An Optimization Method and System for Relay Pump Layout of a Fluid-Based Shield Machine

Abstract

The present invention discloses an optimization method and system for the layout of relay pumps of a shield machine based on fluid, belonging to the technical field of slurry shield machines. A fluid control model is constructed to obtain the theoretical pressure values at the inlet and outlet of the pipeline. Based on the pipeline parameters and fluid characteristics, a pipeline pressure loss function is constructed to obtain the total pipeline pressure loss. The efficient working range of the water pump is determined, and the high-efficiency working range of the head and the high-efficiency working range of the flow rate of the relay pump are matched, and the layout mode of the relay pump is selected. Based on the layout mode of the relay pump, a three-dimensional model is re-modeled, and a fluid dynamics analysis model is constructed for the three-dimensional model. The actual pressure value at the pipeline input port is substituted into the fluid dynamics analysis model as a known condition, and the selected relay pump distribution mode is verified through simulation calculation. The pipeline pressure loss function of the present invention distinguishes between the frictional pressure loss and the local pressure loss, and can accurately calculate the total pipeline pressure loss, providing reliable pressure reference data for the layout of the relay pump and the operation of the entire system.

Description

Technical Field

[0001] The present invention belongs to the technical field of the layout method of relay pumps for slurry balance shield machines, and particularly relates to an optimization method and system for the layout of relay pumps of a shield machine based on fluid. Background Art

[0002] The slurry circulation system of a slurry shield mainly consists of slurry inlet and outlet pumps, slurry inlet and outlet pipelines, control valves, quarry boxes, and pipeline extension mechanisms, etc. It is one of the core systems of the total slurry shield system and plays an important role in aspects such as maintaining the stability of the excavation face, transporting muck, and tunnel sewage discharge. The slurry shield transports muck to the outside of the tunnel through the slurry circulation method. Therefore, transporting muck by slurry is an important feature of slurry shield construction, and the slurry transportation system is also the key system of the slurry shield. The configuration and selection of the slurry circulation system directly determine the construction efficiency of the slurry shield. Due to the frictional loss along the pipeline, the relationship between uphill and downhill, and the influence on the slurry balance of the shield, the pressure distribution in the pipeline is always changing. In order to reasonably arrange the position of the slurry pump, it is necessary to analyze the pressure distribution in the slurry pipeline.

[0003] Regarding the current state of the existing technology, when facing complex geological conditions, such as encountering special geological conditions such as gravel, hard rock, or boulders, there are still obvious deficiencies in the selection and layout of relay pumps. Specifically, in the process of selection and layout, the slag-carrying capacity of the slurry and the change of pipeline resistance are often not fully and comprehensively considered. In this way, it is extremely easy to cause the performance of the selected relay pump in actual construction to not match the complex and changeable construction requirements, thereby affecting the efficiency and quality of the entire slurry shield construction.

[0004] In addition, in past construction practices, the estimation of the pressure loss of the inlet and outlet slurry mostly relied on manual experience. This method has great limitations. It is difficult to fully consider the dynamic change of the pipeline pressure loss caused by the formation change during the actual tunneling process, and it is even more unable to properly handle the many difficulties brought by the layout of multi-section pipelines. The existence of these problems has restricted the efficient development of slurry shield construction under complex geological conditions to a certain extent and urgently needs to be solved by more scientific and reasonable technical means. Summary of the Invention

[0005] The present invention provides an optimization method and system for the layout of relay pumps of a shield machine based on fluid to solve the technical problems existing in the above background art.

[0006] The present invention adopts the following technical solutions: An optimization method for the layout of relay pumps of a shield machine based on fluid, comprising the following steps:

[0007] Build a fluid control model for incompressible fluid, and solve the fluid control model to obtain the theoretical pressure values at the inlet and outlet of the pipeline;

[0008] Build a pipeline pressure loss function based on pipeline parameters and fluid characteristics, and calculate the total pipeline pressure loss using the pipeline pressure loss function;

[0009] Determine the efficient working range of the water pump, and match the high-efficiency working range of the head and the high-efficiency working range of the flow rate of the relay pump in the efficient working range of the water pump in combination with the total pipeline pressure loss; in the high-efficiency working range of the head, adopt a layout mode based on the conveying distance for the relay pump, and in the high-efficiency working range of the flow rate, adopt a layout mode based on the conveying flow rate for the relay pump;

[0010] Re-model based on the layout mode of the relay pump to obtain a three-dimensional model, and build a fluid dynamics analysis model for the three-dimensional model. The fluid dynamics analysis model takes into account the working characteristics of the relay pump and its interaction with the fluid in the pipeline, and sets corresponding boundary conditions;

[0011] Substitute the actual pressure value at the pipeline inlet as a known condition into the fluid dynamics analysis model, and conduct a preliminary verification on the selected relay pump distribution mode through simulation calculation.

