A method for optimizing design of condenser extraction port position based on CFD simulation

Abstract

The present application relates to the technical field of ship condenser simulation, and particularly relates to a condenser suction port position optimization design method based on CFD simulation, comprising the following steps: S1, obtaining geometric parameters, cooling tube bundle arrangement parameters and operating condition parameters; S2, performing geometric repair on the condenser shell side flow field geometric model; S3, establishing a mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium; S4, obtaining a condenser shell side flow field calculation model; S5, obtaining flow field data; S6, iterating until the convergence criterion is met and outputting the optimal suction port position; by equivalenting the cooling tube bundle to a porous medium model and loading a porous medium resistance source term and a vapor-liquid two-phase condensation mass transfer model, a global optimization algorithm is used for dynamic iterative optimization, and finally the optimal suction port position is output, so that the calculation cost is greatly reduced while ensuring the condensation heat exchange simulation precision, and the quantitative and automatic optimization of the suction port position is realized.

Description

Technical Field

[0001] This invention relates to the technical field of ship condenser simulation, and in particular to a method for optimizing the design of condenser exhaust port location based on CFD simulation. Background Technology

[0002] The condenser is the core heat exchanger in a steam power system, and its shell-side flow field characteristics directly determine the cycle thermal efficiency and economy of the power system. With the development of computational fluid dynamics (CFD) technology, using numerical simulation to perform refined analysis of the condensation heat transfer process on the condenser shell side has become an important approach for the optimized design of large industrial condensers. In current engineering practice, the simulation of the condenser shell-side flow field typically uses the multiphase flow model built into CFD software, simplifying the steam condensation process into a heat and mass transfer process for solution, and combining the distribution of non-condensable gases on the shell side to design and evaluate the location of the extraction port.

[0003] However, existing CFD simulation methods still have shortcomings in the detailed modeling of the condenser shell-side flow field. Firstly, the multiphase flow models built into CFD software tend to oversimplify the latent heat release, heat and mass transfer coupling, and phase interface evolution during steam phase change condensation, making it difficult to accurately characterize the shell-side condensation heat transfer process. Secondly, condenser cooling tube bundles are densely packed; when using full-size modeling, the geometry of each tube bundle needs to be constructed individually, resulting in unstructured meshes on the order of tens of millions, leading to an exponential increase in computational resource requirements, which is insufficient for practical engineering applications. These two shortcomings make it difficult for existing simulation methods to achieve both high accuracy in condensation heat transfer process simulation while maintaining acceptable computational costs.

[0004] Furthermore, the spatial distribution of non-condensable gases has a decisive impact on condenser operating efficiency. Local accumulation of these gases forms a gaseous thermal resistance layer on the cooling tube walls, hindering direct heat transfer between steam and the cooling medium and causing vacuum degradation. The location of the evacuation port directly determines the effectiveness of non-condensable gas removal. However, current engineering practices mainly rely on physical experiments and empirical judgment to determine the optimal location of the evacuation port. Physical experiments are costly and time-consuming, while empirical judgment is highly subjective and lacks universality. Summary of the Invention

[0005] To address the technical problems existing in the background art, this invention proposes a method for optimizing the design of condenser extraction port location based on CFD simulation, the specific scheme of which is as follows: A method for optimizing the location of condenser exhaust ports based on CFD simulation includes the following steps: S1. Obtain the geometric parameters of the condenser shell-side flow field, the cooling tube bundle arrangement parameters, and the operating condition parameters; S2. Construct a geometric model of the condenser shell-side flow field based on the geometric parameters, and perform geometric repair on the geometric model of the condenser shell-side flow field. S3. The geometric model of the condenser shell-side flow field after geometric repair is meshed to generate a computational domain, wherein the area where the cooling tube bundle is located is determined as the tube bundle region, and the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium is established. S4. The tube bundle region is equivalent to a porous medium model. A porous medium resistance source term is loaded in the tube bundle region, and a vapor-liquid two-phase condensation mass transfer model is loaded in the calculation domain to obtain the condenser shell-side flow field calculation model. S5. Based on the preset boundary conditions and initial conditions of the operating condition parameters, the condenser shell-side flow field calculation model is numerically solved to obtain flow field data including velocity field, non-condensable gas concentration field, pressure field and temperature field. S6. Based on the flow field data, establish an energy efficiency objective function characterized by the ratio of heat exchange to pressure drop. Based on the non-condensable gas concentration field and the velocity field, identify candidate regions for the extraction port. Use a global optimization algorithm to update the position of the extraction port within the candidate regions and perform numerical solution again. Iterate until the convergence criterion is met, and then output the optimal extraction port position.

[0006] Furthermore, in S3, the geometrically repaired condenser shell-side flow field geometric model is meshed to generate a computational domain, including: Based on the characteristic structure of the condenser shell-side flow field geometric model, the computational domain is divided into an air-cooled region, a main condensation region, and a non-tube bundle region, wherein the region where the cooling tube bundle is located specifically includes the air-cooled region and the main condensation region. The loading region of the porous media resistance source term is specifically the air-cooled zone and the main condensation zone, while the non-tube bundle zone is not loaded with the porous media resistance source term.

[0007] Furthermore, in S3, the cooling tube bundle arrangement parameters include the outer diameter of the tubes and the spacing between the tubes; when the cooling tube bundle is arranged in a triangular pattern, establishing the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium includes: Calculate the ratio of the outer diameter of the pipe to the distance between the pipes, and square the ratio to obtain the geometric percentage square value; The quotient obtained by dividing pi by two square roots of three is used as the constant coefficient. The porosity of the porous medium in the tube bundle region is obtained by subtracting the product of the constant coefficient and the square of the geometric proportion.

