An intelligent design method for heat exchangers based on deep reinforcement learning

Abstract

The present invention provides an intelligent design method for a heat exchanger based on deep reinforcement learning, which relates to the technical field of data processing. The method includes: establishing a flow model that simulates the flow path, velocity distribution, and pressure change of the fluid inside the heat exchanger according to the geometric shape of the heat exchanger, the physical properties of the heat exchanger material, and the fluid flowing inside the heat exchanger; establishing a heat transfer model according to the geometric shape of the heat exchanger, the physical properties of the heat exchanger material, and the fluid flowing inside the heat exchanger, in combination with the thermodynamic principle, to calculate the heat transfer efficiency under different fluid flow conditions; coupling the flow model and the heat transfer model to obtain a coupled performance evaluation model for evaluating the overall performance of the heat exchanger under specific working conditions; determining the state space, action space, and reward function according to the flow and heat transfer intensity coupled performance evaluation model. The present invention improves the design efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to an intelligent design method for heat exchangers based on deep reinforcement learning. Background Art

[0002] Traditional heat exchanger design methods mostly rely on empirical formulas and experimental verification, which not only have a long design cycle but also are difficult to optimize performance under complex and variable working conditions. In recent years, with the rapid development of computing technology, design methods based on numerical simulation have gradually become mainstream, but there are still some deficiencies in processing large-scale data and optimization decisions.

[0003] In the field of heat exchanger design, deep reinforcement learning technology has great application potential, especially when dealing with complex systems with multiple variables, non-linearity, and time-varying characteristics. However, in practical applications, deep reinforcement learning methods still face some key challenges in data processing.

[0004] For example, reinforcement learning models require a large amount of sample data during the training process, and in the field of heat exchanger design, high-quality experimental or simulation data is often difficult to obtain, which limits the application scope and performance improvement of reinforcement learning models. Summary of the Invention

[0005] The present invention provides an intelligent design method for heat exchangers based on deep reinforcement learning, which improves the design efficiency and optimizes the performance of heat exchangers.

[0006] To solve the above technical problems, the technical solution of the present invention is as follows:

[0007] An intelligent design method for heat exchangers based on deep reinforcement learning, the method comprising:

[0008] Determine the geometric shape of the heat exchanger, including plate type and shell and tube type, and record the key dimensions, including plate thickness, pipe diameter, and tube pitch;

[0009] Obtain the physical properties of the heat exchanger material, including thermal conductivity, density, and specific heat capacity;

[0010] For the fluid flowing in the heat exchanger, obtain its thermodynamic and fluid mechanical properties, including viscosity, density, specific heat capacity, thermal conductivity, and phase change characteristics;

[0011] According to the geometric shape of the heat exchanger, the physical properties of the heat exchanger material, and the fluid flowing in the heat exchanger, establish a flow model for simulating the flow path, velocity distribution, and pressure change of the fluid inside the heat exchanger;

[0012] According to the geometry of the heat exchanger, the physical properties of the heat exchanger material, and the fluids flowing inside the heat exchanger, combined with the principles of thermodynamics, a heat transfer model is established to calculate the heat exchange efficiency under different fluid flow conditions;

[0013] Couple the flow model and the heat transfer model to obtain a coupled performance evaluation model for evaluating the overall performance of the heat exchanger under specific working conditions;

[0014] Determine the state space, action space, and reward function according to the coupled performance evaluation model of flow and heat transfer intensity;

[0015] According to the state space, action space, and reward function, use the reinforcement learning algorithm to construct a reinforcement learning model, conduct reinforcement learning model training, and obtain the optimal policy network;

[0016] Input the working condition parameters of the design working condition into the policy network to obtain the final structural design parameters.

[0017] Furthermore, according to the geometry of the heat exchanger, the physical properties of the heat exchanger material, and the fluids flowing inside the heat exchanger, establish a flow model for simulating the flow path, velocity distribution, and pressure change of the fluid inside the heat exchanger, including:

[0018] Receive and process the geometric parameter data of the heat exchanger, including geometric dimensions, pipe layout, and fin structure;

[0019] Receive the physical property parameters of the fluid and material, including density, viscosity, and thermal conductivity;

[0020] Construct a virtual representation of the heat exchanger in the simulation environment according to the input geometric shape data;

[0021] Set the initial conditions, such as the initial velocity, pressure, temperature distribution, etc. of the fluid, as the starting point of the simulation;

[0022] Automatically identify and set the boundary conditions of the flow model, including the fluid characteristics at the inlet and outlet, wall conditions, and thermal boundary conditions, to obtain the flow model;

[0023] Select the Navier-Stokes equation and the finite volume solution method according to the simulation requirements, run the flow simulation, and solve the flow equation through iterative calculation to obtain the flow state of the fluid inside the heat exchanger.

[0024] Furthermore, select the Navier-Stokes equation and the finite volume solution method according to the simulation requirements, run the flow simulation, and solve the flow equation through iterative calculation to obtain the flow state of the fluid inside the heat exchanger, including:

[0025] Obtain the geometric model data of the heat exchanger, including its shape, size, and internal structure; set the physical parameters for the simulation, including the density, viscosity, and thermal conductivity of the fluid; define the boundary conditions for the simulation, including the inlet velocity, temperature, outlet pressure, and the thermal boundary conditions of the wall surface;

[0026] Generate a computational grid according to the geometry of the heat exchanger, and the computational grid is used to discretize the Navier-Stokes equations;

[0027] Discretize the Navier-Stokes equations using the finite volume method, converting the continuous partial differential equations into discrete algebraic equations;

[0028] Extract the flow state information of the fluid inside the heat exchanger according to the discrete algebraic equations, including the velocity distribution, pressure distribution, and temperature distribution.

[0029] Furthermore, generate a computational grid according to the geometry of the heat exchanger, and the computational grid is used to discretize the Navier-Stokes equations, including:

[0030] Obtain all the geometric data of the heat exchanger according to its geometry, including the diameter, length, arrangement of the pipes, the size, spacing, thickness of the fins, and the outer and inner walls of the heat exchanger;

[0031] Generate a computational grid according to all the geometric data of the heat exchanger;

[0032] Divide the geometric space inside the heat exchanger into grid cells, and the physical quantities within each grid cell are regarded as constants;

[0033] Convert the Navier-Stokes equations from the continuous differential form to the corresponding numerical form, and the steps include:

[0034] Each control volume contains a certain amount of fluid, and the boundary of the control volume is defined by grid nodes. Integrate the Navier-Stokes equations within each control volume, and the integrated equation is expressed as:

[0035] ;

[0036] where, is the velocity field, is the viscosity of the fluid, is the external force; is the integrated volume region, representing a certain region in a three-dimensional space; is the velocity field with respect to time partial derivative; is the unit normal vector; is the integrated surface region, representing the boundary surface; is the area of the surface element; is the velocity field gradient; is the dot product of the velocity gradient tensor with itself; is the gradient operator, representing the rate of change in space.

