Design method of cooling water guide structure of compact condenser

Abstract

The application belongs to the technical field of condensers, and particularly relates to a design method of a cooling water flow guide structure of a compact condenser, which comprises the following steps: according to a condenser inlet water chamber structure design scheme, the type and size of the condenser inlet water chamber, the type and size of the cooling water inlet of the inlet water chamber are given; according to a condenser overall thermal design scheme, the condenser cooling water pipe bundle inlet velocity uniformity requirement and the condenser cooling water flow guide structure resistance requirement are given. The Latin hypercube sampling method is adopted, and the space filling curve technology is combined to realize the maximum space coverage with the least sample points, so that the obtained flow guide structure parameter combinations are as various and representative as possible. In the optimization process, a dynamic adjustment factor is introduced to dynamically adjust the sampling parameters, an intelligent sampling strategy based on a genetic algorithm is adopted, the sampling point distribution density is dynamically adjusted according to the previous sampling result, and the sampling efficiency and accuracy are improved.

Description

Technical Field

[0001] This invention belongs to the field of condenser technology, specifically a design method for a cooling water flow guiding structure of a compact condenser. Background Technology

[0002] The core function of a condenser is to cool and liquefy steam into a liquid, maintaining a certain degree of vacuum while efficiently recovering heat. It is primarily used to receive exhaust steam from the turbine, create and maintain a vacuum, enabling the steam to perform work effectively, ensuring system efficiency, and recovering condensate for boiler feedwater. Condensers are widely used in power generation, petrochemicals, nuclear energy, marine engineering, food and pharmaceuticals, HVAC, and environmental protection.

[0003] A compact design for condensers can effectively save space and reduce ancillary costs. The compact overall structure of the condenser facilitates its installation in narrow compartments, skids, or equipment mezzanines. A compact overall design means reduced equipment weight and lower foundation strength requirements; for mobile platforms such as offshore platforms, lightweighting leads to reduced energy consumption. The compact design also ensures structural rigidity, allowing for the use of specialized materials. For example, in marine engineering, titanium, which is corrosion-resistant but less rigid than steel, is used; a more compact structure makes it easier to maintain equipment rigidity. The compact design of the condenser also makes it easier to integrate the equipment with piping and valves as modules, facilitating factory prefabrication and reducing on-site installation difficulty and time.

[0004] Compact design of the condenser water chamber is a crucial aspect of compact condenser design. Optimizing the thickness of the condenser inlet water chamber is key to achieving this compactness. Therefore, optimizing the condenser inlet water chamber thickness effectively reduces the space occupied by the condenser, lowers the requirements for the condenser installation foundation, increases the structural rigidity of the condenser, and reduces the overall cost of the steam system equipment. The flow characteristics of the cooling water are one of the key factors affecting the condenser's condensation efficiency. Ensuring consistent flow within different heat exchange tubes is extremely important during the optimization of the condenser inlet water chamber thickness; this is a vital means of guaranteeing the condenser's heat exchange efficiency.

[0005] Reducing the thickness of the inlet water chamber in a compact condenser leads to uneven flow distribution within the cooling water tube bundle. Heat exchange tubes with low flow rates have low internal water velocities, resulting in a significant decrease in the heat transfer coefficient on the inner side of the tubes. These tubes are quickly heated to near steam temperature, causing a sharp drop in heat transfer capacity, transforming them into "thermal resistance" tubes. While heat exchange tubes with high flow rates offer better heat transfer, they cannot fully compensate for the performance losses caused by low-flow tubes. The result is a deterioration in the overall heat transfer performance of the condenser, increased unit heat consumption, and reduced output power, ultimately affecting the overall performance of the steam system. Therefore, reducing the thickness of the inlet water chamber in a compact condenser while maintaining condenser heat exchange efficiency is of great significance.

[0006] Current technologies typically increase the thickness of the condenser inlet water chamber to ensure uniform flow distribution within the cooling water tube bundle. However, this technology currently fails to meet the cost-saving and efficiency-improvement requirements of modern condenser design. Incorporating a flow-guiding structure inside the condenser inlet water chamber can both reduce its thickness and maintain uniform flow distribution within the cooling tube bundle. Currently, methods for designing cooling water flow-guiding structures for compact condensers are lacking, representing a pressing issue in the field of compact condensers. Summary of the Invention