[0012] In a further embodiment, the following steps are further included:

[0013] Build a neural network model, and incorporate the fluid control model and the fluid dynamics analysis model as constraint conditions into the neural network model;

[0014] Use the pipeline parameters, fluid characteristics, and relay pump parameters as the input layer, and the flow state, pressure distribution, and velocity distribution in the pipeline as the output layer; use the theoretical pressure values at the inlet and outlet of the pipeline and the actual pressure value at the pipeline inlet to train the neural network model to obtain an optimized neural network model;

[0015] Use the optimized neural network model to conduct a simulation verification on the distribution mode of the relay pump to obtain a simulation result, and optimize the parameters in the process of confirming the distribution mode of the relay pump according to the simulation result.

[0016] In a further embodiment, the expression form of the fluid control model for incompressible fluid is as follows:

[0017] In three-dimensional space The mass conservation equation of the incompressible fluid is:

[0018] ; where , [[ID=,40]]and are the fluid in the three axial directions Velocity vector on

[0019] For an incompressible fluid, the expression form of the momentum conservation equation in each direction is as follows:

[0020] ; where and represent any direction, represents time, , are respectively the components of the velocity in the direction and the direction ; represents the pressure, is the fluid density, represents the component of the external force acting on the unit volume of the fluid in the direction ; and are the lengths in the direction and the direction ; , represents the fluid viscosity;

[0021] For an incompressible fluid, the expression form of the energy conservation equation is as follows:

[0022] ; where represents the total energy per unit mass, is the heat conduction flux, represents the dissipation rate caused by viscosity, represents the divergence operation, is the divergence operation on the vector ; is the divergence operation on the heat conduction flux, is the velocity vector of the fluid, is the external force per unit volume of the fluid.

[0023] In a further embodiment, the expression of the pipeline pressure loss function is as follows:

[0024] ; where is the total pipeline pressure loss;

[0025] is the frictional pressure loss: , is the fluid density, is the acceleration due to gravity, is the frictional head loss, which is a function of the pipeline length and the pipeline inner diameter;

[0026] is the local pressure loss: , is the local head loss.

[0027] In a further embodiment, the process for determining the efficient operating range of the water pump is as follows:

[0028] The self-performance of the water pump is expressed by the following formula: , where , and are constants and are the performance parameters of the water pump; is the performance flow rate of the water pump, is the head corresponding to the self-performance of the water pump;

[0029] Calculate the flow rate difference and head difference of the water pump:

[0030] ; where and are respectively the head and flow rate at the highest efficiency point of the water pump;

[0031] The boundaries of the efficient operating range of the water pump are determined by the following formula:

[0032] and ; where and are respectively the given head tolerance and flow rate tolerance.

[0033] In a further embodiment, the matching process of the head efficient operating range of the relay pump is as follows:

[0034] According to the total pipeline pressure loss calculate the corresponding head loss :

[0035] , is the fluid density, is the acceleration due to gravity;

[0036] The actual total head of the water pump is : , is the head corresponding to the self-performance of the water pump;

[0037] If the actual total head of the water pump meets the head efficient operating range judgment condition, it means that the corresponding point is the head efficient operating range of the relay pump, and the layout mode based on the conveying distance should be adopted; the expression form of the head efficient operating range judgment condition is:

[0038] ;

[0039] Where, It is the head of the pump at its highest efficiency point. is the given head tolerance.

[0040] In a further embodiment, the matching process of the flow efficient working area of the relay pump is as follows:

[0041] Considering the impact of the total pressure loss in the pipeline, the actual flow rate of the pump is calculated using the following formula: :

[0042] ;

[0043] Where, 、 and is a constant, which is the performance parameter of the water pump; is the coefficient reflecting the pipeline resistance. is the actual total head of the pump;

[0044] If the actual flow of the pump If the flow rate efficient working area judgment condition is met, it means that the corresponding point is the flow rate efficient working area of the relay pump, and the layout mode based on the delivery flow rate should be adopted; the expression form of the flow rate efficient working area judgment condition is:

[0045] ;

[0046] Where, It is the flow rate at the highest efficiency point of the pump. is the given flow tolerance.

[0047] In a further embodiment, the fluid dynamics analysis model is expressed as follows:

[0048] In three-dimensional space The mass conservation equation for incompressible fluid is:

[0049] Where, 、 and The fluid in the model is analyzed in three directions. Velocity vector on ;

[0050] For an incompressible fluid, the conservation equation for momentum in each direction is expressed as follows:

[0051] ;

[0052] in, represents the fluid density, Represents pressure, and are the components of the velocity in the directions and respectively, and are the lengths in the directions and respectively, , is the fluid viscosity, represents the component of the external force acting on the unit volume of the fluid in the direction respectively, is the additional force exerted on the unit volume of the fluid by the relay pump in the direction ; is the pressure increment generated by the relay pump, is the length of the infinitesimal element in the direction respectively, is the Dirac function, represents the coordinate position of any point, is the coordinate position of the relay pump;

[0053] For an incompressible fluid, the expression form of the energy conservation equation is as follows:

[0054] ;

[0055] In the formula, represents the total energy per unit mass, is the velocity vector of the fluid, is the heat conduction flux, represents the dissipation rate caused by viscosity, is the divergence operation on the vector respectively, is the divergence operation on the heat conduction flux, is the external force per unit volume of the fluid, is the energy input by the relay pump to the unit volume of the fluid; is the rated flow rate of the relay pump, [[ID=[]71]]is the infinitesimal volume containing the pump.