[0037] Furthermore, the Navier-Stokes equations are discretized using the finite volume method, converting the continuous partial differential equations into discrete algebraic equations, including:

[0038] The computational domain is divided into multiple control volumes, and the conserved quantities on each control volume are integrated to obtain their discrete algebraic equations; in two-dimensional or three-dimensional space, non-overlapping grids are used to divide the computational domain. In the two-dimensional case, the entire region is divided into multiple rectangular or quadrilateral control volumes; for each control volume, assuming its volume is V i , the face is S i , and there is a normal vector n i on the outer boundary of each control volume, and there is a scalar value at the center of each control volume; assuming the discretization of a certain conserved quantity, within the control volume, the conservation equation has the following form:

[0039] ;

[0040] where, is the conserved quantity, including velocity or temperature or mass; is the mass flux, including momentum flux or energy flux, etc.; is the source term, including body force or heat source;

[0041] For a certain conserved quantity , its time and space integrals are performed, and Gauss's theorem is applied to convert the volume integral of the flux into a surface integral;

[0042] An explicit time discretization method is used for time stepping;

[0043] For the momentum equation of the fluid, the mass flux is expressed as the amount of fluid flowing through the surface;

[0044] The flux is calculated according to the boundary conditions of the flow problem; after discretizing the conservation equations of each control volume, an algebraic system of equations is obtained.

[0045] Furthermore, applying Gauss's theorem to convert the volume integral of the flux into a surface integral specifically includes:

[0046] For a certain conserved quantity , perform time and space integration on it, apply Gauss's theorem, convert the volume integral of the flux into a surface integral, and obtain:

[0047] ;

[0048] where, is the control volume, is the outer boundary surface of the control volume, is the outer normal vector.

[0049] Furthermore, use the explicit time discretization method for time stepping, and its corresponding calculation formula is:

[0050] ;

[0051] where, is the conserved quantity at the current time step moment, is the value at the next time step, is the time step size.

[0052] Furthermore, the calculation formula for the mass flux is:

[0053] Assume that the mass flux is calculated, where and represent the density and velocity of the fluid respectively; use the upwind differencing method to calculate the mass flux on the boundary, and obtain the following approximation:

[0054] ;

[0055] where, is the mass flux through the surface of the control volume ; is the density inside the control volume ; is the velocity vector inside the control volume ; is the density inside the control volume ; is the velocity vector inside the control volume .

[0056] Furthermore, the calculation process of the algebraic equation system includes:

[0057] Assume that for a certain conserved quantity , its discretized form inside the control volume is:

[0058] ;

[0059] where, is the value at time step at time is the flux, representing the flow rate through the control volume and its neighborhood of the flow is the source term, representing the volumetric source term within the control volume ; the algebraic equations are solved by an explicit method to obtain the conserved quantities for all control volumes over the entire computational domain; is the value at time step at time

[0060] Furthermore, the flow model and the heat transfer model are coupled to obtain a coupled performance evaluation model for evaluating the overall performance of the heat exchanger under specific operating conditions, including:

[0061] Establish a data exchange interface so that the flow model and the heat transfer model can identify and receive the data output by each other;

[0062] Determine the common parameters in the flow model and the heat transfer model, including the temperature, pressure, and velocity of the fluid;

[0063] Integrate the flow model and the heat transfer model into a simulation environment so that the flow model and the heat transfer model can operate on the same time and space scales;

[0064] Since the flow and heat transfer processes affect each other, during the coupling process, the final solution is found through iterative solution. In each iterative step, the flow model calculates the flow state of the fluid and transfers the flow state to the heat transfer model; the heat transfer model calculates the heat exchange situation based on the flow state and feeds the result back to the flow model, repeating until the predetermined number of iterations is reached to obtain the final coupled performance evaluation model.

[0065] The above solution of the present invention has at least the following beneficial effects:

[0066] Through an automated and intelligent design process, the present invention can significantly reduce the manual calculation and trial-and-error time required in traditional design methods, thereby greatly improving the design efficiency of the heat exchanger. The deep reinforcement learning model can automatically explore and optimize the design parameters of the heat exchanger during the training process to maximize the reward function, thereby ensuring that the designed heat exchanger has excellent overall performance under specific operating conditions. Due to the strong self-learning and adaptive capabilities of the reinforcement learning model, the present invention can easily handle various complex and changing operating conditions and provide a flexible and efficient solution for heat exchanger design. Description of the Drawings

[0067] Figure 1 is a schematic flow chart of the intelligent design method of a heat exchanger based on deep reinforcement learning provided by an embodiment of the present invention. Detailed Embodiment

[0068] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0069] As Figure 1 shown, an embodiment of the present invention provides an intelligent design method for a heat exchanger based on deep reinforcement learning, and the method includes:

[0070] Determine the geometry of the heat exchanger, including plate type and shell-and-tube type, and record the key dimensions, including plate thickness, pipe diameter, and tube pitch;

[0071] Obtain the physical properties of the heat exchanger material, including thermal conductivity, density, and specific heat capacity;

[0072] For the fluid flowing in the heat exchanger, obtain its thermodynamic and fluid mechanical properties, including viscosity, density, specific heat capacity, thermal conductivity, and phase change characteristics;

[0073] According to the geometry of the heat exchanger, the physical properties of the heat exchanger material, and the fluid flowing in the heat exchanger, establish a flow model that simulates the flow path, velocity distribution, and pressure change of the fluid inside the heat exchanger;

[0074] According to the geometry of the heat exchanger, the physical properties of the heat exchanger material, and the fluid flowing in the heat exchanger, combine the thermodynamic principles to establish a heat transfer model to calculate the heat transfer efficiency under different fluid flow conditions;

[0075] Couple the flow model and the heat transfer model to obtain a coupled performance evaluation model for evaluating the overall performance of the heat exchanger under specific working conditions;

[0076] According to the flow and heat transfer intensity coupled performance evaluation model, determine the state space, action space, and reward function;

[0077] According to the state space, action space, and reward function, use a reinforcement learning algorithm to construct a reinforcement learning model, perform reinforcement learning model training, and obtain an optimal policy network;

[0078] Input the working condition parameters of the design working condition into the policy network to obtain the final structural design parameters.