[0007] A design method for a cooling water flow guide structure of a compact condenser includes the following steps: S1. Based on the structural design scheme of the condenser inlet water chamber, provide the type and size of the condenser inlet water chamber and the type and size of the cooling water inlet of the inlet water chamber; Based on the overall thermodynamic design of the condenser, the requirements for the uniformity of the inlet velocity of the condenser cooling water tube bundle and the resistance requirements of the condenser cooling water guide structure are given. S2. Based on the flow characteristics of fluid in a variable cross-section channel, it is proposed to assume that the static pressure on the streamline increases linearly within the fluid structure in order to suppress flow separation to the greatest extent and ensure the uniformity and stability of the fluid during the flow process. S3. Under the assumption of inviscid and incompressible fluid, the difference in static pressure between the inlet and outlet of the flow guiding structure is derived based on Bernoulli's equation theory, providing data input for subsequent velocity and cross-section design. S4. Based on Bernoulli's equation, the expression for the velocity distribution at any cross section along the axial direction within the flow guiding structure is theoretically derived, and a mathematical model of the velocity variation with position is established. S5. Based on the flow velocity distribution, the required cross-sectional radius of the guide structure at each axial position is further derived, and a radius variation function with the axial coordinate as the independent variable is established to provide a theoretical curve for the wall profile of the guide structure. S6. Based on the theoretical curve of the guide structure wall profile obtained from the above theoretical derivation, the ideal form of the guide structure under the assumption of inviscid and incompressible fluid is obtained. S7. Based on the actual engineering conditions and processing requirements, select the wall curve form that most closely approximates the theoretical result from the commonly used curve types.

[0008] S8. The key geometric parameters of the guide structure wall curve, the external dimensions of the guide structure, the positional relationship between the guide structure and the cooling water inlet of the inlet water chamber, and the positional relationship between the guide structure and the tube sheet are used as optimization design variables to lay the foundation for subsequent optimization. S9. Scientifically sample within the design variable space to construct a representative set of simulation conditions, so as to cover as many and representative parameter combinations as possible. S10. Based on the selected flow guide structure type and condenser inlet water chamber structure design scheme, establish a three-dimensional parametric model of the condenser inlet water chamber, including the flow guide structure. S11. Based on the tube bundle layout scheme, set the boundary conditions for the outlet of the three-dimensional simulation model of the condenser inlet water chamber; S12. Using computational fluid dynamics, a three-dimensional numerical simulation of the inlet water chamber flow field under different parameter combinations is performed to obtain the flow state in the flow field. The velocity distribution of the outlet section is extracted from the simulation results, and the flow resistance characteristics of the cooling water guide structure of the condenser are calculated. S13. Based on simulation sample data, use optimization algorithms to find the optimal combination of design variables, including the geometric parameters of the guide structure wall curve, the external dimensions of the guide structure, the positional relationship between the guide structure and the cooling water inlet of the inlet water chamber, and the positional relationship between the guide structure and the tube sheet.

[0009] Furthermore, in S2, when it is assumed that the static pressure on the streamlines within the fluid structure increases linearly, the effect of fluid viscosity is considered, i.e., a viscosity correction term is introduced. The static pressure distribution obtained based on the inviscid assumption is corrected, and the corrected static pressure distribution expression is as follows: in, For fluid dynamic viscosity, For the velocity gradient, For reference pressure, For fluid density, It is the acceleration due to gravity. For height coordinates, The static pressure coefficient increases linearly.

[0010] Furthermore, the radius of the cross-section of the flow guiding structure described in S5 at various axial positions, expressed as a radius variation function with the axial coordinate as the independent variable, is as follows: in, The inlet radius of the flow guide structure. The outlet radius of the flow guide structure, It represents the distance from any cross-section along the axial direction to the inlet cross-section of the guide structure. This is the distance from the inlet section to the outlet section of the flow guide structure. Let be the cross-sectional radius of the flow guide structure at each axial position.

[0011] Furthermore, the commonly used curve types in S7 include hyperbolas, parabolas, and spline curves. The selection of curves takes into account the ease of processing and manufacturing costs, and a processing cost coefficient is introduced. Under the premise of meeting the flow field performance requirements, the processing cost coefficient should be selected first. Smaller curve types serve as the wall curve form of the flow guiding structure, while The determination is based on the complexity of the processing technology and material consumption factors of different curve types.

[0012] Furthermore, S9 uses the Latin hypercube sampling method to sample the working conditions of the simulation calculation, using the fewest sample points to achieve the maximum spatial coverage, so that the obtained flow guiding structure parameter combinations are as diverse and representative as possible. In the Latin hypercube sampling process, the space-filling curve technique is combined to map the multidimensional design variable space to a one-dimensional space, then perform uniform sampling in the one-dimensional space, and finally map it back to the multidimensional design variable space, thereby ensuring that the distribution of sampling points in the design variable space is more uniform and reasonable.

[0013] Furthermore, in S12, when performing three-dimensional numerical simulation of the inlet water chamber flow field under different parameter combinations using computational fluid dynamics, dynamic mesh technology is adopted to consider the influence of vibration and deformation that may occur in the condenser during actual operation on the flow field. By updating the mesh in real time, the actual flow field can be simulated more accurately, thereby improving the reliability of the simulation results and the adaptability to actual working conditions.