[0056] In a further embodiment, the boundary conditions at least include:

[0057] ;

[0058] In the formula, is the outlet pressure of the relay pump, is the inlet pressure of the relay pump, is the pressure increment generated by the relay pump, is the rated flow rate of the relay pump, is the inlet flow rate of the relay pump, It should be noted that there may be an error in the original text where "[[ID=[]71]]" is presented. I have translated it as "is the infinitesimal volume containing the pump." according to the context. Please check the original text for accuracy. is the outlet flow rate of the relay pump.

[0059] A fluid-based relay pump layout optimization system for a shield machine, which is used to implement the relay pump layout optimization method for the shield machine as described above, includes:

[0060] The first module is configured to construct a fluid control model for incompressible fluid, and solve the fluid control model to obtain the theoretical pressure values at the inlet and outlet of the pipeline;

[0061] The second module is configured to construct a pipeline pressure loss function based on pipeline parameters and fluid characteristics, and calculate the total pipeline pressure loss by using the pipeline pressure loss function;

[0062] The third module is configured to determine the efficient working range of the water pump, and match the high-efficiency working range of the head and the high-efficiency working range of the flow rate of the relay pump in the efficient working range of the water pump in combination with the total pipeline pressure loss; in the high-efficiency working range of the head, a layout mode based on the conveying distance is adopted for the relay pump, and in the high-efficiency working range of the flow rate, a layout mode based on the conveying flow rate is adopted for the relay pump;

[0063] The fourth module is configured to re-model based on the determined pipeline distribution and the selected relay pump to obtain a three-dimensional model including the relay pump; construct a fluid dynamics analysis model for the three-dimensional model, and the fluid dynamics analysis model takes into account the working characteristics of the relay pump and its interaction with the fluid in the pipeline, and set corresponding boundary conditions;

[0064] The fifth module is configured to substitute the actual pressure value at the pipeline input port as a known condition into the fluid dynamics analysis model, and preliminarily verify the selected relay pump distribution mode through simulation calculation.

[0065] Advantages of the present invention: By constructing a fluid control model based on incompressible fluid, covering the equations of mass conservation, momentum conservation and energy conservation, the present invention can comprehensively and accurately describe the actual flow characteristics and physical state changes of the fluid in the shield machine pipeline. This provides a solid theoretical foundation for subsequent analysis, makes the simulation of fluid behavior closer to the actual working conditions, and helps to deeply understand and grasp the movement law of the fluid in the entire slurry shield mud circulation system.

[0066] The pipeline pressure loss function constructed by the present invention based on pipeline parameters and fluid characteristics carefully distinguishes the frictional pressure loss and the local pressure loss, and clarifies the influence relationship of each factor (such as friction factor, pipe diameter, local resistance coefficient, etc.) on the pressure loss. In this way, the total pipeline pressure loss can be accurately calculated, avoiding the errors brought by traditional estimation methods, and providing reliable pressure reference data for the layout of the relay pump and the operation of the entire system.

[0067] On the basis of determining the efficient working range of the water pump, the head efficient working range and the flow rate efficient working range of the booster pump are accurately matched in combination with the total pipeline pressure, and the layout mode based on the conveying distance and the conveying flow rate is correspondingly adopted. This method fully considers the complex and changeable working conditions in actual construction and the demand differences of flow rate and head at different positions, ensures the optimization of the layout position of the booster pump, effectively improves the collaborative working efficiency of the booster pump in the whole system, and avoids the performance waste or the situation that the construction requirements cannot be met caused by unreasonable layout.

[0068] By accurately matching the working range and reasonably arranging, the booster pump can operate stably within its efficient working range, maximizing its performance advantages. This not only helps to improve the muck conveying efficiency and ensure the continuity of the slurry shield construction, but also can reduce energy consumption, extend the service life of the booster pump, reduce the construction cost from the perspective of equipment operation, and improve the overall economic benefits.

[0069] Finally, when constructing the fluid dynamics analysis model, the working characteristics of the booster pump and its interaction with the fluid in the pipeline are fully considered, and the corresponding boundary conditions are set. This enables the model to truly reflect the complex dynamic changes of the fluid in the actual pipeline system including the booster pump, provides a powerful tool for accurately simulating and analyzing the fluid state under different working conditions, and improves the accuracy of the assessment of the operation state of the entire slurry shield mud circulation system.

[0070] Substituting the actual pressure value at the pipeline input port into the fluid dynamics analysis model for preliminary verification, combined with the further simulation verification using the neural network model subsequently, a complete verification system from theory to practice, from preliminary judgment to in-depth optimization is formed. Description of the Drawings

[0071] Figure 1 is the flowchart of the layout optimization method of the shield machine booster pump based on fluid in Embodiment 1.

[0072] Figure 2 is the architecture diagram of the PINNs physics-informed neural network in Embodiment 2. Detailed Implementation Modes

[0073] The present invention will be further described below in conjunction with the accompanying drawings of the specification and embodiments.