[0079] In the embodiments of the present invention, traditional heat exchanger design methods usually require a large amount of manual calculations and experimental verifications, which are time-consuming and inefficient. However, by introducing deep reinforcement learning technology, the present invention realizes the automation and intelligence of the design process, thus greatly improving the design efficiency and shortening the product development cycle. During the training process, the deep reinforcement learning model can continuously try and learn to optimize, automatically searching for the best combination of design parameters. This enables the present invention to design heat exchangers with better performance under specific working conditions, improving the heat exchange efficiency and overall performance of the products. The reinforcement learning model constructed by the present invention has strong adaptive capabilities and can automatically adjust the design strategy according to different working conditions. This enables the method to flexibly respond to various complex and changing design requirements, enhancing the market competitiveness of the products. By reducing the number of physical tests and prototype productions, as well as optimizing material selection and structural design, the present invention can effectively reduce production costs while ensuring design quality.

[0080] Determine the geometry of the heat exchanger, including plate type and shell-and-tube type, and record the key dimensions, including plate thickness, pipe diameter, and tube pitch; Obtain the physical properties of the heat exchanger materials, including thermal conductivity, density, and specific heat capacity; For the fluids flowing inside the heat exchanger, obtain their thermodynamic and fluid mechanical properties, including viscosity, density, specific heat capacity, thermal conductivity, and phase change characteristics, specifically including:

[0081] Suppose we choose to design a shell-and-tube heat exchanger. The key dimensions include:

[0082] Pipe diameter: For example, select a pipe diameter of 25 mm, which is a common size applicable to various fluids and heat exchange scenarios.

[0083] Tube pitch: Set it to 30 mm to ensure sufficient fluid channels and heat exchange area.

[0084] Plate thickness (for plate heat exchangers): If designing a plate heat exchanger, the plate thickness may be selected as 0.5 mm to provide good heat conduction performance and structural strength.

[0085] Suppose we choose carbon steel as the main material of the heat exchanger. Its physical properties may include:

[0086] Thermal conductivity: The thermal conductivity of carbon steel is usually between 45 - 50 W / (m·K), indicating its good heat conduction ability. Density: The density of carbon steel is approximately 7850 kg / m³, which is necessary for calculating the weight and structural strength of the heat exchanger. Specific heat capacity: The specific heat capacity of carbon steel is about 0.46 kJ / (kg·K), which helps to understand the temperature change of the material during heat absorption or heat release. Obtain the thermodynamic and fluid mechanical properties of the fluid. In this step, collect the relevant properties of the fluid flowing inside the heat exchanger.

[0087] Further, based on the geometry of the heat exchanger, the physical properties of the heat exchanger material, and the fluid flowing inside the heat exchanger, a flow model is established to simulate the flow path, velocity distribution, and pressure change of the fluid inside the heat exchanger, including:

[0088] Receiving and processing the geometric parameter data of the heat exchanger, including geometric dimensions, pipe layout, and fin structure;

[0089] Receiving the physical property parameters of the fluid and the material, including density, viscosity, and thermal conductivity;

[0090] Constructing a virtual representation of the heat exchanger in a simulation environment according to the input geometric shape data;

[0091] Setting initial conditions, such as the initial velocity, pressure, and temperature distribution of the fluid, as the starting point of the simulation;

[0092] Automatically identifying and setting the boundary conditions of the flow model, including the fluid characteristics at the inlet and outlet, wall conditions, and thermal boundary conditions, to obtain the flow model;

[0093] Selecting the Navier-Stokes equation and the finite volume method according to the simulation requirements, running the flow simulation, and solving the flow equation through iterative calculation to obtain the flow state of the fluid inside the heat exchanger.

[0094] In the embodiments of the present invention, by establishing a refined flow model, designers can more accurately predict the flow behavior of the fluid inside the heat exchanger, including the flow path, velocity distribution, and pressure change. This helps to discover and correct potential problems during the design stage, thereby ensuring that the performance of the heat exchanger meets expectations during actual operation. The flow model can reveal the dynamic characteristics of the fluid in the heat exchanger, such as eddy currents, dead zones, and backflows. By identifying these adverse flow patterns, designers can specifically adjust the geometry or parameters of the heat exchanger to optimize the fluid flow state and improve the heat transfer efficiency. Traditional heat exchanger designs often require a large number of physical tests to verify performance. By establishing an accurate flow model, the operating state of the heat exchanger can be simulated on a computer, thereby reducing the dependence on physical tests. This not only reduces the test cost but also shortens the product development cycle. By simulating and evaluating the performance of the heat exchanger under various working conditions through the flow model, potential failure modes or performance bottlenecks can be discovered at an early stage of design. This helps to take improvement measures in a timely manner and enhance the reliability and durability of the product. The flow model can quickly respond to design changes. When it is necessary to adjust the geometry, material, or fluid parameters of the heat exchanger, the model can quickly recalculate and display the new flow state.

[0095] Receiving and processing the geometric parameter data of the heat exchanger, and the specific implementation process is as follows:

[0096] Receive detailed data on the geometric dimensions of the heat exchanger (such as length, width, height, pipe diameter, etc.), pipe layout (such as pipe spacing, arrangement method, etc.), and fin structure (such as fin thickness, spacing, etc.) through the user interface or data file; check the integrity and rationality of the input data to ensure that all data is within the valid range and meets the design requirements of the heat exchanger; perform necessary unit conversions and format adjustments on the input data; save the processed geometric parameter data in a suitable data structure, such as an array, list, or object, for easy access during the simulation process.

[0097] Receive the physical property parameters of the fluid and materials, and the specific implementation process is as follows:

[0098] Receive the physical property parameters of the fluid (such as water, oil, etc.) and the heat exchanger material (such as carbon steel, stainless steel, etc.), including density, viscosity, thermal conductivity, etc. through the user interface or database. Verify whether the input physical property parameters are accurate and ensure that all parameters use a unified unit system. Store the verified physical property parameters in a data structure for quick access to these values during the simulation calculation.