[0008] Furthermore, in S3, the cooling tube bundle arrangement parameters include the outer diameter of the tubes, the radial tube center distance, the local arc radius, and the corresponding central angle; when the cooling tube bundle is arranged radially, establishing the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium includes: The reference cross-sectional area is obtained by multiplying pi by the square of the outer diameter of the tube. The reference control area is obtained by multiplying four times the radial tube center distance, the local arc radius, and the corresponding central angle. The porosity of the porous medium at different radial positions in the tube bundle region is calculated by subtracting the ratio of the reference cross-sectional area to the reference control area.

[0009] Furthermore, in S4, a two-phase vapor-liquid condensation mass transfer model is loaded into the computational domain to obtain a condenser shell-side flow field calculation model, including: The local water vapor temperature, cooling water temperature, heat transfer area, latent heat of vaporization, control volume, and total thermal resistance of the flow field are obtained. The heat transfer temperature difference is obtained by calculating the difference between the local water vapor temperature and the cooling water temperature; The product of the heat transfer temperature difference and the heat transfer area is calculated as the heat transfer driving term, and the product of the latent heat of vaporization, the volume of the control body, and the total thermal resistance is calculated as the thermal resistance term. The condensation rate is obtained by dividing the heat transfer driving term by the thermal resistance term. The negative value of the condensation rate is used as the vapor phase side source term characterizing the reduction of steam condensation, and the positive value of the condensation rate is used as the liquid phase side source term characterizing the increase of liquid water.

[0010] Furthermore, in S4, the tube bundle region is equivalent to a porous medium model, and a porous medium resistance source term is loaded into the tube bundle region, including: Obtain the tube spacing and tube outer diameter from the cooling tube bundle arrangement parameters; The frictional resistance coefficient in the first direction is obtained based on the Reynolds number of the fluid in the first direction, and the frictional resistance coefficient in the second direction is obtained based on the Reynolds number of the fluid in the second direction. Calculate twice the ratio of the frictional resistance coefficient in the first direction to the pipe spacing to obtain the first multiplier in the first direction; Calculate the product of the pipe spacing and the porosity of the porous medium, divide the product by the difference between the pipe spacing and the outer diameter of the pipe, and square the resulting quotient to obtain the second multiplier; Multiplying the first multiplier in the first direction with the second multiplier yields the porous medium resistance coefficient in the first direction; Multiply the first-direction porous medium drag coefficient, fluid density, velocity vector magnitude and the first-direction velocity component together to obtain the first-direction porous medium drag component. Perform the same calculation steps as for the first direction on the second direction to obtain the porous medium resistance component in the second direction.

[0011] Furthermore, in S5, based on the preset boundary and initial conditions of the operating condition parameters, the condenser shell-side flow field calculation model is numerically solved, including: Set the boundary conditions, wherein the wall is set as an adiabatic boundary, the outlet absolute pressure boundary condition is set, and the mass fraction of the non-condensable gas recirculation component at the outlet is set to be consistent with the mass fraction of the non-condensable gas at the inlet or extrapolated by the adjacent unit, and the inlet flow rate and the inlet non-condensable gas mass fraction are set. The initial conditions are set, including setting an absolute pressure close to the actual operating conditions of the condenser as the initial absolute pressure, and setting the initial velocity, initial turbulent kinetic energy and initial specific dissipation rate in each direction of the flow field. The numerical solution is considered to have converged when the residuals of the continuity equation, momentum equation, energy equation, and component equation are all below the set residual threshold, and the relative change of the outlet mass flow rate with iteration is less than the set change threshold.

[0012] Furthermore, in S6, an energy efficiency objective function characterized by the ratio of heat exchange to pressure drop is established based on the flow field data, including: Extract the condensate mass flow rate, inlet pressure data, and outlet pressure data from the flow field data; The heat exchange rate is obtained by multiplying the condensation mass flow rate by the latent heat of vaporization of the fluid. The pressure drop is obtained by subtracting the inlet pressure data from the outlet pressure data. The quotient obtained by dividing the heat exchange by the pressure drop is used as the value of the energy efficiency objective function.

[0013] Furthermore, in S6, identifying candidate regions for the extraction port based on the non-condensable gas concentration field and the velocity field includes: Calculate the average mass fraction and standard deviation of noncondensable gases across the entire noncondensable gas concentration field, and calculate the average velocity across the entire velocity field. When traversing all flow field regions, if the local non-condensable gas mass fraction in a region is higher than the sum of the average non-condensable gas mass fraction across the entire field and one time the standard deviation of the concentration, and the local velocity is less than half of the average velocity across the entire field, then that region is designated as a candidate region for the extraction port.

[0014] Furthermore, in S6, a global optimization algorithm is used to update the position of the air extraction port within the candidate region and the numerical solution is performed again. After iterating until the convergence criterion is met, the optimal air extraction port position is output, including: A preset number of initial sample point coordinates are generated within the candidate region of the air extraction port using Latin hypercube sampling. The positions of the extraction ports in the condenser shell-side flow field calculation model are updated according to the coordinates of each initial sample point, and the corresponding energy efficiency objective function values ​​are obtained by numerical solution, thereby constructing the initial sample set. Based on the initial sample set, a Kriging surrogate model is constructed, and a genetic algorithm is used to search for candidate point coordinates on the Kriging surrogate model that make the predicted energy efficiency objective function value optimal. Substitute the candidate point coordinates into the condenser shell-side flow field calculation model to perform a real numerical solution, obtain the real energy efficiency objective function value, and supplement it to the sample set to update the Kriging proxy model. Repeatedly perform the proxy model search and update operation. When the relative change of the actual energy efficiency objective function value corresponding to the optimal candidate point within a set number of consecutive times is less than the set convergence threshold, it is determined that the convergence criterion is met and the iteration stops.