[0074] Embodiment 1

[0075] This embodiment provides a layout optimization method of the shield machine booster pump based on fluid to solve the technical problems existing in the above-mentioned background technology, as Figure 1 shown, including the following steps:

[0076] Build a fluid control model for an incompressible fluid, and solve the fluid control model to obtain the theoretical pressure values at the inlet and outlet of the pipeline;

[0077] Construct a pipeline pressure loss function based on pipeline parameters and fluid characteristics, and calculate the total pipeline pressure loss using the pipeline pressure loss function;

[0078] Determine the efficient working range of the water pump, and match the high-efficiency working range of the head and the high-efficiency working range of the flow rate of the relay pump in the efficient working range of the water pump in combination with the total pipeline pressure loss; in the high-efficiency working range of the head, adopt a layout mode based on the conveying distance for the relay pump, and in the high-efficiency working range of the flow rate, adopt a layout mode based on the conveying flow rate for the relay pump;

[0079] Re-model based on the layout mode of the relay pump to obtain a three-dimensional model, and construct a fluid dynamics analysis model for the three-dimensional model. The fluid dynamics analysis model takes into account the working characteristics of the relay pump and its interaction with the fluid in the pipeline, and sets corresponding boundary conditions;

[0080] Substitute the actual pressure value at the pipeline inlet as a known condition into the fluid dynamics analysis model, and preliminarily verify the selected relay pump distribution mode through simulation calculation.

[0081] In a further embodiment, the expression form of the fluid control model for an incompressible fluid is as follows:

[0082] In three-dimensional space The mass conservation equation of an incompressible fluid is:

[0083] ; where , and Are the velocity vectors of the fluid in the three-axis directions respectively;

[0084] For an incompressible fluid, the expression form of the momentum conservation equation in each direction is as follows:

[0085] ; where and Represent any direction, Represents time, , Are the components of the velocity in the direction and the direction respectively, Represents the pressure, Is the fluid density, Represents the component of the external force acting on the unit volume of the fluid in the direction respectively, and is the direction and the direction the length in, , represents the fluid viscosity;

[0086] For an incompressible fluid, the expression form of the energy conservation equation is as follows:

[0087] ; where, represents the total energy per unit mass, is the heat conduction flux, represents the dissipation rate caused by viscosity, represents the divergence operation, is the divergence operation on the vector of, is the divergence operation on the heat conduction flux, is the velocity vector of the fluid, is the external force per unit volume of the fluid. , where, are the components of the shear stress tensor.

[0088] It should be noted that the fluid control model in this embodiment is for an incompressible fluid mainly because the fluid in the pipeline is a low-speed liquid, so there will be no compression in practice, and in this way, the calculation equation of the fluid will be appropriately simplified.

[0089] In a further embodiment, the expression of the pipeline pressure loss function is as follows:

[0090] ; where, is the total pipeline pressure loss;

[0091] is the frictional pressure loss: , is the fluid density, is the acceleration due to gravity, is the frictional head loss, which is a function of the pipeline length and the pipeline inner diameter.

[0092] Further expressed as, , is the friction factor, is the pipeline length, is the pipeline inner diameter, is the average fluid velocity. By using the pipeline pressure loss function of this embodiment, different parameters are changed (such as adjusting the pipe diameter, planning the pipeline length, predicting the situation under different flow velocities, etc.) to simulate and analyze the pressure loss situation, and then optimize the pipeline layout and the selection and layout of relay pumps, so as to achieve the purpose of reducing the total pressure loss and improving the fluid transportation efficiency. At the same time, when problems occur in the operation of the pipeline system (such as finding that the actual pressure does not match the expected value, the fluid transportation is unsmooth, etc.), with the help of this function, the reasons are investigated from the links represented by each parameter, and it is judged whether the pressure loss abnormality is caused by too long pipeline length, unreasonable pipe diameter, too large local resistance of pipe fittings or other factors, so as to quickly locate the problem and take corresponding solutions to ensure the normal operation of the fluid transportation system such as the slurry circulation of the shield machine.

[0093] Correspondingly, is the local pressure loss: , is the local head loss. , where is the local resistance coefficient. It is worth mentioning that the local resistance coefficient is mainly considered because: for pipe fittings of different types, specifications and materials (such as 90° elbows, tees, gate valves, etc.), the internal flow channel shapes, roughness and other factors are different. For example, a pipe fitting with a complex structure, a large degree of flow channel bending and a relatively rough inner wall will have a much larger local resistance coefficient than a pipe fitting with a simple structure and a smooth inner wall, resulting in a greater local pressure loss. By determining the local resistance coefficients of different pipe fittings, the local pressure losses generated by them in the pipeline system can be accurately calculated, so that the pressure loss situation caused by the existence of pipe fittings in the entire pipeline system can be accurately quantified, providing a strong basis for accurately analyzing and optimizing the pipeline layout.

[0094] In another embodiment, the determination process of the high-efficiency working range of the water pump is as follows:

[0095] The following formula is used to represent the performance of the water pump: , where , and are constants and are the performance parameters of the water pump; is the performance flow rate of the water pump, is the head corresponding to the performance of the water pump itself;

[0096] Calculate the flow rate difference and head difference of the water pump: ; where and are the head and flow rate at the highest efficiency point of the water pump respectively;

[0097] The boundaries of the high-efficiency working range of the water pump are determined by the following formula:

[0098] And ; In the formula, And Are the given head tolerance and flow tolerance respectively.