[0099] Construct a virtual representation of the heat exchanger in the simulation environment according to the input geometric shape data, and the specific implementation process is as follows:

[0100] Use computer graphics or computational geometry techniques to construct a three-dimensional model or two-dimensional grid representation of the heat exchanger in the simulation environment according to the input geometric dimensions and shape data; if numerical simulation methods such as the finite volume method (FVM) are used, a computational grid needs to be generated. This can be done through an automatic grid generation algorithm or professional software to ensure the quality and adaptability of the grid. Check whether the constructed virtual representation accurately reflects the geometric characteristics of the actual heat exchanger and make model corrections or optimizations if necessary.

[0101] Set parameters such as the initial velocity, pressure, and temperature distribution of the fluid according to the design requirements or actual operating scenarios. These parameters can be used as the starting point for the simulation calculation. Verify whether the input initial conditions are reasonable and adjust them as needed to ensure the accuracy and effectiveness of the simulation. Save the verified initial conditions in the simulation environment for loading these values at the start of the simulation.

[0102] Automatically identify and set the boundary conditions of the flow model, and the specific implementation process is as follows:

[0103] Based on the geometric model and flow characteristics of the heat exchanger, automatically identify the positions where boundary conditions need to be set, such as inlets, outlets, walls, etc. According to the identification results, set the corresponding fluid properties (such as velocity, pressure, temperature, etc.), wall conditions (such as no-slip, partial-slip, etc.), and thermal boundary conditions (such as constant temperature, constant heat flux density, etc.) for each boundary position. Check whether the set boundary conditions conform to physical reality and simulation requirements, and make adjustments and optimizations based on the feedback. Apply the verified boundary conditions to the flow model to ensure that the simulation process can correctly reflect the real flow of the fluid inside the heat exchanger.

[0104] Furthermore, select the Navier-Stokes equations and the finite volume method according to the simulation requirements, run the flow simulation, and solve the flow equations through iterative calculations to obtain the flow state of the fluid inside the heat exchanger, including:

[0105] Obtain the geometric model data of the heat exchanger, including its shape, size, and internal structure; set the physical parameters of the simulation, including the density, viscosity, and thermal conductivity of the fluid; define the boundary conditions of the simulation, including the inlet flow velocity, temperature, outlet pressure, and the thermal boundary conditions of the wall;

[0106] Generate a computational grid according to the geometric shape of the heat exchanger, and the computational grid is used to discretize the Navier-Stokes equations;

[0107] Discretize the Navier-Stokes equations using the finite volume method, and transform the continuous partial differential equations into discrete algebraic equations;

[0108] Extract the flow state information of the fluid inside the heat exchanger according to the discrete algebraic equations, including the velocity distribution, pressure distribution, and temperature distribution.

[0109] In the embodiments of the present invention, by using the Navier-Stokes equations and the finite volume solution method, the flow behavior of fluids in complex geometries can be accurately simulated, so as to accurately predict the performance of heat exchangers at the design stage. The simulation results provide detailed information on the fluid flow state, including velocity distribution, pressure distribution, and temperature distribution. This information helps to identify adverse regions in the flow, such as dead zones, vortices, or high-pressure drop regions, thereby guiding designers to optimize the heat exchanger structure to improve fluid flow and enhance heat exchange efficiency. Traditional heat exchanger designs usually require a large number of physical tests to verify performance. By means of numerical simulation methods, these tests can be virtually carried out on a computer, thus significantly reducing the number of physical tests and related costs. This not only speeds up the product development cycle but also reduces the overall development cost. The numerical simulation technology allows designers to explore a variety of different design schemes within a short period of time and evaluate their performance. This provides a powerful tool for product innovation, enabling designers to customize high-performance heat exchanger products according to customer requirements or specific application scenarios.

[0110] Further, according to the geometry of the heat exchanger, a computational grid is generated, and the computational grid is used to discretize the Navier-Stokes equations, including:

[0111] All geometric data of the heat exchanger are obtained according to its geometry, including the diameter, length, arrangement of pipes, dimensions, spacing, thickness of fins, and the outer and inner walls of the heat exchanger; a computational grid is generated according to all geometric data of the heat exchanger; the geometric space inside the heat exchanger is divided into grid cells, and the physical quantities within each grid cell are regarded as constants, specifically including:

[0112] Use measuring tools (such as calipers, laser rangefinders, etc.) or design drawings to obtain the detailed geometric data of the heat exchanger; record the diameter, length, and arrangement pattern (such as parallel, staggered, etc.) of the pipes; measure the dimensions (such as length, width), spacing, and thickness of the fins; determine the dimensions and shapes of the outer and inner walls of the heat exchanger. Organize the collected data into a standard format, such as a spreadsheet or a CAD file, to ensure the accuracy and integrity of the data; according to the complexity of the heat exchanger and the calculation requirements, select a suitable mesh generation tool, such as a professional CFD (Computational Fluid Dynamics) preprocessing software; import the organized geometric data into the mesh generation tool, and if necessary, perform necessary simplification or repair on the geometric model to ensure the smooth progress of mesh generation; define the density, type, and distribution pattern of the mesh to adapt to the geometric characteristics and flow characteristics of the heat exchanger, and set finer meshes in key areas (such as areas with complex fluid flow or high heat transfer efficiency) to improve the accuracy of the simulation; run the mesh generation tool to automatically generate the computational mesh according to the set parameters, check the quality of the generated mesh, and ensure that there are no deformed or overly distorted mesh elements to guarantee the stability and accuracy of the calculation; in the generated mesh, each mesh element represents a control volume, which is the basic computational unit in fluid dynamics simulation, and the boundary of the control volume is defined by the nodes and edges of the mesh; within each control volume, discretize the continuous physical quantities (such as velocity, pressure, temperature, etc.) into constants representing the average values within that volume. This discretization method enables complex partial differential equations to be transformed into algebraic equations that are easier to solve, and store the physical quantity data of each mesh element in a suitable data structure for efficient access and update during the simulation process.

[0113] Transform the Navier-Stokes equations from the continuous differential form into the corresponding numerical form, and the steps include:

[0114] Each control volume contains a certain amount of fluid, and the boundary of the control volume is defined by the mesh nodes. Integrate the Navier-Stokes equations within each control volume, and the integrated equation is expressed as:

[0115] ;

[0116] Among them, is the velocity field, is the viscosity of the fluid, is the external force; is the volume region of integration, representing a certain region in three-dimensional space; is the velocity field with respect to time partial derivative; is the unit normal vector; is the surface region of integration, representing the boundary surface; is the area of the surface element; is the velocity field gradient; is the dot product of the velocity gradient tensor with itself; is the gradient operator, representing the rate of change in space.