[0015] Compared with the prior art, the present invention can achieve at least the following beneficial effects: 1. This invention equates the cooling tube bundle to a porous medium model and loads a porous medium resistance source term and a vapor-liquid two-phase condensation mass transfer model. It obtains flow field data by numerically solving under near-vacuum boundary conditions. Then, based on the energy efficiency objective function of the heat transfer to pressure drop ratio, it identifies candidate regions for the extraction port by combining the concentration field and velocity field of non-condensable gases. It uses a global optimization algorithm to dynamically iterate and find the optimal position of the extraction port, thereby significantly reducing the computational cost while ensuring the accuracy of condensation heat transfer simulation and realizing the quantitative and automated optimization of the extraction port position.

[0016] 2. This invention treats the region where the cooling tube bundle is located as an equivalent porous medium model, and uses porosity and resistance characteristics to characterize each discrete cooling tube bundle, avoiding full-size tube-by-tube modeling. This significantly reduces the mesh size of the tube bundle region from tens of millions of units, thereby significantly reducing the computational resource requirements and simulation costs while ensuring the accuracy of the flow field solution. Furthermore, by loading a vapor-liquid two-phase condensation mass transfer model into the control equation and introducing a porous medium resistance source term for correction, this invention compensates for the shortcomings of traditional CFD software's built-in phase change model in characterizing the steam condensation heat and mass transfer process, thus improving the simulation accuracy of the shell-side condensation heat transfer process. Boundary and initial conditions under near-vacuum conditions are set based on actual operating conditions, and convergence is determined by both residual and outlet mass flow rate to ensure that the obtained flow field data are stable physical solutions and improve the reliability of flow field simulation results. An energy efficiency objective function characterized by the ratio of heat transfer to pressure drop is constructed. Candidate regions for the extraction port are identified based on the concentration and velocity fields of non-condensable gases, and a global optimization algorithm is used for dynamic iterative optimization to achieve quantitative and automated determination of the extraction port location, eliminating the reliance on physical experiments and empirical judgments. The parameters required for modeling and solving are determined based on the actual structure and operating conditions of the condenser, so that the entire optimization design method can be adapted to condensers with different structural forms, improving the engineering universality of the method. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart of the overall method of the present invention.

[0018] Figure 2 This is a flowchart of the mesh generation and porosity mapping process of the present invention.

[0019] Figure 3 This is a flowchart of the tube bundle equivalent modeling and mass transfer model loading process of the present invention. Detailed Implementation

[0020] Embodiments of the present invention are described in detail below. Examples of these embodiments are illustrated in the accompanying drawings, wherein the same or similar symbols denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0021] Please refer to Figures 1 to 3 As shown, this invention provides a method for optimizing the location of a condenser extraction port based on CFD simulation, comprising the following steps: S1. Obtain the geometric parameters of the condenser shell-side flow field, the cooling tube bundle arrangement parameters, and the operating condition parameters.

[0022] It should be noted that the three types of parameters obtained correspond to the input sources for subsequent geometric modeling, tube bundle equivalent modeling, and boundary condition setting, respectively, and their values ​​are determined based on the actual structure and operating conditions of the condenser.

[0023] The geometric parameters of the condenser shell-side flow field include the external dimensions and relative positions of the main condensation zone and the air-cooled zone, as well as the dimensions and arrangement of characteristic structures such as the steam inlet, main steam passage, auxiliary air intake passage, baffle, exhaust port, non-condensable gas collection passage, and liquid phase outlet. The cooling tube bundle arrangement parameters include the arrangement method of the tube bundle (triangular arrangement or radial arrangement), the outer diameter of the tubes, and the spacing between the tubes. The operating condition parameters include the inlet steam pressure and temperature, inlet flow velocity, cooling water temperature, and the mass fraction of non-condensable gases leaking in with the steam.

[0024] It should be noted that the above three types of parameters can be obtained directly from condenser design drawings, equipment nameplates and operation monitoring records, or they can be calculated based on design selection data. This scheme does not limit the specific source of the parameters, as long as they can fully describe the geometric boundaries, tube bundle structure and operating conditions required for subsequent modeling and solving.

[0025] S2. Construct a geometric model of the condenser shell-side flow field based on the geometric parameters, and perform geometric repair on the geometric model of the condenser shell-side flow field.

[0026] It should be noted that a two-dimensional geometric model of the condenser shell side is constructed based on the geometric parameters obtained from S1. This two-dimensional geometric model includes characteristic structures such as the main condensation zone, air-cooled zone, steam inlet, main steam passage, auxiliary air intake passage, baffle, exhaust port, non-condensable gas collection passage, and liquid phase outlet. These characteristic structures are identified and named so that their respective parameters can be set during subsequent mesh generation and physical model loading.

[0027] It should be noted that the internal structure of the condenser shell is complex and contains numerous features. Directly imported geometric models often contain discontinuities such as redundant points, defective surfaces, and overlapping boundaries. If these are not addressed, they will directly affect the mesh generation quality and even the solution convergence. The geometric repair described in this solution involves deleting redundancies and repairing defects in the constructed geometric model, eliminating redundant geometric elements, ensuring the integrity and continuity of the model structure, and providing defect-free geometric input for subsequent mesh generation.

[0028] Please see Figure 2 S3. The geometric model of the condenser shell-side flow field after geometric repair is meshed to generate a computational domain, wherein the area where the cooling tube bundle is located is determined as the tube bundle region, and the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium is established.

[0029] It should be noted that the geometrically repaired model is meshed to form a computational domain for numerical solutions, and the dense region containing the cooling tube bundle is separately defined as the tube bundle region. The tube bundle region consists of a large number of finely arranged cooling tubes. If the geometry is built and meshed for each tube individually, the number of meshes will reach tens of millions, consuming huge computational resources. This scheme treats the tube bundle region as a whole during the meshing stage, laying the foundation for subsequently treating this region as an equivalent porous medium model and significantly reducing the mesh size.

[0030] It should be noted that the porosity of porous media is a physical quantity characterizing the proportion of fluid-flowable space within the tube bundle region, and its magnitude directly affects the value of the subsequent porous media resistance source term. This scheme establishes a mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous media. That is, the porosity is quantitatively determined based on the tube bundle arrangement and geometric dimensions, so that the porosity can be automatically updated as the tube bundle structure changes, ensuring the correspondence between the equivalent model and the actual tube bundle structure.