[0099] Correspondingly, the matching process of the high-efficiency working range of the head of the relay pump is as follows:

[0100] According to the total pressure loss of the pipeline Calculate the corresponding head loss :

[0101] , Is the fluid density, Is the acceleration of gravity;

[0102] The actual total head of the water pump is : , Is the head corresponding to the performance of the water pump itself;

[0103] If the actual total head of the water pump Meets the judgment condition of the high-efficiency working range of the head, it means that the corresponding point is the high-efficiency working range of the head of the relay pump, and the layout mode based on the conveying distance should be adopted; The expression form of the judgment condition of the high-efficiency working range of the head is:

[0104] ;

[0105] In the formula, Is the head at the highest efficiency point of the water pump, Is the given head tolerance.

[0106] It should be noted that when calculating the actual total head of the water pump as Fully consider the influence of the total pressure loss of the pipeline on the head of the water pump, incorporate the actual pressure loss situation in the pipeline system, and comprehensively and accurately reflect the total head required for the water pump to overcome the pipeline resistance and maintain the normal flow of the fluid under actual working conditions, making the calculation result more in line with the real engineering scenario and avoiding the error caused by only relying on the performance parameters of the water pump itself and ignoring the actual pressure loss of the pipeline.

[0107] Based on this, by Determine the high-efficiency working range of the head of the relay pump, and this method provides a clear and definite standard for judging whether the relay pump is in an efficient operating state.

[0108] Among them, And For the highest efficiency point and head tolerance of a given water pump, a reasonable range can be defined through subsequent models for optimization.

[0109] The matching process of the flow efficient working area of the relay pump is as follows: Considering the influence of the total pipeline pressure loss, the actual flow rate of the water pump is calculated using the following formula :

[0110] ;

[0111] In the formula, , and are constants and are the performance parameters of the water pump; is the coefficient reflecting the pipeline resistance situation, is the actual total head of the water pump.

[0112] During the operation of the system, the pipeline resistance may change over time and usage conditions. For example, the wear of the pipeline inner wall, the accumulation of dirt, and the damage of pipe fittings will all cause the resistance coefficient to change. This formula can calculate the actual flow rate of the water pump in real time and accurately according to the dynamic change of the value. In contrast, the existing technology lacks this dynamic adaptability to pipeline working conditions changes and cannot adjust the flow calculation results in time, which may lead to unreasonable selection and layout of water pumps and affect the operation efficiency and stability of the system.

[0113] Correspondingly, the parameters , and reflect its unique flow - head characteristic curve and also combine the total head requirements under actual working conditions, and a reasonable value can be defined through subsequent models for optimization.

[0114] If the actual flow rate of the water pump meets the flow efficient working area judgment condition, it means that the corresponding point is the flow efficient working area of the relay pump, and the layout mode based on the conveying flow rate should be adopted; the expression form of the flow efficient working area judgment condition is:

[0115] ;

[0116] In the formula, is the flow rate at the highest efficiency point of the water pump, is the given flow tolerance, and a reasonable range can be defined through subsequent models for optimization.

[0117] Through To determine the high-efficiency flow working range of the relay pump so that the relay pump can operate within its high-efficiency flow range. This helps improve the operating efficiency of the water pump, reduce energy consumption, and extend the service life of the water pump. It effectively solves the problem in the prior art that due to inaccurate flow calculation, the water pump operates in an inefficient range for a long time, not only wasting energy but also increasing equipment wear and maintenance costs.

[0118] According to the above technical solution, the layout mode of the relay pump is determined. It is laid out based on the high-efficiency working range of the water pump and the accurately matched high-efficiency working range of the lift and flow of the relay pump, so that the relay pump can be in the high-efficiency working state to the greatest extent during operation. Within the high-efficiency working range, the relay pump can achieve a good fluid transportation effect with lower energy consumption, reduce energy waste, lower operating costs, and also help extend the service life of the relay pump, improving the economy and stability of the entire system from the perspective of equipment operation. For example, through accurate calculation and layout, it is avoided that the relay pump exceeds the high-efficiency working range due to too high lift or too large flow, resulting in unnecessary energy loss and excessive equipment wear.

[0119] Multiple relay pumps in the entire system can work together in their respective high-efficiency working areas, making the energy utilization of the entire slurry shield mud circulation system more reasonable and efficient. Compared with the situation of randomly laying out relay pumps or not accurately considering their high-efficiency working areas, this layout mode can effectively integrate the functions of each relay pump, avoid the phenomenon that some equipment has high energy consumption but low efficiency, realize the optimal allocation of energy, meet the requirements of modern engineering construction for energy conservation and emission reduction, and help improve the overall benefit of the project.

[0120] Therefore, in order to verify the layout mode of the relay pump obtained by the above method, a new model is built based on the determined pipeline distribution and the selected relay pump. Therefore, in the current scenario, the hydrodynamic analysis model for three-dimensional model construction should consider the working characteristics of the relay pump and its interaction with the fluid in the pipeline.