[0014] Furthermore, in step S12, if the obtained flow guide structure resistance meets the predetermined standard, then the subsequent related steps are executed; if it does not meet the predetermined standard, then the part about simulation calculation and resistance judgment in steps S9-S12 is repeated until the flow guide structure resistance meets the predetermined standard. Furthermore, a dynamic adjustment factor is introduced during repeated execution. The sampling parameters are dynamically adjusted, and when the difference in flow guide structure resistance obtained from two consecutive simulations is less than a preset threshold... Dynamic adjustment factor according to Attenuation adjustments are made to more accurately approximate the flow guide structure resistance that meets predetermined standards. This is the dynamic adjustment factor used in the previous sampling. This is the preset attenuation coefficient.

[0015] Furthermore, in S13, if the optimal combination of design variables for the uniformity-optimized flow guide structure wall curve geometric parameters, flow guide structure external dimensions, flow guide structure and inlet cooling water inlet position relationship, and flow guide structure and tube sheet position relationship is not on the sampling boundary, then the subsequent related steps are executed. If the result is a local optimum on the sampling boundary, repeat steps S9-S13 regarding sampling, simulation calculation, and determination of optimization variables until the design variable combination with the best uniformity is no longer on the sampling boundary. Meanwhile, during the repeated sampling process, an intelligent sampling strategy based on genetic algorithms is adopted. According to the distribution of the optimized variable combination obtained from the previous sampling, the distribution density of sampling points is dynamically adjusted, and the number of sampling points is increased in areas where the global optimum may exist.

[0016] Furthermore, in S13, if the optimal combination of design variables with the best uniformity is not on the sampling boundary and the uniformity meets the minimum preset requirement, then the design of the cooling water guide structure of the compact condenser is completed. If the optimal design variable combination with the best uniformity is not on the sampling boundary and the overall uniformity does not meet the minimum preset requirement, but only the velocity amplitude in a local area is too large, then for the local flow non-uniform area at the inlet of the condenser cooling water tube bundle, after fine-tuning the tube bundle arrangement scheme to avoid the area with large flow non-uniformity, repeat the part about setting boundary conditions, simulation calculation and uniformity judgment in steps S11-S13, and re-perform the simulation calculation to ensure that the local adjustment of the tube bundle area has a small enough impact on the overall flow field until the uniformity meets the minimum preset requirement; If the optimal combination of design variables with the best uniformity is not on the sampling boundary, the overall uniformity does not meet the minimum preset requirement, and the uniformity cannot be made to meet the preset requirement by adjusting the tube bundle arrangement locally, then repeat the parts of steps S7-S13 regarding the selection of wall curve form, sampling, simulation calculation and optimization until the design requirements are met. Furthermore, during the repeated execution of the above steps, a multi-objective optimization model is introduced to comprehensively consider multiple objectives, including the resistance of the guide structure, the uniformity of the inlet velocity of the cooling water tube bundle, and the manufacturing cost of the guide structure. The weights of each objective are determined by weighted summation or analytic hierarchy process to seek the optimal guide structure design scheme under the comprehensive consideration of multiple objectives.

[0017] Furthermore, in S13, a surrogate model is established using Kriging, radial basis function, or multinomial regression methods to characterize the mapping relationship between design variables and the uniformity of the condenser cooling water tube bundle inlet velocity, and the optimal combination of design variables is sought using optimization algorithms. In the process of building the surrogate model, cross-validation is used to evaluate the accuracy of the surrogate model. The sample dataset is divided into training set and validation set. Through multiple random partitioning and model training and validation, the surrogate model with the smallest prediction error is selected as the final mapping model to improve the reliability of the optimization results.