[0117] In the embodiments of the present invention, by obtaining in detail all the geometric data of the heat exchanger, including the diameter, length, and arrangement of the pipes, the size, spacing, and thickness of the fins, as well as the outer and inner walls of the heat exchanger, etc., a highly accurate geometric model of the heat exchanger can be established. This makes it possible to perform flow and heat transfer simulations on heat exchangers with complex internal structures, thereby more accurately predicting their performance. The generated computational grid can reasonably divide the geometric space inside the heat exchanger into grid cells, and the physical quantities within each grid cell are regarded as constants. This discretization method not only simplifies complex flow problems but also makes the numerical solution process more efficient while optimizing the utilization of computing resources. By integrating the Navier-Stokes equations within each control volume, the continuous fluid dynamics problem can be accurately transformed into a discrete numerical problem. This transformation preserves the main physical characteristics of the original equations, such as the inertia, viscosity of the fluid, and the influence of external forces, thus being able to more realistically reflect the flow state of the fluid inside the heat exchanger. With the help of accurate numerical simulation results, designers can deeply understand the flow details of the fluid in the heat exchanger, such as velocity distribution, pressure change, and heat transfer efficiency, etc. Traditional heat exchanger design often requires multiple physical experiments to verify performance. However, through numerical simulation methods, these experiments can be virtually carried out on a computer, thereby significantly shortening the R & D cycle and reducing the experimental cost. At the same time, numerical simulation also allows for comparison and optimization of multiple schemes during the design stage, further improving the efficiency and accuracy of the design.

[0118] Furthermore, the finite volume method is used to discretize the Navier-Stokes equations, transforming the continuous partial differential equations into discrete algebraic equations, including:

[0119] The computational domain is divided into multiple control volumes, and the conserved quantities on each control volume are integrated to obtain their discrete algebraic equations; in two-dimensional or three-dimensional space, non-overlapping grids are used to divide the computational domain. In the two-dimensional case, the entire region is divided into multiple rectangular or quadrilateral control volumes; for each control volume, assuming its volume is V i , the surface is S i , and there is a normal vector n i, and there is a scalar value at the center of each control volume; assuming the discretization of a certain conserved quantity, within the control volume, the conservation equation has the following form:

[0120] ;

[0121] where, is the conserved quantity, including velocity or temperature or mass; is the mass flux, including momentum flux or energy flux, etc.; is the source term, including body force or heat source;

[0122] For a certain conserved quantity , integrating it in time and space, applying Gauss's theorem, and converting the volume integral of the flux into a surface integral;

[0123] Using an explicit time discretization method for time stepping;

[0124] For the momentum equation of the fluid, the mass flux represents the amount of fluid flowing through the surface;

[0125] Calculating the flux according to the boundary conditions of the flow problem; after discretizing the conservation equation of each control volume, an algebraic equation system is obtained.

[0126] In the embodiment of the present invention, the finite volume method is based on the conservation principle, integrates the physical quantities within the control volume, and ensures the numerical stability and convergence during the discretization process. This means that during the solution process, the numerical solution can gradually approach the true solution, reducing the calculation error and uncertainty. Since the finite volume method strictly satisfies the conservation equation on each control volume, it can accurately capture the local changes during the fluid flow process, such as velocity gradients, pressure mutations, etc. This guarantee of local conservation makes the simulation results more realistic and reliable. The finite volume method is applicable to grids of various shapes, including structured grids and unstructured grids. This makes the method highly flexible and versatile when dealing with heat exchangers with complex geometries. At the same time, it can easily handle different boundary conditions and physical parameter settings. By discretizing the Navier-Stokes equation and converting the continuous partial differential equation into a discrete algebraic equation, the solution difficulty and computational complexity can be significantly reduced. This means that with the same computing resources, the finite volume method can give simulation results faster, improving the computational efficiency.

[0127] Furthermore, applying Gauss's theorem to convert the volume integral of the flux into a surface integral specifically includes:

[0128] For a certain conserved quantity , integrating it in time and space, applying Gauss's theorem, and converting the volume integral of the flux into a surface integral, obtaining:

[0129] ;

[0130] wherein, is the control volume, is the outer boundary surface of the control volume, is the outer normal vector.

[0131] In the embodiments of the present invention, by converting the volume integral into a surface integral, the complexity and computational amount in the calculation can be significantly reduced. In numerical simulations, the integration of the surface area is usually easier to implement than the integration of the volume because the dimension of the surface area is lower than that of the volume, thus simplifying the process of numerical integration. The calculation of the surface integral usually involves fewer grid points and computational elements, so compared with the volume integral, the calculation speed is faster and the memory occupancy is lower. This means that under the same computational resources, larger-scale or more refined simulation problems can be processed, thereby improving the overall computational efficiency. Gauss's theorem is one of the basic principles in physics, which ensures that the conservation properties (such as mass, momentum, energy, etc.) of the fluid within the control volume are strictly adhered to during the discretization process. By applying Gauss's theorem, it can be ensured that the results of the numerical simulation are physically reasonable and reliable. Converting the continuous partial differential equation into a discrete algebraic equation based on the control volume, combined with the application of Gauss's theorem, helps to enhance the stability of the numerical solution.

[0132] Furthermore, using the explicit time discretization method for time stepping, its corresponding calculation formula is:

[0133] ;

[0134] wherein, is the conserved quantity at the current time step at time is the value at the next time step, is the time step size. The mass flux has the following calculation formula:

[0135] Assume that the mass flux is calculated, wherein, and respectively represent the density and velocity of the fluid; the upwind difference method is used to calculate the mass flux at the boundary, and the following approximation is obtained:

[0136] ;

[0137] wherein, is the mass flux through the surface of the control volume is the density within the control volume ; is the control volume The velocity vector within is the control volume the density within; is the control volume the velocity vector within.

[0138] In an embodiment of the present invention, the explicit time discretization method allows the direct calculation of the solution at the next time step at each time step, without the need to solve a complex linear system. This method is simple and fast in calculation. Especially when dealing with large-scale problems, it can significantly improve the calculation efficiency. The algorithm structure of the explicit method is relatively simple and easy to be programmed and implemented. In addition, since the calculation at each time step does not depend on the results of other time steps, the explicit method is easier to be parallelized, thereby further accelerating the calculation process. Compared with the implicit method, the explicit time discretization method usually does not need to store a large amount of coefficient matrices or intermediate variables, thus reducing the memory occupation. This enables the handling of larger-scale problems in a computing environment with limited resources. At an appropriate time step, the explicit method can provide a stable numerical solution. By reasonably selecting the time step Δt, the numerical stability during the simulation process can be ensured, and the divergence or non-physical behavior of the solution can be avoided. The upwind difference method, as a boundary treatment method, can flexibly handle the mass flux of the fluid at the boundary. It selects the physical quantity at the upstream according to the flow direction of the fluid to calculate the flux, thereby more accurately simulating the behavior of the fluid at the boundary.