[0121] In a further embodiment, the expression form of the hydrodynamic analysis model is as follows:

[0122] In three-dimensional space the mass conservation equation of an incompressible fluid is:

[0123] ; where , and are the velocity vectors of the fluid in the hydrodynamic analysis model in the three axial directions respectively;

[0124] For an incompressible fluid, the expression form of the momentum conservation equation in each direction is as follows:

[0125] ;

[0126] wherein, represents the fluid density, represents the pressure, and are respectively the components of the velocity in the directions and ; and are the lengths in the directions and ; , is the fluid viscosity, represents the component of the external force acting on the unit volume of the fluid in the direction ; is the additional force exerted on the unit volume of the fluid by the relay pump in the direction ; is the pressure increment generated by the relay pump, is the differential length in the direction ; is the Dirac function, represents the coordinate position of any point, is the coordinate position of the relay pump;

[0127] For an incompressible fluid, the expression form of the energy conservation equation is as follows:

[0128] ;

[0129] In the formula, represents the total energy per unit mass, is the velocity vector of the fluid, is the heat conduction flux, represents the dissipation rate caused by viscosity, is the divergence operation on the vector ; is the divergence operation on the heat conduction flux, is the external force per unit volume of the fluid, is the energy input by the relay pump to the unit volume of the fluid; is the rated flow rate of the relay pump, is the differential volume containing the pump.

[0130] The model closely conforms to the true flow characteristics of fluids and the working conditions of relay pumps in actual projects such as slurry shield tunneling from a physical essence. Compared with some models in the prior art that are overly simplified or lack strong generality, this model can more accurately simulate various physical phenomena of fluids in a complex pipeline system under the action of relay pumps, such as fluid acceleration, pressure change, energy conversion, etc., providing a more reliable theoretical basis for verifying the layout mode of relay pumps.

[0131] In addition to the conventional inlet and outlet flow conditions and inlet and outlet velocity conditions, the boundary conditions in this embodiment also include: ;

[0132] In the formula, is the outlet pressure of the relay pump, is the inlet pressure of the relay pump, is the pressure increment generated by the relay pump, is the rated flow of the relay pump, is the inlet flow of the relay pump, is the outlet flow of the relay pump.

[0133] It clarifies the relationship between the pressure difference and flow rate at the inlet and outlet of the relay pump. In actual projects, the inlet and outlet states of the relay pump play a key role in the fluid flow of the entire pipeline system, while the prior art may not be comprehensive or accurate enough in setting boundary conditions to accurately reflect these key factors. This model can better simulate the flow behavior of fluids in the relay pump and its surrounding pipeline areas under actual working conditions, improving the fit between the simulation results and the actual situation, and thus more effectively verifying the rationality of the relay pump layout mode.

[0134] Embodiment 2

[0135] Based on the method for optimizing the layout of the relay pump of the shield machine based on fluid disclosed in Embodiment 1, this embodiment also discloses the following steps:

[0136] Construct a neural network model. In this embodiment, the Physics-Informed Neural Networks (PINNs) is taken as an example. The fluid control model and the fluid dynamics analysis model are incorporated into the neural network model as constraint conditions.

[0137] Taking pipeline parameters, fluid characteristics, and relay pump parameters as the input layer, and the flow state, pressure distribution, and velocity distribution inside the pipeline as the output layer; using the theoretical pressure values at the inlet and outlet of the pipeline and the actual pressure value at the input port of the pipeline to train the neural network model to obtain an optimized neural network model;

[0138] Use the optimized neural network model to simulate and verify the distribution mode of the relay pump to obtain simulation results, and optimize the parameters in the process of confirming the distribution mode of the relay pump according to the simulation results.

[0139] In other words, the PINNs (Physics-Informed Neural Networks) at least include: a neural network model, a loss function L, and residuals. The structural layer includes: an input layer, a hidden layer, and an output layer.

[0140] Furthermore, a PINNs physics-informed neural network model is constructed, and physical laws (such as mass conservation, momentum conservation, etc.) and boundary conditions in the fluid control equation and the fluid dynamics analysis model are incorporated as constraint conditions into the structure and training process of the neural network. Pipeline parameters, fluid characteristics, relay pump parameters, etc. input are used as the input layer, and the pressure distribution and flow velocity distribution in the pipeline, etc. are used as the output layer; theoretical calculation data (such as the theoretical pressure values at the inlet and outlet of the pipeline, the total pressure loss of the pipeline, etc.) and actual engineering measurement data (such as the actual pressure value at the input port of the pipeline) are used to train the PINNs model. During the training process, the weights and biases of the neural network are continuously adjusted to minimize the error between the model output result and the actual data, while ensuring that the model output satisfies the incorporated physical laws.

[0141] The trained PINNs model is used to simulate and verify different relay pump distribution patterns to obtain simulation results. According to the simulation results, parameters (such as some set parameters and range values in Embodiment 1, etc.) in the process of confirming the relay pump distribution pattern are optimized, and finally the best relay pump distribution pattern that meets the engineering requirements, has high fluid transportation efficiency and conforms to physical laws is determined.

