative value of the flow velocity in the i-th pipeline segment is expressed in m / s. In engineering practice, it is considered that too low a gas flow velocity can easily lead to gas stagnation, while too high a flow velocity will cause a surge in friction resistance. Considering the economical flow velocity range, the appropriate flow velocity range is selected. m / s; The design flow rate for the extraction system is expressed in cubic meters per second (m³). 3 / s; Let be the cross-sectional area of ​​the i-th pipe segment, in meters. 2 ; This represents the current iteration number; Let k be the pressure balance residual function; This represents the pressure at the (i-1)th node after the k-th iteration, in Pa. The pressure at node i after the k-th iteration is expressed in Pa. Let be the friction coefficient of the i-th pipeline segment; Let be the inner diameter of the i-th pipe segment, in meters. The density of the mixed gas in the i-th pipeline segment after k iterations is given in kg / m³. 3 ; is the flow velocity of the i-th pipe segment after the k-th iteration, in m / s; g is the acceleration due to gravity. The elevation difference between the two nodes of the i-th pipeline segment is expressed in meters. Let this be the (k+n)th flow balance residual function; This is the sum of the volumetric flow rates of all inflow branches to the i-th node after the k-th iteration, in m³. 3 / s; This is the sum of the volumetric flow rates of all outflow branches of node i after the k-th iteration, in m³. 3 / s; This is the global residual vector; It is a topologically constrained sparse Jacobian matrix; This is the correction vector for the k-th iteration; It is a relaxation factor; Let be the iteration vector for the (k+1)th iteration.

[0044] The core feature of the coupled adaptive numerical iterative algorithm is the synchronous iteration of coupling parameters. During the iteration process, node pressure, branch flow, gas concentration, and various coupled parameters are simultaneously corrected. The iteration convergence criterion uses a dual-threshold criterion of residual accuracy coupling threshold and parameter accuracy coupling threshold, with the residual accuracy coupling threshold set to 10. -5 ~10 -6 Pa, parameter precision coupling threshold is set to 10. -4 ~10 -5 Pa, while setting the maximum number of iterations to 50 as a fallback, avoids the pseudo-convergence situation of "small residuals but large fluctuations in coupling parameters"; The specific definition of iterative convergence criterion is: in, The Euclidean norm of the updated residual vector; This is the m-th component of the global residual vector; The residual accuracy threshold is set to 10. -5 ~10 -6If Pa satisfies the above conditions, then the residual is considered to be small enough, and the system of equations satisfies the equilibrium conditions of the laws of conservation of energy and conservation of mass. This represents the relative change of the updated unknown vector; Let be the relative change of the t-th unknown vector; The parameter precision threshold is set to 10. -4 ~10 -5 Pa.

[0045] The coupled adaptive numerical iterative algorithm with dual threshold convergence criteria introduces a topologically constrained sparse Jacobian matrix. The matrix elements are the first-order partial derivatives of the residual function with respect to each unknown. The partial derivatives are non-zero only when there is a physical coupling relationship between the unknown and the equation, and are zero otherwise. The efficiency of coupled iteration is improved through sparsification.

[0046] The termination verification process of the coupled adaptive numerical iterative algorithm with dual threshold convergence determination consists of three levels of determination: First, if both the residual accuracy coupling threshold and the parameter accuracy coupling threshold are satisfied simultaneously, the iteration is terminated, and the final solution for the pressure and gas flow rate at each node is output. Secondly, if the double threshold is not met, let the iteration number k = k + 1, and return to recalculate the residual vector and the topologically constrained sparse Jacobian matrix; Third, if the number of iterations exceeds the maximum number of iterations (50) and still fails to converge, a backtracking check is triggered to verify in sequence whether the initial value conforms to the physical law of underground negative pressure extraction, whether the sign of the residual function is correct, and whether the calculation of the partial derivative of the Jacobian matrix is ​​correct.