[0139] Furthermore, the calculation process of the algebraic equation set includes:

[0140] Assume that for a certain conserved quantity , its discretized form within the control volume is:

[0141] ;

[0142] wherein, is the value at time step , is the flux, representing the flow rate through the control volume and its neighborhood , is the source term, representing the volume source term within the control volume ; Solve the algebraic equation set by the explicit method to obtain the conserved quantity of all control volumes on the entire calculation domain; is the value at time step .

[0143] In the embodiments of the present invention, the explicit method allows for the direct calculation of the solution at the next time step without the need to solve complex linear systems or iterative processes, thus significantly improving the computational efficiency. Since the explicit method does not require storing large coefficient matrices or intermediate variables, it occupies less memory, which is particularly important for large-scale computations or situations with limited resources. The algorithmic structure of the explicit method is relatively simple and easy to implement programmatically. Additionally, the calculation at each time step is independent, which makes parallelization easy and can further accelerate the computational process. The explicit method has high flexibility in dealing with complex boundary conditions and source terms and can more easily adapt to different physical models and scenarios. At appropriate time steps, the explicit method can provide stable numerical solutions. By reasonably choosing the time step, numerical stability during the simulation can be ensured, avoiding divergence or non-physical behavior of the solution.

[0144] According to the geometry of the heat exchanger, the physical properties of the heat exchanger material, and the fluids flowing within the heat exchanger, and in combination with the principles of thermodynamics, a heat transfer model is established to calculate the heat transfer efficiency under different fluid flow conditions, specifically including:

[0145] The heat exchanger consists of two mutually isolated channels for the flow of two fluids respectively (for example, a hot fluid and a cold fluid); the fluids flow in a one-dimensional direction within their respective channels; the wall surface of the heat exchanger is uniform; the physical properties of the fluids (such as specific heat capacity, density, etc.) remain constant within their respective operating temperature ranges, and the logarithmic mean temperature difference (LMTD) method is used to estimate the temperature difference between the fluids.

[0146] For each fluid, calculate its heat capacity rate, which will be used to determine the amount of heat that the fluid can absorb or release. Use the LMTD method to calculate the average temperature difference between the two fluids. LMTD takes into account the temperature changes of the fluids at the inlet and outlet of the heat exchanger and provides a single value representing the temperature difference throughout the heat exchanger.

[0147] Use the thermal conductivity of the heat exchanger material, the surface area of the heat exchanger (calculated based on geometric dimensions), and the LMTD to calculate the total heat transfer amount. The heat transfer efficiency can be defined as the ratio of the actual heat transferred to the maximum possible heat that could be transferred theoretically. In this model, the actual heat transferred can be compared with the maximum possible heat calculated based on the fluid inlet temperature difference and the minimum heat capacity rate, and the calculated total heat transfer amount and heat transfer efficiency are output.

[0148] Furthermore, the flow model and the heat transfer model are coupled to obtain a coupled performance evaluation model for evaluating the overall performance of the heat exchanger under specific operating conditions, including:

[0149] Establish a data exchange interface so that the flow model and the heat transfer model can identify and receive the data output by each other;

[0150] Determine the common parameters in the flow model and the heat transfer model, including the temperature, pressure, and velocity of the fluid;

[0151] Integrate the flow model and the heat transfer model into a simulation environment so that the flow model and the heat transfer model operate on the same time and space scales;

[0152] Since the flow and heat transfer processes affect each other, during the coupling process, the final solution is found through iterative solving. In each iterative step, the flow model calculates the flow state of the fluid and transfers the flow state to the heat transfer model; the heat transfer model calculates the heat exchange situation based on the flow state and feeds back the results to the flow model, and repeats until the predetermined number of iterations is reached to obtain the final coupled performance evaluation model.

[0153] In the embodiments of the present invention, by coupling the flow model and the heat transfer model, the actual physical process inside the heat exchanger can be simulated more accurately. This method that comprehensively considers the mutual influence of flow and heat exchange can more precisely predict the performance of the heat exchanger, including heat exchange efficiency, fluid flow characteristics, etc. The coupled performance evaluation model can provide a comprehensive performance analysis for the design of the heat exchanger. Designers can optimize the structure, material selection, fluid flow path, etc. of the heat exchanger according to the simulation results to improve the overall performance and reduce costs. Traditional heat exchanger performance evaluations usually rely on a large number of experimental tests. The coupled performance evaluation model can simulate the performance of the heat exchanger under various working conditions on a computer, thereby reducing the need for physical experiments, saving time, resources, and costs. By quickly simulating and evaluating the performance of different design schemes, enterprises can find the best solution faster, accelerate the new product development process, and improve market competitiveness.

[0154] The above steps are implemented through the following steps:

[0155] Define the data format and determine the data format transmitted between the flow model and the heat transfer model. This includes determining the type of data (such as floating-point numbers, vectors, matrices, etc.), units (such as meters, seconds, degrees Celsius, etc.), and the structure of the data (such as arrays, objects, etc.). In the flow model and the heat transfer model, create interface functions for sending and receiving data respectively. These functions should be able to handle data format conversion, error checking, and synchronous or asynchronous transmission of data.

[0156] Implement data transmission between the flow model and the heat transfer model through methods such as file exchange, memory sharing, network communication (such as using the TCP / IP protocol). The choice of which method depends on the complexity of the model, the distribution of computing resources, and the real-time requirements of the simulation. Before integration, conduct unit tests on the interface to ensure that data can be correctly transmitted between the two models and there is no data loss or format error.

[0157] Analyze the inputs and outputs of the flow model and the heat transfer model, identify the common parameters such as the temperature, pressure, velocity of the fluid, etc. Define clear physical meanings, units, and value ranges for each common parameter. Ensure that these definitions are consistent in the two models. In the data exchange interface, establish the mapping relationship of the common parameters between the flow model and the heat transfer model. In this way, when one model updates a certain parameter, the other model can immediately sense this change. Design a mechanism to ensure that during the simulation process, the common parameters are synchronized between the two models, which requires data exchange and update at the end of each simulation time step.