[0031] In an optional embodiment, in S3, the geometrically repaired condenser shell-side flow field geometric model is meshed to generate a computational domain, including: Based on the characteristic structure of the condenser shell-side flow field geometric model, the computational domain is divided into an air-cooled region, a main condensation region, and a non-tube bundle region, wherein the region where the cooling tube bundle is located specifically includes the air-cooled region and the main condensation region. The loading region of the porous media resistance source term is specifically the air-cooled zone and the main condensation zone, while the non-tube bundle zone is not loaded with the porous media resistance source term.

[0032] It should be noted that the air-cooled zone is the area in the condenser used for separating and accumulating non-condensable gases, the main condensation zone is the main area for condensation heat exchange between steam and cooling water, and both are areas with densely arranged cooling tube bundles; the non-tube bundle area refers to the steam passages and other areas outside the tube bundles. After dividing the computational domain into these three zones, porous media resistance source terms are only applied to the air-cooled zone and the main condensation zone. The non-tube bundle area, treated as a normal flow channel, is not subject to porous media resistance, thus avoiding the incorrect addition of resistance to the non-tube bundle area and resulting in flow field distortion. Taking a practical engineering example, in a certain ship's condenser, the porosity of the tube bundle area in the air-cooled zone and the main condensation zone is set to 0.42, while the porosity of the non-tube bundle area is treated as 1.

[0033] In an optional embodiment, in S3, the cooling tube bundle arrangement parameters include the outer diameter of the tubes and the spacing between the tubes; when the cooling tube bundle is arranged in a triangular pattern, establishing the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium includes: Calculate the ratio of the outer diameter of the pipe to the distance between the pipes, and square the ratio to obtain the geometric percentage square value; The quotient obtained by dividing pi by two square roots of three is used as the constant coefficient. The porosity of the porous medium in the tube bundle region is obtained by subtracting the product of the constant coefficient and the square of the geometric proportion.

[0034] It should be noted that the triangular arrangement is a method of arranging tube bundles along the vertices of an equilateral triangle. Its porosity depends on the relative relationship between the outer diameter of the tubes and the spacing between them. The denser the tube bundle arrangement and the larger the outer diameter of the tubes, the smaller the proportion of pores through which fluid can flow, and the lower the porosity. In this scheme, the ratio of the outer diameter of the tubes to the spacing between them is first calculated and squared, reflecting the proportion of the cross-sectional area of ​​a single tube in the area of ​​the tube bundle unit. The quotient obtained by dividing pi by 2√3 is the geometric conversion constant for the triangular arrangement, derived from the geometric relationship between the unit area and the cross-sectional area of ​​a circular tube under an equilateral triangle arrangement. Subtracting the product of this constant and the square of the geometric proportion from 1 yields the porosity of the tube bundle region. Taking a triangular arrangement with an outer diameter of 0.025 meters and a spacing of 0.032 meters as an example, its porosity is approximately between 0.4 and 0.45.

[0035] In an optional embodiment, in S3, the cooling tube bundle arrangement parameters include the outer diameter of the tubes, the radial tube center distance, the local arc radius, and the corresponding central angle; when the cooling tube bundle is arranged radially, establishing the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium includes: The reference cross-sectional area is obtained by multiplying pi by the square of the outer diameter of the tube. The reference control area is obtained by multiplying four times the radial tube center distance, the local arc radius, and the corresponding central angle. The porosity of the porous medium at different radial positions in the tube bundle region is calculated by subtracting the ratio of the reference cross-sectional area to the reference control area.

[0036] It should be noted that the radial arrangement refers to the simultaneous radial and circumferential distribution of the tube bundles, spreading out radially. The density of the tube bundles varies at different radial positions, resulting in a radial variation in porosity, with different porosity values ​​corresponding to different radial positions. In this scheme, the product of pi and the square of the tube's outer diameter is the reference cross-sectional area of ​​a single tube; the product of four times the radial tube center distance, the local arc radius, and the corresponding central angle is the reference control area of ​​a single control volume, reflecting the fan-shaped area occupied by each tube bundle unit at that radial position; subtracting the ratio of the reference cross-sectional area to the reference control area from 1 yields the porosity at that radial position. Unlike the single porosity of the triangular arrangement, the radial arrangement yields a porosity distribution that varies with radial position, thus enabling anisotropic resistance characterization in subsequent steps.

[0037] Please see Figure 3S4. The tube bundle region is equivalent to a porous medium model. A porous medium resistance source term is loaded in the tube bundle region, and a vapor-liquid two-phase condensation mass transfer model is loaded in the calculation domain to obtain the condenser shell-side flow field calculation model.

[0038] It should be noted that the tube bundle region defined in S3 is equivalent to a porous media model, that is, the discrete cooling tube bundles are replaced by an equivalent continuous medium with porosity and resistance characteristics. The cooling tubes in the tube bundle region impede the steam flow, and their effect is characterized by adding a resistance source term to the momentum equation of the porous media model. This allows for the reconstruction of the flow hindrance effect of the tube bundle region at a lower computational cost, avoiding the tens of millions of mesh scales resulting from tube-by-tube modeling.

[0039] It should be noted that this scheme also loads a vapor-liquid two-phase condensation mass transfer model into the computational domain. The essence of the condenser shell side is the phase change process of steam condensing on the cooling tube wall. The vapor-liquid two-phase condensation mass transfer model is used to describe the mass transfer of steam condensing into liquid water and the accompanying latent heat release, which is crucial for ensuring the simulation accuracy of the condensation heat transfer process. Traditional CFD software's built-in phase change models are insufficient in characterizing the heat and mass transfer coupling of steam condensation; the vapor-liquid two-phase condensation mass transfer model loaded in this scheme is a targeted correction to this deficiency.