[0047] This step achieves accurate inversion of the parameters of the entire pipeline network under sparse monitoring through a multi-objective weighted objective function and an adaptive iterative algorithm. The dual threshold determination effectively avoids the problem of pseudo-convergence in traditional algorithms, ensuring that the solution process is stable and controllable.

[0048] S6: Coupling-type solution reliability quantification verification: Based on the coupling characteristics, a gas-air coupled solution reliability quantification index system is constructed, including three quantitative indicators: the solution accuracy rate of the extraction pipeline network, the deviation rate of coupling parameters, and the coupling error of pump station operation. Multi-dimensional coupling verification is performed on the full-domain inversion results to determine whether the solution results meet the engineering application standards.

[0049] The reliability quantification index system includes three quantitative indicators: the accuracy rate of the extraction pipeline network solution, the deviation rate of coupling parameters, and the coupling error of pump station operation; the specific meanings and standards of each indicator are as follows: The accuracy rate of the extraction pipeline network calculation is the proportion of the number of coupled nodes that meet the engineering qualification standards to the total number of coupled monitoring nodes. The engineering qualification standards are that the relative error between the calculated negative pressure of the node and the actual value measured by the sensor is ≤5%, and the relative error between the calculated value of gas concentration and the actual value is ≤3%. The coupling parameter deviation rate is the deviation rate between the calculated coupling parameters and the values ​​calculated by the quantization formula of the pre-extracted coupling features of S1. A deviation rate of ≤2% is considered acceptable. The pump station operation coupling error is the deviation between the average negative pressure under stable operating conditions and the operating negative pressure under coupled solution conditions. A deviation ≤ 1 kPa is considered acceptable. Only when all three quantitative indicators meet the qualification criteria can the coupling verification result be deemed to conform to the engineering application standard.

[0050] The reliability metric is specifically defined as follows: In the formula, This indicates the accuracy rate of the extraction pipeline network calculation; The total number of all nodes in the pipeline network; This represents the number of nodes with a relative error of less than 5%. Let be the solution error for the i-th node; The measured negative pressure at node i is expressed in Pa. The theoretical negative pressure for solving the i-th node is expressed in Pa. This refers to the relative error in pump station operation. The average negative pressure of the pumping station under stable operating conditions for a period of time is expressed in Pa. The negative pressure of the pump station under the calculated conditions is expressed in Pa. To account for the relative error, the maximum value of the single-node solution error and the pump station operation error is taken as the upper limit of the overall solution error.

[0051] The quantitative verification system established in this step solves the problem of traditional technology lacking quantitative accuracy judgment, ensuring that the calculation results meet the engineering safety requirements for mine gas extraction.

[0052] S7: Adaptive Coupling Correction under Operating Conditions: If the coupling verification result of step S6 meets the engineering standards, output the coupling inversion solution result of the entire pipeline network; if not, locate the low-precision solution section based on the quantification index of the reliability of the coupled solution, perform adaptive dynamic correction of the coupling parameters of the gas-air steady-state coupled control equation set, return to step S5 to perform coupling iteration solution again, and form a closed-loop solution system of coupling inversion-verification-correction.

[0053] The specific implementation process is as follows: Based on the non-compliant items of the reliability quantification index of the coupled solution, the core problem of the low-precision solution section is located. The coupling parameter dynamic optimization algorithm is used to adaptively adjust the coupling coefficient, resistance coefficient, and viscosity coefficient in the gas-air steady-state coupled control equation set. The adjustment range is ±5% to ±10% of the original parameters. After adjustment, return to S5 to re-perform the coupled iteration solution until the coupling verification result meets the engineering standard. When the accuracy of the extraction pipeline solution is ≥90%, the solution result is directly output for pipeline resistance analysis, extraction efficiency optimization, and gas disaster early warning.

[0054] This step, through adaptive coupling correction under working conditions, forms a closed-loop solution system of coupled inversion-verification-correction, dynamically optimizing the solution accuracy and giving the method strong adaptability to downhole working conditions and engineering practicality.