[0142] Embodiment 3

[0143] To implement the shield machine relay pump layout optimization method as described in Embodiment 1 and Embodiment 2, this embodiment discloses a shield machine relay pump layout optimization system based on fluid, including:

[0144] The first module is configured to construct a fluid control model for an incompressible fluid and solve the fluid control model to obtain the theoretical pressure values at the inlet and outlet of the pipeline;

[0145] The second module is configured to construct a pipeline pressure loss function based on pipeline parameters and fluid characteristics and calculate the total pipeline pressure loss using the pipeline pressure loss function;

[0146] The third module is configured to determine the high-efficiency working range of the water pump, and match the high-efficiency working range of the head and the high-efficiency working range of the flow rate of the relay pump in the high-efficiency working range of the water pump in combination with the total pipeline pressure loss; in the high-efficiency working range of the head, a layout mode based on the transportation distance is adopted for the relay pump, and in the high-efficiency working range of the flow rate, a layout mode based on the transportation flow rate is adopted for the relay pump;

[0147] The fourth module is configured to re-model based on the determined pipeline distribution and the selected relay pumps to obtain a three-dimensional model including the relay pumps; construct a hydrodynamic analysis model for the three-dimensional model, where the hydrodynamic analysis model takes into account the operating characteristics of the relay pumps and their interaction with the fluid in the pipeline, and set corresponding boundary conditions;

[0148] The fifth module is configured to substitute the actual pressure value at the pipeline inlet as a known condition into the hydrodynamic analysis model, and preliminarily verify the selected relay pump distribution pattern through simulation calculations.

[0149] The sixth module is configured to construct a neural network model and incorporate the fluid control model and the hydrodynamic analysis model as constraint conditions into the neural network model;

[0150] The seventh module is configured to use the pipeline parameters, fluid characteristics, and relay pump parameters as the input layer, and the flow state, pressure distribution, and velocity distribution in the pipeline as the output layer; train the neural network model using the theoretical pressure values at the pipeline inlet and outlet and the actual pressure value at the pipeline inlet to obtain an optimized neural network model;

[0151] The eighth module is configured to use the optimized neural network model to simulate and verify the relay pump distribution pattern to obtain a simulation result, and optimize the parameters in the process of confirming the relay pump distribution pattern according to the simulation result.

Claims

1. An optimization method for the layout of relay pumps of a shield machine based on fluid, characterized in that, It includes the following steps: Construct a fluid control model for an incompressible fluid, and solve the fluid control model to obtain the theoretical pressure values at the inlet and outlet of the pipeline; Construct a pipeline pressure loss function based on pipeline parameters and fluid characteristics, and calculate the total pipeline pressure loss by using the pipeline pressure loss function; The expression of the pipeline pressure loss function is as follows: ; where, is the total pressure loss of the pipeline; is the frictional pressure loss: , is the fluid density, is the acceleration due to gravity, is the frictional head loss, which is a function of the pipe length and the internal diameter of the pipe; is the local pressure loss: , is the local head loss; Determine the efficient working range of the water pump, and match the efficient working range of the head and the efficient working range of the flow rate of the relay pump in the efficient working range of the water pump in combination with the total pipeline pressure loss; In the efficient working range of the head, adopt a layout mode based on the conveying distance for the relay pump, and in the efficient working range of the flow rate, adopt a layout mode based on the conveying flow rate for the relay pump; Re-model based on the layout mode of the relay pump to obtain a three-dimensional model, and construct a fluid dynamics analysis model for the three-dimensional model. The fluid dynamics analysis model takes into account the working characteristics of the relay pump and its interaction with the fluid in the pipeline, and sets corresponding boundary conditions; Substitute the actual pressure value at the pipeline inlet as a known condition into the fluid dynamics analysis model, and preliminarily verify the selected distribution mode of the relay pump through simulation calculation; The determination process of the efficient working range of the water pump is as follows: The self-performance of the water pump is expressed by the following formula: , where , and are constants, which are the performance parameters of the water pump and reflect the flow-head characteristic curve of the water pump; is the performance flow of the water pump, is the head corresponding to the self-performance of the water pump. Calculate the flow difference and head difference of the water pump: ; where and are the head and flow rate at the highest efficiency point of the water pump respectively; The boundaries of the high-efficiency working range of the water pump are determined by the following formula: and ; where and are the given head tolerance and flow tolerance respectively; The matching process of the high-efficiency working range of the head of the relay pump is as follows: According to the total pressure loss of the pipeline the corresponding head loss is calculated : ; The actual total head of the water pump is : ; If the actual total head of the water pump meets the judgment conditions of the high-efficiency working area of the head, it means that the corresponding point is the high-efficiency working area of the head of the relay pump, and the layout mode based on the conveying distance should be adopted; The expression form of the judgment condition for the high-efficiency working range of the lift is as follows: ; The matching process of the flow efficient working area of the relay pump is as follows: Considering the influence brought by the total pipeline pressure loss, the actual flow rate of the water pump is calculated using the following formula : ; wherein, is a coefficient reflecting the pipeline resistance situation; if the actual flow rate meets the judgment conditions of the high-efficiency flow working area, it means that the corresponding point is the high-efficiency flow working area of the relay pump, and the layout mode based on the conveying flow rate should be adopted; The expression form of the judgment condition for the efficient working range of the flow rate is: ; wherein, is the given flow tolerance.