[0040] In an optional embodiment, in S4, a vapor-liquid two-phase condensation mass transfer model is loaded into the computational domain to obtain a condenser shell-side flow field computational model, including: The local water vapor temperature, cooling water temperature, heat transfer area, latent heat of vaporization, control volume, and total thermal resistance of the flow field are obtained. The heat transfer temperature difference is obtained by calculating the difference between the local water vapor temperature and the cooling water temperature; The product of the heat transfer temperature difference and the heat transfer area is calculated as the heat transfer driving term, and the product of the latent heat of vaporization, the volume of the control body, and the total thermal resistance is calculated as the thermal resistance term. The condensation rate is obtained by dividing the heat transfer driving term by the thermal resistance term. The negative value of the condensation rate is used as the vapor phase side source term characterizing the reduction of steam condensation, and the positive value of the condensation rate is used as the liquid phase side source term characterizing the increase of liquid water.

[0041] It should be noted that the driving force of the condensation process comes from the heat transfer temperature difference between the local steam temperature and the cooling water temperature; the larger the temperature difference, the higher the condensation rate. This scheme first multiplies the heat transfer temperature difference by the heat transfer area to obtain the heat transfer driving term, characterizing the scale of heat transfer per unit temperature difference. Then, it multiplies the latent heat of vaporization, the control volume, and the total thermal resistance to obtain the thermal resistance term. The latent heat of vaporization reflects the heat released by the condensation of a unit mass of steam, the total thermal resistance reflects the obstruction to heat transfer by the cooling water side and the pipe wall side, and the control volume is used to normalize the condensation rate to a unit volume. Dividing the heat transfer driving term by the thermal resistance term yields the condensation rate per unit volume. A negative value of this condensation rate is taken as the vapor phase source term, indicating a decrease in steam due to condensation; a positive value is taken as the liquid phase source term, indicating an increase in liquid water due to condensation. The two terms are equal in magnitude but opposite in sign, ensuring the conservation of mass between the vapor and liquid phases. Taking near-vacuum conditions as an example, when the cooling water temperature is about 300 Kelvin and the local water vapor temperature is slightly higher than the cooling water temperature by a few Kelvin, the condensation rate is maintained at a low reasonable level and approaches zero as the temperature difference decreases.

[0042] In an optional embodiment, in S4, the tube bundle region is equivalent to a porous medium model, and a porous medium resistance source term is loaded into the tube bundle region, including: Obtain the tube spacing and tube outer diameter from the cooling tube bundle arrangement parameters; The frictional resistance coefficient in the first direction is obtained based on the Reynolds number of the fluid in the first direction, and the frictional resistance coefficient in the second direction is obtained based on the Reynolds number of the fluid in the second direction. Calculate twice the ratio of the frictional resistance coefficient in the first direction to the pipe spacing to obtain the first multiplier in the first direction; Calculate the product of the pipe spacing and the porosity of the porous medium, divide the product by the difference between the pipe spacing and the outer diameter of the pipe, and square the resulting quotient to obtain the second multiplier; Multiplying the first multiplier in the first direction with the second multiplier yields the porous medium resistance coefficient in the first direction; Multiply the first-direction porous medium drag coefficient, fluid density, velocity vector magnitude and the first-direction velocity component together to obtain the first-direction porous medium drag component. Perform the same calculation steps as for the first direction on the second direction to obtain the porous medium resistance component in the second direction.

[0043] It should be noted that the resistance to flow from the tube bundle is directional; the resistance experienced by steam along the axial and transverse directions of the tube bundle differs. Therefore, this step calculates the resistance components in the first and second directions separately to characterize the anisotropy of the tube bundle resistance. The frictional resistance coefficient is determined by the Reynolds number in the corresponding direction, reflecting the influence of the flow state on the resistance. The first multiplier is formed by twice the ratio of the frictional resistance coefficient to the tube spacing. The second multiplier is formed by dividing the product of the tube spacing and porosity by the difference between the tube spacing and the outer diameter of the tubes. Multiplying these two multipliers yields the resistance coefficient in that direction, which comprehensively reflects the combined effects of the tube bundle geometry, porosity, and flow state. Multiplying the resistance coefficient by the fluid density, the magnitude of the velocity vector, and the velocity component in that direction yields the resistance component in that direction, which is added as a source term in the momentum equation. The second direction is calculated using the same steps, thus simultaneously reflecting the resistance effects in both directions in the momentum equation.

[0044] S5. Based on the preset boundary conditions and initial conditions of the operating condition parameters, the condenser shell-side flow field calculation model is numerically solved to obtain flow field data including velocity field, non-condensable gas concentration field, pressure field and temperature field.

[0045] It should be noted that, based on the operating condition parameters obtained in S1, boundary conditions and initial conditions are set, and the shell-side flow field calculation model obtained in S4 is numerically solved to obtain complete flow field data. The condenser shell side is under near-vacuum conditions, with steam and non-condensable gases mixing and flowing, accompanied by condensation phase change. The flow field has the characteristics of strong nonlinearity and multi-physics coupling. It is necessary to set boundary conditions and initial conditions that fit the actual operating conditions to ensure the physical rationality and convergence of the solution.

[0046] In an optional embodiment, in S5, based on the preset boundary conditions and initial conditions of the operating condition parameters, the condenser shell-side flow field calculation model is numerically solved, including: Set the boundary conditions, wherein the wall is set as an adiabatic boundary, the outlet absolute pressure boundary condition is set, and the mass fraction of the non-condensable gas recirculation component at the outlet is set to be consistent with the mass fraction of the non-condensable gas at the inlet or extrapolated by the adjacent unit, and the inlet flow rate and the inlet non-condensable gas mass fraction are set. The initial conditions are set, including setting an absolute pressure close to the actual operating conditions of the condenser as the initial absolute pressure, and setting the initial velocity, initial turbulent kinetic energy and initial specific dissipation rate in each direction of the flow field. The numerical solution is considered to have converged when the residuals of the continuity equation, momentum equation, energy equation, and component equation are all below the set residual threshold, and the relative change of the outlet mass flow rate with iteration is less than the set change threshold.