[0158] Select a simulation platform that can support the integration of the flow model and the heat transfer model, such as ANSYS Workbench, COMSOL Multiphysics, or other professional CFD (Computational Fluid Dynamics) and heat transfer simulation software. Import the established flow model and heat transfer model into the simulation platform. This may require converting the models into a specific format supported by the simulation platform. In the simulation platform, set the simulation time step, spatial resolution, and other relevant simulation parameters to ensure that the flow model and the heat transfer model operate on the same time and space scales. Use the tools or scripts provided by the simulation platform to connect the flow model and the heat transfer model to form an integrated coupled model. Ensure that the data exchange interface works properly and that the two models can recognize and receive each other's data. Set the initial conditions for the flow model and the heat transfer model, such as the initial temperature, pressure, velocity distribution of the fluid, etc. In each iteration step, first run the flow model to calculate the flow state of the fluid, including the velocity field, pressure field, etc. Transfer the flow state data calculated by the flow model to the heat transfer model through the data exchange interface.

[0159] Based on the received flow state data, the heat transfer model calculates the heat exchange situation, including the temperature distribution, heat flux, etc. Feed back the heat exchange data calculated by the heat transfer model to the flow model and update the boundary conditions or physical properties parameters (such as temperature-dependent fluid viscosity) of the flow model. Check whether the predetermined number of iterations is reached or the convergence condition is satisfied. If the condition is met, stop the iteration; otherwise, return to continue the iteration. After the iteration is completed, output the final coupled performance evaluation results, including the overall heat exchange efficiency of the heat exchanger, fluid flow characteristics, and other performance indicators of concern.

[0160] The following is the specific implementation process for each step:

[0161] The implementation process of determining the state space, action space, and reward function is as follows:

[0162] Define the state space, which is the set of all possible values that describe the current state of the system; in the coupled performance evaluation model of flow heat transfer intensity, the state space can include the real-time values of key physical quantities such as the temperature, pressure, and velocity of the fluid, as well as the structural parameters of the heat exchanger (such as pipe diameter, fin pitch, etc.). These state variables will be used to describe the complete state of the heat exchanger at a certain moment.

[0163] Define the action space, which is the set of all possible actions that the agent can take. In this scenario, the action space can be defined as the adjustment amount of the heat exchanger structure design parameters, such as changing the pipe diameter, adjusting the fin pitch, etc. Each action corresponds to a specific structural adjustment aimed at optimizing the performance of the heat exchanger.

[0164] Define the reward function, which is used to evaluate the performance of the agent after taking a certain action in a certain state. In the coupled performance evaluation model of flow heat transfer intensity, the reward function can be defined based on performance indicators such as heat transfer efficiency, pressure drop, and fluid flow stability. For example, if a certain action leads to an increase in heat transfer efficiency, a positive reward is given; conversely, if the performance decreases, a negative reward is given.

[0165] Use the reinforcement learning algorithm to construct a reinforcement learning model, and the implementation process is as follows:

[0166] Select a suitable reinforcement learning algorithm according to the characteristics of the problem (such as the dimensions of the state space and action space, whether they are continuous, etc.), such as Q-learning, Deep Q-Network (DQN), Proximal Policy Optimization (PPO), etc. Use a deep learning framework (such as TensorFlow or PyTorch) to construct a neural network model, which will be used to approximate the value function or policy function; for the value function method (such as DQN), the network outputs the state or action-state value; for the policy gradient method (such as PPO), the network outputs the action probability distribution; randomly initialize the weights and biases of the neural network, or use a pre-trained model for initialization.

[0167] Train the reinforcement learning model in an interactive environment to obtain the optimal policy network, and the implementation process is as follows:

[0168] Create a simulator or an actual system as an interaction environment to simulate the operation of the heat exchanger under various working conditions; the environment should be able to provide status information and return new status and rewards according to the actions of the agent; in each training iteration, the agent selects an action according to the current policy and applies it to the environment; the environment executes the action and returns new status and rewards, and the agent uses the new status and rewards to update its policy network or value function network. Repeat this process until the stopping condition is met (such as reaching the maximum number of iterations, the policy performance is stable, etc.). During the training process, regularly evaluate the performance of the policy and record the policy network parameters with the best performance. After the training is completed, save the policy network that reaches the best performance for subsequent use.

[0169] Input the working condition parameters of the design condition into the policy network to obtain the final structural design parameters. The implementation process is as follows:

[0170] According to the actual application scenario, determine the design condition parameters of the heat exchanger, such as the inlet fluid temperature, flow rate, target heat transfer efficiency, etc.; input the design condition parameters as the state into the trained optimal policy network; the policy network will output the adjusted amount of the optimized structural design parameters under the current working condition, apply these adjusted amounts to the initial structure design to obtain the final structural design parameters, and use simulation or experimental methods to verify the performance of the heat exchanger according to the structural design parameters output by the policy network.

[0171] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art of this technology, without departing from the principle described in the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A heat exchanger intelligent design method based on deep reinforcement learning, characterized in that: The method comprises: Determine heat exchanger geometry, including plate and shell and tube, and record key dimensions, including plate thickness, tube diameter, and tube spacing; Obtain the physical properties of heat exchanger materials, including thermal conductivity, density, and specific heat capacity; For the fluid flowing in the heat exchanger, obtain its thermodynamic and fluid mechanics properties, including viscosity, density, specific heat capacity, thermal conductivity and phase change characteristics; According to the geometric shape of the heat exchanger, the physical properties of the heat exchanger material and the fluid flowing in the heat exchanger, a flow model is established to simulate the flow path, velocity distribution and pressure change of the fluid inside the heat exchanger; According to the geometric shape of the heat exchanger, the physical properties of the heat exchanger material and the fluid flowing in the heat exchanger, combined with the principles of thermodynamics, a heat exchange model is established to calculate the heat exchange efficiency under different fluid flow conditions; The flow model and the heat transfer model are coupled to obtain a coupled performance evaluation model for evaluating the overall performance of the heat exchanger under specific operating conditions; Determine the state space, action space and reward function according to the flow heat transfer intensity coupling performance evaluation model; According to the state space, action space and reward function, the reinforcement learning algorithm is used to construct a reinforcement learning model, and the reinforcement learning model is trained to obtain the optimal strategy network; the working condition parameters of the design condition are input into the strategy network to obtain the final structural design parameters.

2. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 1 is characterized in that: According to the geometric shape of the heat exchanger, the physical properties of the heat exchanger material and the fluid flowing in the heat exchanger, a flow model is established to simulate the flow path, velocity distribution and pressure change of the fluid inside the heat exchanger, including: Receive and process the geometric and positional parameter data of the heat exchanger, including geometric dimensions, pipe layout and fin structure; Receive physical properties of fluids and materials, including density, viscosity, and thermal conductivity; Based on the input geometry data, a virtual representation of the heat exchanger is constructed in a simulation environment; Set initial conditions, such as initial velocity, pressure, and temperature distribution of the fluid, as the starting point of the simulation; Automatically identify and set the boundary conditions of the flow model, including the fluid properties of the inlet and outlet, the wall conditions and the thermal boundary conditions, to obtain the flow model; According to the simulation requirements, the Navier-Stokes equations and the finite volume solution method are selected to run the flow simulation. The flow equations are solved through iterative calculations to obtain the flow state of the fluid inside the heat exchanger.

3. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 2 is characterized in that: Select the Navier-Stokes equations and the finite volume solver according to the simulation requirements, run the flow simulation, solve the flow equations through iterative calculations, and obtain the flow state of the fluid inside the heat exchanger, including: Obtain the geometric model data of the heat exchanger, including its shape, size and internal structure; set the physical parameters of the simulation, including the density, viscosity and thermal conductivity of the fluid; define the boundary conditions of the simulation, including the inlet flow rate, temperature, outlet pressure, and thermal boundary conditions of the wall; Generate a computational grid based on the geometry of the heat exchanger, which is used to discretize the Navier-Stokes equations; The finite volume method is used to discretize the Navier-Stokes equations and transform the continuous partial differential equations into discrete algebraic equations. According to the discrete algebraic equations, the flow state information of the fluid inside the heat exchanger is extracted, including velocity distribution, pressure distribution and temperature distribution.

4. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 3 is characterized in that: Based on the geometry of the heat exchanger, a computational mesh is generated. The computational mesh is used to discretize the Navier-Stokes equations, including: Obtain all geometric data of the heat exchanger according to its geometric shape, including the diameter, length, arrangement of the pipe, the size, spacing, thickness of the fins, and the outer and inner walls of the heat exchanger; Generate computational mesh based on all geometric data of the heat exchanger; The geometric space inside the heat exchanger is divided into grid cells, and the physical quantities in each grid cell are regarded as constants; The steps to convert the Navier-Stokes equations from continuous differential form to the corresponding numerical form include: Each control volume contains a certain amount of fluid. The boundaries of the control volume are defined by mesh nodes. The Navier-Stokes equations are integrated within each control volume. The integrated equations are expressed as: ； in, is the velocity field, is the viscosity of the fluid, is an external force; is the volume region of the integration, representing a region in a three-dimensional space; is the velocity field About Time The partial derivative of is the unit normal vector; is the surface area of ​​the integration, representing the boundary surface; is the area of ​​the surface element; is the velocity field The gradient of is the dot product of the velocity gradient tensor with itself; is the gradient operator, which represents the rate of change in space.

5. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 4 is characterized in that: The finite volume method is used to discretize the Navier-Stokes equations and transform the continuous partial differential equations into discrete algebraic equations, including: The computational domain is divided into multiple control volumes, and the conservation quantity on each control volume is integrated to obtain its discretized algebraic equation; in two-dimensional or three-dimensional space, non-overlapping grids are used to divide the computational domain. In the two-dimensional case, the entire region is divided into multiple rectangular or quadrilateral control volumes; for each control volume, assume that its volume is V i , the surface is S i , each control volume has a normal vector on its outer boundary n i , and each control volume has a scalar value at its center; assuming that a conserved quantity is discretized, within the control volume, the conservation equation has the following form: ； in, is a conserved quantity, including velocity or temperature or mass; is the mass flux, including momentum flux or energy flux, etc.; is the source term, including body forces or heat sources; For a conserved quantity , integrate it in time and space, and apply Gauss's theorem to convert the volume integral of the flux into a surface integral; Time stepping using explicit time discretization methods; For the momentum equation of the fluid, the mass flux Expressed as the amount of fluid flowing over a surface; The fluxes are calculated based on the boundary conditions of the flow problem; the conservation equations for each control volume are discretized to obtain a system of algebraic equations.

6. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 5 is characterized in that: Apply Gauss's theorem to convert the volume integral of the flux into a surface integral, including: For a conserved quantity , integrate it in time and space, and apply Gauss's theorem to convert the volume integral of the flux into the surface integral, and we get: ； in, is the control volume, is the outer boundary surface of the control volume, is the outward normal vector.

7. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 6 is characterized in that: The corresponding calculation formula for time stepping using the explicit time discretization method is: ； in, is the current time step The conservation of time, is the value of the next time step, is the time step.

8. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 7 is characterized in that: Mass flux The calculation formula is: Assuming the calculated mass flux ,in, and represent the density and velocity of the fluid respectively; the upwind difference method is used to calculate the mass flux on the boundary, and the following approximation is obtained: ； in, By controlling the volume Mass flux on the surface; is the control volume Density inside; is the control volume The velocity vector inside; is the control volume Density inside; is the control volume The velocity vector inside.

9. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 8, characterized in that: The calculation process of the algebraic equation system includes: Assume that for a conserved quantity , and its discretized form in the control volume is: ； in, is the time step The value of the moment, is the flux, which represents the flow through the control volume and its neighbors of traffic, is the source term, which means that The volume source term in the control volume is solved by explicit method to obtain the conservation quantities of all control volumes in the whole computational domain. is the time step The value of the moment.

10. The heat exchanger intelligent design method based on deep reinforcement learning according to claim 9, characterized in that: The flow model and the heat transfer model are coupled to obtain a coupled performance evaluation model for evaluating the overall performance of the heat exchanger under specific operating conditions, including: Establish a data exchange interface so that the flow model and heat transfer model can recognize and receive the data output by each other; Determine the common parameters in the flow model and heat transfer model, including the temperature, pressure, and velocity of the fluid; Integrate the flow model and heat transfer model into one simulation environment so that the flow model and heat transfer model can run on the same time and space scale; Since the flow and heat transfer processes influence each other, in the coupling process, the final solution is found through iterative solution. In each iterative step, the flow model calculates the flow state of the fluid and transfers the flow state to the heat transfer model; the heat transfer model calculates the heat exchange according to the flow state and feeds the result back to the flow model. This is repeated until the predetermined number of iterations is reached to obtain the final coupling performance evaluation model.

Images