2. The optimized method for arranging the relay pumps of a shield machine based on fluid according to claim 1, wherein It further includes the following steps: Construct a neural network model, and incorporate the fluid control model and the fluid dynamics analysis model as constraint conditions into the neural network model; Use the pipeline parameters, fluid characteristics and relay pump parameters as the input layer, and the flow state, pressure distribution and velocity distribution in the pipeline as the output layer; Train the neural network model with the theoretical pressure values at the inlet and outlet of the pipeline and the actual pressure value at the pipeline inlet to obtain an optimized neural network model; Use the optimized neural network model to simulate and verify the distribution mode of the relay pump to obtain a simulation result, and optimize the parameters in the confirmation process of the distribution mode of the relay pump according to the simulation result.

3. An optimization method for arranging relay pumps of a shield machine based on fluid according to claim 1, characterized in that, The expression form of the fluid control model for an incompressible fluid is as follows: In three-dimensional space The mass conservation equation for an incompressible fluid is as follows: ; wherein, , and are the velocity vectors of the fluid in the three axial directions respectively; For an incompressible fluid, the expression form of the momentum conservation equation in each direction is as follows: ; wherein, and represent any direction, represents time, 、 are respectively the components of the velocity in the directions and ; represents pressure, is the fluid density, represents the component of the external force acting on the unit volume of the fluid in the direction ; and are the lengths in the directions and ; , represents the fluid viscosity; For an incompressible fluid, the expression form of the energy conservation equation is as follows: ; where, represents the total energy per unit mass, is the heat conduction flux, represents the dissipation rate caused by viscosity, represents the divergence operation, is the divergence operation on the vector , is the divergence operation on the heat conduction flux, is the velocity vector of the fluid, is the external force per unit volume of the fluid.

4. An optimization method for arranging relay pumps of a shield machine based on fluid according to claim 1, characterized in that, The expression form of the fluid dynamics analysis model is as follows: In three-dimensional space The mass conservation equation of an incompressible fluid is as follows: ; wherein, , and are the velocity vectors of the fluid in the hydrodynamic analysis model in the three axial directions respectively; For an incompressible fluid, the expression form of the momentum conservation equation in each direction is as follows: ； Among them, represents the fluid density, represents the pressure, and are respectively the components of the velocity in the directions and ; and are the lengths in the directions and ; , is the fluid viscosity, represents the component of the external force acting on the unit volume of the fluid in the direction ; is the additional force exerted by the relay pump on the unit volume of the fluid in the direction ; is the pressure increment generated by the relay pump, is the differential length in the direction ; is the Dirac function, represents the coordinate position of any point, is the coordinate position of the relay pump; For an incompressible fluid, the expression form of the energy conservation equation is as follows: ； wherein, represents the total energy per unit mass, is the velocity vector of the fluid, is the heat conduction flux, represents the dissipation rate caused by viscosity, is the divergence operation on the vector , is the divergence operation on the heat conduction flux, is the external force per unit volume of the fluid, is the energy input per unit volume of the fluid by the relay pump; is the rated flow rate of the relay pump, is the differential volume containing the pump.

5. An optimization method for arranging relay pumps of a shield machine based on fluid according to claim 4, characterized in that The boundary conditions at least include: ； In the formula, is the outlet pressure of the relay pump, is the inlet pressure of the relay pump, is the pressure increment generated by the relay pump, is the rated flow rate of the relay pump, is the inlet flow rate of the relay pump, is the outlet flow rate of the relay pump.

6. A fluid-based relay pump layout optimization system for a shield machine, which is used to implement the relay pump layout optimization method for a shield machine described in any one of claims 1 to 5, characterized in that, Include: The first module is configured to construct a fluid control model for an incompressible fluid, and solve the fluid control model to obtain the theoretical pressure values at the inlet and outlet of the pipeline; The second module is configured to construct a pipeline pressure loss function based on pipeline parameters and fluid characteristics, and calculate the total pipeline pressure loss by using the pipeline pressure loss function; The third module is configured to determine the high-efficiency working range of the water pump, and match the high-efficiency working range of the head and the high-efficiency working range of the flow rate of the relay pump in the high-efficiency working range of the water pump in combination with the total pressure loss of the pipeline; in the high-efficiency working range of the head, a layout mode based on the conveying distance is adopted for the relay pump, and in the high-efficiency working range of the flow rate, a layout mode based on the conveying flow rate is adopted for the relay pump; The fourth module is configured to re-model based on the determined pipeline distribution and the selected relay pump to obtain a three-dimensional model including the relay pump; a computational fluid dynamics analysis model is constructed for the three-dimensional model, and the computational fluid dynamics analysis model takes into account the working characteristics of the relay pump and its interaction with the fluid in the pipeline, and corresponding boundary conditions are set; The fifth module is configured to substitute the actual pressure value at the pipeline input port as a known condition into the computational fluid dynamics analysis model, and preliminarily verify the selected relay pump distribution mode through simulation calculation.

Images