[0047] It should be noted that the boundary conditions use an adiabatic boundary for the wall because the condenser shell is well insulated, and the heat loss to the outside is negligible compared to the heat released during condensation. An absolute pressure is set at the outlet to reflect a near-vacuum environment; for example, the outlet absolute pressure is set to 5700 Pa. The mass fraction of the non-condensable gas recirculation component at the outlet is set to avoid non-physical component distributions caused by recirculation at the outlet during iterative solutions. Its value can be consistent with the inlet non-condensable gas mass fraction or extrapolated from neighboring elements at the outlet. The inlet flow velocity and inlet non-condensable gas mass fraction are set according to the operating conditions; for example, the inlet flow velocity is set to 64 m / s, and the inlet non-condensable gas mass fraction is set to 8.28 × 10⁻⁶ m / s. -5 This reflects the rated amount of non-condensable gas that leaks in.

[0048] It should be noted that the initial conditions are set to accelerate solution convergence. The initial absolute pressure is set to a value close to the actual operating conditions of the condenser, such as a gauge pressure of 101325 Pa. The initial velocity, initial turbulent kinetic energy, and initial specific dissipation rate are set to reasonable initial values. For example, the velocity component perpendicular to the steam inlet direction is set to zero, the velocity component along the inlet direction is set to the inlet velocity, and the turbulent kinetic energy and specific dissipation rate are taken as initial values ​​that match the inlet turbulence intensity. Reasonable initial values ​​can shorten the number of iterations required for convergence and avoid excessive deviation of the initial values, which could lead to solution divergence.

[0049] It should be noted that the convergence criterion considers both residuals and flow rate: on the one hand, the residuals of the continuity equation, momentum equation, energy equation, and component equation are all required to be below a set residual threshold; on the other hand, the relative change in outlet mass flow rate with iteration is required to be less than a set change threshold. If only the residuals decrease while the flow rate continues to fluctuate, it indicates that the flow field is not yet stable; convergence is only determined when both conditions are met, to ensure that the obtained flow field data is a stable physical solution.

[0050] S6. Based on the flow field data, establish an energy efficiency objective function characterized by the ratio of heat exchange to pressure drop. Based on the non-condensable gas concentration field and the velocity field, identify candidate regions for the extraction port. Use a global optimization algorithm to update the position of the extraction port within the candidate regions and perform numerical solution again. Iterate until the convergence criterion is met, and then output the optimal extraction port position.

[0051] It should be noted that the quality of the exhaust port location directly affects the discharge effect of non-condensable gases and the vacuum degree of the condenser. This scheme uses the ratio of heat exchange to pressure drop as the energy efficiency objective function. This ratio takes into account both heat exchange capacity and flow loss. The exhaust port location corresponding to the maximum value is the optimal location.

[0052] It should be noted that the identification of candidate regions for the extraction port is based on the concentration field and velocity field of non-condensable gases. Regions with high non-condensable gas mass fraction and low flow velocity are precisely where non-condensable gases tend to accumulate and are difficult to expel. Defining these regions as candidate regions can narrow down the optimization scope, avoid blind global search, and ensure that the optimization results are physically reasonable.

[0053] In an optional embodiment, in S6, an energy efficiency objective function characterized by the ratio of heat transfer to pressure drop is established based on the flow field data, including: Extract the condensate mass flow rate, inlet pressure data, and outlet pressure data from the flow field data; The heat exchange rate is obtained by multiplying the condensation mass flow rate by the latent heat of vaporization of the fluid. The pressure drop is obtained by subtracting the inlet pressure data from the outlet pressure data. The quotient obtained by dividing the heat exchange by the pressure drop is used as the value of the energy efficiency objective function.

[0054] It should be noted that the heat transfer is obtained by multiplying the condensation mass flow rate and the latent heat of vaporization in the flow field data, reflecting the actual heat release from steam condensation on the shell side; the pressure drop is obtained by subtracting the inlet pressure from the outlet pressure, reflecting the flow resistance loss on the shell side. Dividing the heat transfer by the pressure drop yields the energy efficiency objective function value, which physically represents the heat transfer per unit pressure drop, i.e., the effective heat transfer gain per unit flow loss. A larger function value indicates a stronger heat transfer capacity under the same flow loss, and a more optimal extraction port location. It should be noted that there is a coupling between the heat transfer and pressure drop; the extraction port location affects the distribution of non-condensable gases, thus affecting condensation heat transfer and flow resistance. Therefore, this objective function can comprehensively reflect the advantages and disadvantages of the extraction port location.

[0055] In an optional embodiment, in S6, identifying candidate regions for the extraction port based on the non-condensable gas concentration field and the velocity field includes: Calculate the average mass fraction and standard deviation of noncondensable gases across the entire noncondensable gas concentration field, and calculate the average velocity across the entire velocity field. When traversing all flow field regions, if the local non-condensable gas mass fraction in a region is higher than the sum of the average non-condensable gas mass fraction across the entire field and one time the standard deviation of the concentration, and the local velocity is less than half of the average velocity across the entire field, then that region is designated as a candidate region for the extraction port.

[0056] It should be noted that the average mass fraction of non-condensable gases, the standard deviation of their concentration, and the average velocity across the entire field are first calculated as the judgment criteria. The determination of candidate regions adopts a dual-condition criterion: if the local mass fraction of non-condensable gases is higher than the sum of the overall average and one standard deviation of concentration, it indicates that the region is significantly enriched with non-condensable gases; if the local velocity is less than half of the overall average velocity, it indicates that the region is in a state of flow stagnation, where the gas is difficult to be carried away by the mainstream. Regions that simultaneously meet both conditions are precisely the locations where non-condensable gases are most likely to accumulate and where extraction ports are most needed; these are designated as candidate regions. Using one standard deviation and half the average velocity as thresholds ensures that candidate regions cover the actual accumulation areas while avoiding overly broad thresholds that might mistakenly include normal flow areas, thus balancing optimization efficiency and accuracy.