[0055] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A real-time solution method for global inversion of coupling characteristics of gas extraction pipeline networks, characterized in that: Includes the following steps: S1: Pre-extraction of gas-air coupling characteristics: Based on the flow characteristics of mixed gas in the pipeline network, extract the coupling correlation characteristics between gas concentration and mixed gas density, dynamic viscosity, friction coefficient, and Reynolds number, and determine the quantitative correlation formulas of each coupling parameter; S2: Coupled global topology abstraction modeling: Based on graph theory and pipeline topology, coupling features are incorporated to abstract pipelines into coupled directed flow path units carrying coupling parameters. Nodes are divided into basic coupled nodes and feature-enhanced coupled nodes to complete the parameterized representation of coupling and construct a coupled global topology abstraction model. S3: Construction of the gas-air steady-state coupled control equation set: Based on the coupling features extracted in step S1, a gas-air steady-state coupled control equation set is established that satisfies the three assumptions of one-dimensional compressibility, isothermal steady state, and continuous and consistent nodal pressure. Gas concentration is used as the core coupling variable to realize multi-parameter coupling constraints of pressure-flow-concentration-density. S4: Measurement point data acquisition and preprocessing: Strongly constrained measurement points with coupling characteristics are set up in key sections of the pipeline network to collect core monitoring data. The coupled quantization formula in step S1 is used to preprocess the data to obtain the coupled characteristic monitoring dataset, which serves as the coupling input constraint for the global inversion. S5: Constructing the objective function and solving iteratively: The gas-air steady-state coupled control equations are transformed into standard nonlinear coupled function equations. A coupled multi-objective weighted global inversion objective function is constructed and solved using a coupled adaptive numerical iterative algorithm with dual threshold convergence criteria to achieve global parameter inversion and coupled state reconstruction of the pipeline network. S6: Quantitative verification of reliability of coupled solution: Based on the coupling characteristics, a quantitative index system is constructed that includes the solution accuracy of the extraction pipeline network, the deviation rate of coupling parameters, and the coupling error of pump station operation, and multi-dimensional coupling verification of the full-domain inversion results is carried out. S7: Adaptive Coupling Correction under Working Conditions: If the coupling verification result in step S6 meets the standard, the solution result is output; if it does not meet the standard, the low-precision section is located, the coupling parameters are dynamically corrected under working conditions, and the process returns to step S5 to perform coupling iteration again, forming a closed-loop solution system of coupling inversion-verification-correction.

2. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 1, characterized in that: The specific process of pre-extracting the gas-air coupling characteristics in step S1 is as follows: by combining downhole field tests and laboratory simulations, the variation laws of the density, dynamic viscosity, friction coefficient, and Reynolds number of the mixed gas under different gas concentrations are determined. A nonlinear fitting method is used to establish a quantitative correlation formula between each parameter and the gas concentration. The density of the mixed gas is linearly negatively correlated with the gas concentration, while the dynamic viscosity, friction coefficient, and Reynolds number are nonlinearly exponentially correlated with the gas concentration.

3. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 1, characterized in that: The coupling parameters carried by the coupled directional flow path unit in step S2 include density, dynamic viscosity, friction coefficient, local resistance equivalent length, and Reynolds number associated with gas concentration; the basic coupling node is the intersection of the gas source and sink positions of the coal seam extraction borehole, gas pump station inlet, and vent with the coupled directional flow path unit, integrating gas convergence / diversion coupling flow characteristics; The enhanced coupling node is the location of the auxiliary component, integrating local resistance coupling loss characteristics; The basic coupling nodes and feature-enhanced coupling nodes together constitute the global coupling solution node system of the gas extraction pipeline network.

4. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 1, characterized in that: The three assumptions mentioned in step S3 are as follows: One-dimensional compressible steady-state assumption: Ignore the pipe path velocity gradient and non-uniform parameter distribution, retain only the axial flow variable, and realize the coupled change of gas density with pressure and gas concentration through a coupled quantization formula; Isothermal steady-state assumption: Based on the environmental characteristics of the underground roadway being closed and constant in temperature with sufficient heat exchange, significant changes in gas temperature are ignored, the energy equation constraints are simplified, and only the basic influence of temperature on the coupling parameters is retained; Assumption of continuous and consistent node pressure: At any coupled solution node in the pipeline network, the terminal pressure of each branch flowing into the node remains equal to the initial pressure of each branch flowing out of the node, ensuring the physical consistency of the coupled flow field.

5. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 4, characterized in that: The gas-air steady-state coupled control equation set is a four-parameter coupled equation set of pressure-flow rate-velocity-concentration, specifically including: the coupled equation of the one-dimensional compressible motion state of the mixed gas in the pipeline, the coupled equation of mass conservation and concentration transport of gas components, the coupled equation of the relationship between total flow rate and velocity of the mixed gas in the pipeline cross section, the coupled equation of flow convergence balance at the node, and the coupled equation of friction resistance and local resistance. The gas-air steady-state coupled control equations are designed for the series convergence conditions from node n to node n-1 and from node n+1 to node n-1 in the pipeline network. They define pressure drop, flow rate, and concentration correlation formulas that include gas concentration coupling terms. The integrated parameters include friction coefficient, local resistance equivalent length, and absolute pipe wall roughness, thereby achieving accurate characterization of the gas-air mixed gas flow process. Specifically defined as: In the formula, , and These are the mixed gas volume flow rates at the nth node, the (n+1)th node, and the (n-1)th node, respectively; , and These are the gas mixture densities at the nth node, the (n+1)th node, and the (n-1)th node, respectively. , and These represent the mixed gas concentrations at the nth node, the (n+1)th node, and the (n-1)th node, respectively. Let be the gas flow rate in the pipe at node i; Let be the flow velocity within the pipe at node i; Let be the inner diameter of the extraction pipe where the i-th node is located; Let be the gas density in the pipe at point i; Let be the gas pressure inside the pipe at node i; Let be the molar mass of the mixed gas in the pipe at node i; Here is the gas state constant; The temperature of the gas inside the pipe; The mechanical energy loss per unit weight of the mixed gas flow within the i-th pipe segment; Let be the friction coefficient of the i-th pipeline segment; Let be the Reynolds number of the mixed gas flow in the i-th pipeline segment; Let be the dynamic viscosity of the mixed gas in the i-th pipeline segment; The actual length of the i-th pipe segment; The equivalent length of the local resistance within the i-th pipeline segment; denoted as the absolute roughness of the i-th pipe segment; g is the acceleration due to gravity. , and These are the absolute pressures of the mixed gas at nodes n-1, n, and n+1, respectively. , and These are the elevations of the (n-1)th node, the nth node, and the (n+1)th node, respectively. , and These represent the mixed gas flow velocities in the (n-1)th, nth, and (n+1)th pipe segments, respectively. and These are the unit weight mechanical energy losses of the mixed gas flow in the nth and (n+1)th pipe segments, respectively. The system's pumping station extracts negative pressure; This refers to the absolute atmospheric pressure during pump station operation. The absolute pressure of the mixed gas in the pipeline at the pump station interface; Z1 represents the absolute pressure of the mixed gas in the adjacent pipeline section downstream of the pump station; Z2 and Z1 represent the elevations of the pump station interface node and the adjacent downstream node, respectively; V1 and V2 represent the flow velocities of the mixed gas in the pump station interface pipeline section and the adjacent downstream pipeline section, respectively. This refers to the mechanical energy loss per unit weight of the mixed gas flow within the pipeline section from the pump station interface to the downstream adjacent node.

6. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 1, characterized in that: The principle for setting up the coupling characteristic strong constraint measuring points in step S4 is to fully cover the key sections of the coupled flow state. The number of measuring points is 5%-15% of the total number of nodes in the pipeline network. Each measuring point synchronously collects core parameters reflecting the gas-air coupling characteristics. The coupling characteristic data of the measuring points realizes strong constraint on the coupled flow state of the entire pipeline network and prevents interference from invalid data from non-coupled measuring points.

7. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 1, characterized in that: The unknown vector of the coupled multi-objective weighted global inversion objective function described in step S5 includes the pressure and gas velocity of all nodes in the pipeline network to be solved. The objective function is constructed based on the optimal synergy between physical coupling constraints and numerical solution accuracy. The components and their weights are as follows: The core coupling error term is the mean square error between the coupled characteristic monitoring data and the model's coupled calculation value, accounting for 60%-70% of the weight; the pipeline pressure drop coupling constraint term is the deviation threshold between the actual pipeline pressure drop and the theoretical pressure drop of the coupled model, accounting for 20%-25% of the weight; the flow conservation coupling penalty term is the coupling imbalance of the node's confluence flow, accounting for 5%-10% of the weight; each weight coefficient is adaptively adjusted according to the downhole extraction conditions; The specific definition of the coupled multi-objective weighted global inversion objective function is as follows: , where: unknown vector It contains the pressure and gas velocity of all the nodes to be solved, i.e. , This represents the total number of pipeline nodes. It consists of n-1 pressure drop equations and n-1 nodal flow equations. Each equation corresponds to a residual function, which represents the error change between the measured values ​​at finite measuring points and the theoretical calculation values ​​of the model. If the residual function is 0, the equation satisfies the set physical conservation conditions.

8. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 7, characterized in that: The coupled adaptive numerical iterative algorithm described in step S5 is either the Newton iteration method based on topological coupling constraints or the adaptive damped least squares method. The initial values ​​of the algorithm are engineering-based initial values ​​that conform to the physical laws of downhole negative pressure extraction: the initial pressure value is estimated based on the rated negative pressure of the extraction pump, along the pipeline towards the borehole direction according to the linear decay law of friction resistance; the initial flow velocity value is selected from the economic flow velocity range of 4-10 m / s, and is calculated in combination with the design flow rate of the extraction system and the cross-sectional area of ​​the pipeline. The coupled adaptive numerical iterative algorithm is specifically defined as follows: In the formula, This is the initial iteration vector; The initial iterative pressure value for the i-th node; For the first The length of the pipeline section; This is the total length of the pipeline; Let be the initial iterative value of the flow velocity in the i-th pipe segment; Design the flow rate for the extraction system; Let be the cross-sectional area of ​​the i-th pipe segment; This represents the current iteration number; Let k be the pressure balance residual function; The pressure at node i-1 after the k-th iteration; The pressure at node i after the k-th iteration; Let be the friction coefficient of the i-th pipeline segment; Let be the inner diameter of the i-th pipe segment; Let be the density of the mixed gas in the i-th pipeline segment after k iterations; denoted as , where is the flow velocity in the i-th pipe segment after the k-th iteration; g is the acceleration due to gravity. Let be the elevation difference between the two nodes of the i-th pipeline segment; Let this be the (k+n)th flow balance residual function; It is the sum of the volumetric flow rates of all inflow branches to the i-th node after the k-th iteration; It is the sum of the volumetric flow rates of all outflow branches of the i-th node after the k-th iteration; This is the global residual vector; It is a topologically constrained sparse Jacobian matrix; This is the correction vector for the k-th iteration; It is a relaxation factor; Let be the iteration vector for the (k+1)th iteration.

9. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 8, characterized in that: The core feature of the coupled adaptive numerical iterative algorithm is the synchronous iteration of coupling parameters. During the iteration process, node pressure, branch flow, gas concentration, and various coupled parameters are simultaneously corrected. The iteration convergence determination adopts a dual threshold determination of residual accuracy coupling threshold and parameter accuracy coupling threshold, with the residual accuracy coupling threshold set to 10. -5 ~10 -6 Pa, parameter precision coupling threshold is set to 10. -4 ~10 -5 Pa, and set the maximum number of iterations to 50; The specific definition of iterative convergence criterion is: in, The Euclidean norm of the updated residual vector; This is the m-th component of the global residual vector; The residual accuracy threshold is set to 10. -5 ~10 -6 If Pa satisfies the above conditions, then the residual is considered to be small enough, and the system of equations satisfies the equilibrium conditions of the laws of conservation of energy and conservation of mass. This represents the relative change of the updated unknown vector; Let be the relative change of the t-th unknown vector; The parameter precision threshold is set to 10. -4 ~10 - 5 Pa.

10. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 9, characterized in that: The coupled adaptive numerical iterative algorithm introduces a topologically constrained sparse Jacobian matrix, where the matrix elements are the first-order partial derivatives of the residual function with respect to each unknown. The partial derivatives are non-zero only when there is a physical coupling relationship between the unknown and the equation, and are zero otherwise. The efficiency of coupled iteration is improved through sparsification.

11. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 10, characterized in that: The termination verification process of the coupled adaptive numerical iterative algorithm consists of three levels of judgment: First, if both the residual accuracy coupling threshold and the parameter accuracy coupling threshold are satisfied simultaneously, the iteration is terminated, and the final solution for the pressure and gas flow rate at each node is output. Secondly, if the double threshold is not met, let the iteration number k = k + 1, and return to recalculate the residual vector and the topologically constrained sparse Jacobian matrix; Third, if the number of iterations exceeds the maximum number of iterations (50) and still fails to converge, a backtracking check is triggered to verify in sequence whether the initial value conforms to the physical law of underground negative pressure extraction, whether the sign of the residual function is correct, and whether the calculation of the partial derivative of the Jacobian matrix is ​​correct.

12. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 1, characterized in that: The gas-air coupled solution reliability quantification index system described in step S6 includes three quantitative indicators: the solution accuracy of the extraction pipeline network, the deviation rate of the coupling parameters, and the coupling error of the pump station operation. Among them, the accuracy rate of the extraction pipeline network calculation is the proportion of the number of coupled nodes that meet the engineering qualification standards to the total number of coupled monitoring nodes. The engineering qualification standards are that the relative error between the calculated negative pressure of the node and the actual value measured by the sensor is ≤5%, and the relative error between the calculated value of gas concentration and the actual value is ≤3%. The coupling parameter deviation rate is the deviation rate between the calculated coupling parameters and the values ​​calculated by the quantization formula of the pre-extracted coupling features of S1. A deviation rate of ≤2% is considered acceptable. The pump station operation coupling error is the deviation between the average negative pressure under stable operating conditions and the operating negative pressure under coupled solution conditions. A deviation ≤ 1 kPa is considered acceptable. The coupling verification result is deemed to conform to engineering application standards only when all three quantitative indicators meet the qualification criteria; the specific definitions of the reliability quantitative indicators are as follows: In the formula, This indicates the accuracy rate of the extraction pipeline network calculation; The total number of all nodes in the pipeline network; This represents the number of nodes with a relative error of less than 5%. Let be the solution error for the i-th node; The measured negative pressure at node i; The theoretical negative pressure for solving the i-th node; This refers to the relative error in pump station operation. The average negative pressure of the pumping station under stable operating conditions for a period of time; To calculate the operating negative pressure of the pumping station under the given conditions; To account for the relative error, the maximum value of the single-node solution error and the pump station operation error is taken as the upper limit of the overall solution error.

13. The real-time solution method for global inversion of coupling characteristics of gas extraction pipeline network according to claim 1, characterized in that: The specific process of the adaptive coupling correction described in step S7 is as follows: Based on the non-conforming items of the reliability quantification index of the coupled solution, the core problem of the low-precision solution section is located. The coupling coefficient, resistance coefficient, and viscosity coefficient in the gas-air steady-state coupled control equation set are adaptively adjusted using a coupling parameter dynamic optimization algorithm. The adjustment range is ±5% to ±10% of the original parameters. After adjustment, the process returns to S5 to perform coupling iteration solution again until the coupling verification result meets the engineering standard. When the accuracy of the extraction pipeline solution is ≥90%, the solution result is directly output for pipeline resistance analysis, extraction efficiency optimization, and gas disaster early warning.

Images