[0057] In an optional embodiment, in S6, a global optimization algorithm is used to update the position of the air extraction port within the candidate region and the numerical solution is performed again. After iterating until the convergence criterion is met, the optimal air extraction port position is output, including: A preset number of initial sample point coordinates are generated within the candidate region of the air extraction port using Latin hypercube sampling. The positions of the extraction ports in the condenser shell-side flow field calculation model are updated according to the coordinates of each initial sample point, and the corresponding energy efficiency objective function values ​​are obtained by numerical solution, thereby constructing the initial sample set. Based on the initial sample set, a Kriging surrogate model is constructed, and a genetic algorithm is used to search for candidate point coordinates on the Kriging surrogate model that make the predicted energy efficiency objective function value optimal. Substitute the candidate point coordinates into the condenser shell-side flow field calculation model to perform a real numerical solution, obtain the real energy efficiency objective function value, and supplement it to the sample set to update the Kriging proxy model. Repeatedly perform the proxy model search and update operation. When the relative change of the actual energy efficiency objective function value corresponding to the optimal candidate point within a set number of consecutive times is less than the set convergence threshold, it is determined that the convergence criterion is met and the iteration stops.

[0058] It should be noted that, since the computational cost of each flow field numerical solution is high, the computational burden would be unbearable if all possible locations within the candidate region were solved one by one. Therefore, this scheme adopts a surrogate model and global optimization strategy: first, Latin hypercube sampling is used to uniformly generate a preset number of initial sample points within the candidate region. These sample points are spatially uniform, representative, and can comprehensively cover the candidate region; for each sample point, the position of the exhaust port is updated and the actual numerical solution is performed to obtain the corresponding energy efficiency objective function value, which constitutes the initial sample set.

[0059] It should be noted that the Kriging surrogate model is an interpolation-based surrogate model that predicts the function value of unknown locations based on known sample points. It constructs an approximate surface of the energy efficiency objective function within the entire candidate region based on the correspondence between the exhaust port positions and energy efficiency objective function values ​​in the initial sample set. The genetic algorithm, on the other hand, is a global optimization algorithm that simulates the biological evolution process. By searching on the Kriging surrogate model, it quickly finds the coordinates of candidate points that optimize the predicted energy efficiency objective function value, thus locating potential optimal locations at a cost far lower than solving each point individually.

[0060] It should be noted that since the surrogate model is only an approximation, its predicted optimal point may not correspond to the true optimal point. Therefore, the coordinates of the candidate point need to be substituted into the actual flow field calculation model to perform numerical solution, obtain the true energy efficiency objective function value, and then add this point to the sample set to update the Kriging surrogate model, so that the surrogate model gradually approximates the true function. Repeat the loop of surrogate model search, true solution, and model update. When the relative change of the true energy efficiency objective function value corresponding to the optimal candidate point is less than the set convergence threshold within a set number of consecutive iterations, it indicates that the optimization result has stabilized, the convergence criterion is satisfied, the iteration stops, and the position of the exhaust port corresponding to the optimal candidate point is output.

[0061] In summary, this invention combines equivalent porous media modeling of tube bundles with gas-liquid two-phase coupled simulation, and uses the heat exchange ratio to pressure drop ratio as the objective function for global optimization, thereby achieving quantitative automatic optimization of the condenser extraction port position.

Claims

1. A method for optimizing the location of a condenser extraction port based on CFD simulation, characterized in that, Includes the following steps: S1. Obtain the geometric parameters of the condenser shell-side flow field, the cooling tube bundle arrangement parameters, and the operating condition parameters; S2. Construct a geometric model of the condenser shell-side flow field based on the geometric parameters, and perform geometric repair on the geometric model of the condenser shell-side flow field. S3. The geometric model of the condenser shell-side flow field after geometric repair is meshed to generate a computational domain, wherein the area where the cooling tube bundle is located is determined as the tube bundle region, and the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium is established. S4. The tube bundle region is equivalent to a porous medium model. A porous medium resistance source term is loaded in the tube bundle region, and a vapor-liquid two-phase condensation mass transfer model is loaded in the calculation domain to obtain the condenser shell-side flow field calculation model. S5. Based on the preset boundary conditions and initial conditions of the operating condition parameters, the condenser shell-side flow field calculation model is numerically solved to obtain flow field data including velocity field, non-condensable gas concentration field, pressure field and temperature field. S6. Based on the flow field data, establish an energy efficiency objective function characterized by the ratio of heat exchange to pressure drop. Based on the non-condensable gas concentration field and the velocity field, identify candidate regions for the extraction port. Use a global optimization algorithm to update the position of the extraction port within the candidate regions and perform numerical solution again. Iterate until the convergence criterion is met, and then output the optimal extraction port position.

2. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S3, the geometrically repaired condenser shell-side flow field geometric model is meshed to generate a computational domain, including: Based on the characteristic structure of the condenser shell-side flow field geometric model, the computational domain is divided into an air-cooled region, a main condensation region, and a non-tube bundle region, wherein the region where the cooling tube bundle is located specifically includes the air-cooled region and the main condensation region. The loading region of the porous media resistance source term is specifically the air-cooled zone and the main condensation zone, while the non-tube bundle zone is not loaded with the porous media resistance source term.

3. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S3, the cooling tube bundle arrangement parameters include the outer diameter of the tubes and the spacing between the tubes; when the cooling tube bundle is arranged in a triangular pattern, establishing the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium includes: Calculate the ratio of the outer diameter of the pipe to the distance between the pipes, and square the ratio to obtain the geometric percentage square value; Calculate the quotient obtained by dividing pi by two square roots of three, and use it as the constant coefficient; The porosity of the porous medium in the tube bundle region is obtained by subtracting the product of the constant coefficient and the square of the geometric proportion.

4. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S3, the cooling tube bundle arrangement parameters include the outer diameter of the tubes, the radial tube center distance, the local arc radius, and the corresponding central angle; when the cooling tube bundle is arranged radially, establishing the mapping relationship between the cooling tube bundle arrangement parameters and the porosity of the porous medium includes: The reference cross-sectional area is obtained by multiplying pi by the square of the outer diameter of the tube. The reference control area is obtained by multiplying four times the radial tube center distance, the local arc radius, and the corresponding central angle. The porosity of the porous medium at different radial positions in the tube bundle region is calculated by subtracting the ratio of the reference cross-sectional area to the reference control area.

5. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S4, a two-phase vapor-liquid condensation mass transfer model is loaded into the computational domain to obtain a condenser shell-side flow field calculation model, including: The local water vapor temperature, cooling water temperature, heat transfer area, latent heat of vaporization, control volume, and total thermal resistance of the flow field are obtained. The heat transfer temperature difference is obtained by calculating the difference between the local water vapor temperature and the cooling water temperature; The product of the heat transfer temperature difference and the heat transfer area is calculated as the heat transfer driving term, and the product of the latent heat of vaporization, the volume of the control body, and the total thermal resistance is calculated as the thermal resistance term. The condensation rate is obtained by dividing the heat transfer driving term by the thermal resistance term. The negative value of the condensation rate is used as the vapor phase side source term characterizing the reduction of steam condensation, and the positive value of the condensation rate is used as the liquid phase side source term characterizing the increase of liquid water.

6. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S4, the tube bundle region is equivalent to a porous medium model, and a porous medium resistance source term is loaded into the tube bundle region, including: Obtain the tube spacing and tube outer diameter from the cooling tube bundle arrangement parameters; The frictional resistance coefficient in the first direction is obtained based on the Reynolds number of the fluid in the first direction, and the frictional resistance coefficient in the second direction is obtained based on the Reynolds number of the fluid in the second direction. Calculate twice the ratio of the frictional resistance coefficient in the first direction to the pipe spacing to obtain the first multiplier in the first direction; Calculate the product of the pipe spacing and the porosity of the porous medium, divide the product by the difference between the pipe spacing and the outer diameter of the pipe, and square the resulting quotient to obtain the second multiplier; Multiplying the first multiplier in the first direction with the second multiplier yields the porous medium resistance coefficient in the first direction; Multiply the first-direction porous medium drag coefficient, fluid density, velocity vector magnitude and the first-direction velocity component together to obtain the first-direction porous medium drag component. Perform the same calculation steps as for the first direction on the second direction to obtain the porous medium resistance component in the second direction.

7. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S5, based on the preset boundary and initial conditions of the operating condition parameters, the numerical solution of the condenser shell-side flow field calculation model is performed, including: Set the boundary conditions, wherein the wall is set as an adiabatic boundary, the outlet absolute pressure boundary condition is set, and the mass fraction of the non-condensable gas recirculation component at the outlet is set to be consistent with the mass fraction of the non-condensable gas at the inlet or extrapolated by the adjacent unit, and the inlet flow rate and the inlet non-condensable gas mass fraction are set. The initial conditions are set, including setting an absolute pressure close to the actual operating conditions of the condenser as the initial absolute pressure, and setting the initial velocity, initial turbulent kinetic energy and initial specific dissipation rate in each direction of the flow field. The numerical solution is considered to have converged when the residuals of the continuity equation, momentum equation, energy equation, and component equation are all below the set residual threshold, and the relative change of the outlet mass flow rate with iteration is less than the set change threshold.

8. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S6, an energy efficiency objective function characterized by the ratio of heat exchange to pressure drop is established based on the flow field data, including: Extract the condensate mass flow rate, inlet pressure data, and outlet pressure data from the flow field data; The heat exchange rate is obtained by multiplying the condensation mass flow rate by the latent heat of vaporization of the fluid. The pressure drop is obtained by subtracting the inlet pressure data from the outlet pressure data. The quotient obtained by dividing the heat exchange by the pressure drop is used as the value of the energy efficiency objective function.

9. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S6, candidate regions for the extraction port are identified based on the non-condensable gas concentration field and the velocity field, including: Calculate the average mass fraction and standard deviation of noncondensable gases across the entire noncondensable gas concentration field, and calculate the average velocity across the entire velocity field. When traversing all flow field regions, if the local non-condensable gas mass fraction in a region is higher than the sum of the average non-condensable gas mass fraction across the entire field and one time the standard deviation of the concentration, and the local velocity is less than half of the average velocity across the entire field, then that region is designated as a candidate region for the extraction port.

10. The condenser extraction port location optimization design method based on CFD simulation as described in claim 1, characterized in that: In S6, a global optimization algorithm is used to update the position of the air extraction port within the candidate region and the numerical solution is performed again. After iterating until the convergence criterion is met, the optimal air extraction port position is output, including: A preset number of initial sample point coordinates are generated within the candidate region of the air extraction port using Latin hypercube sampling. The positions of the extraction ports in the condenser shell-side flow field calculation model are updated according to the coordinates of each initial sample point, and the corresponding energy efficiency objective function values ​​are obtained by numerical solution, thereby constructing the initial sample set. Based on the initial sample set, a Kriging surrogate model is constructed, and a genetic algorithm is used to search for candidate point coordinates on the Kriging surrogate model that make the predicted energy efficiency objective function value optimal. Substitute the candidate point coordinates into the condenser shell-side flow field calculation model to perform a real numerical solution, obtain the real energy efficiency objective function value, and supplement it to the sample set to update the Kriging proxy model. Repeatedly perform the proxy model search and update operation. When the relative change of the actual energy efficiency objective function value corresponding to the optimal candidate point within a set number of consecutive times is less than the set convergence threshold, it is determined that the convergence criterion is met and the iteration stops.

Images