[0018] The beneficial effects of this invention are as follows: 1. The present invention provides a flow guiding structure inside the inlet water chamber of a compact condenser. This structure precisely determines the shape curve and dimensions of the flow guiding structure, as well as the positional relationship between the flow guiding structure and the cooling water inlet and tube sheet of the inlet water chamber. This allows the cooling water to enter the tube bundle area more evenly and smoothly. With this structure, the thickness of the condenser inlet water chamber can be reasonably reduced, thereby reducing the overall size of the condenser. This not only reduces the space occupied by the condenser and optimizes the system spatial layout, but also reduces the construction cost and operating energy consumption of the system. 2. By performing detailed three-dimensional simulation calculations on the flow field inside the condenser water chamber, the relevant data of the flow field at the outlet of the condenser water chamber, i.e. the flow field at the inlet of the cooling water tube bundle, can be accurately obtained. By deeply analyzing this flow field, it can be ensured that the ratio of the maximum to minimum velocity component along the tube bundle direction is less than a specified value. Based on the flow field analysis results, the tube bundle arrangement of the condenser can be fine-tuned. Specifically, tubes in positions where the ratio of the maximum to minimum velocity component is greater than a specified value can be eliminated. This operation effectively avoids problems such as low local heat exchange efficiency and uneven tube bundle wear caused by uneven flow velocity, thereby improving the overall heat exchange performance and service life of the condenser. 3. The Latin hypercube sampling method is adopted and combined with the space filling curve technique to achieve the maximum spatial coverage with the fewest sample points, so that the obtained flow guiding structure parameter combination is as diverse and representative as possible. During the optimization process, a dynamic adjustment factor is introduced to dynamically adjust the sampling parameters. A smart sampling strategy based on genetic algorithm is adopted to dynamically adjust the sampling point distribution density according to the previous sampling results, which improves the efficiency and accuracy of sampling. 4. After performing simulation calculations on the selected operating conditions, a surrogate model for optimizing design parameters and objectives was constructed. This model allows for large-scale prediction of the uniformity of the condenser cooling water tube bundle inlet velocity within a short timeframe, avoiding lengthy and time-consuming simulation calculations near local optima. This accelerates the design cycle of the condenser's cooling water guide structure, enabling designers to obtain a design solution that meets requirements more quickly, thus improving design efficiency. Furthermore, during the establishment of the surrogate model, cross-validation technology was used to evaluate its accuracy, selecting the surrogate model with the smallest prediction error as the final mapping model, further enhancing the reliability of the optimization results. Attached Figure Description

[0019] Figure 1 This is a flowchart of the present invention; Figure 2 Schematic diagram of input variables for cooling water guide structure; Figure 3 (ad) is a three-dimensional model of the condenser inlet water chamber; Figure 4 Pressure cloud diagram of the flow field in the condenser inlet water chamber; Figure 5This is a vector cloud diagram of the flow field velocity in the condenser inlet water chamber. Figure 6 This is a contour map of the velocity amplitude at the inlet section of the condenser tube bundle region. Detailed Implementation

[0020] The present invention will now be further described with reference to the accompanying drawings.

[0021] Please see Figure 1-6 , This embodiment provides a design method for a cooling water flow guide structure of a compact condenser, including the following steps: S1. Based on the structural design scheme of the condenser inlet water chamber, provide the type and size of the condenser inlet water chamber and the type and size of the cooling water inlet of the inlet water chamber; Based on the overall thermodynamic design of the condenser, the requirements for the uniformity of the inlet velocity of the condenser cooling water tube bundle and the resistance requirements of the condenser cooling water guide structure are given. S2. Based on the flow characteristics of fluid in a variable cross-section channel, it is proposed to assume that the static pressure on the streamline increases linearly within the fluid structure in order to suppress flow separation to the greatest extent and ensure the uniformity and stability of the fluid during the flow process. In S2, when it is assumed that the static pressure along the streamlines within the fluid structure increases linearly, the effect of fluid viscosity is considered, i.e., a viscosity correction term is introduced. The static pressure distribution obtained based on the inviscid assumption is corrected, and the corrected static pressure distribution expression is as follows: in, For fluid dynamic viscosity, For the velocity gradient, For reference pressure, For fluid density, It is the acceleration due to gravity. For height coordinates, The static pressure coefficient increases linearly. S3. Under the assumption of inviscid and incompressible fluid, the difference in static pressure between the inlet and outlet of the flow guiding structure is derived based on Bernoulli's equation theory, providing data input for subsequent velocity and cross-section design. S4. Based on Bernoulli's equation, the expression for the velocity distribution at any cross section along the axial direction within the flow guiding structure is theoretically derived, and a mathematical model of the velocity variation with position is established. S5. Based on the flow velocity distribution, the required cross-sectional radius of the guide structure at each axial position is further derived, and a radius variation function with the axial coordinate as the independent variable is established to provide a theoretical curve for the wall profile of the guide structure. The radius of the cross-section of the flow guide structure at various axial positions, expressed as a radius variation function with the axial coordinate as the independent variable, is as follows: in, The inlet radius of the flow guide structure. The outlet radius of the guide structure, It represents the distance from any cross-section along the axial direction to the inlet cross-section of the guide structure. This is the distance from the inlet section to the outlet section of the flow guide structure. The radius of the cross-section of the flow guiding structure at each axial position; S6. Based on the theoretical curve of the guide structure wall profile obtained from the above theoretical derivation, the ideal form of the guide structure under the assumption of inviscid and incompressible fluid is obtained. S7. Considering actual engineering conditions and processing requirements, select the wall curve form that most closely approximates the theoretical result from commonly used curve types. The hyperbola model is... .

[0022] Commonly used curve types in S7 include hyperbolas, parabolas, and spline curves. The selection of curves takes into account the ease of fabrication and manufacturing costs, incorporating a fabrication cost coefficient. Under the premise of meeting the flow field performance requirements, the processing cost coefficient should be selected first. Smaller curve types serve as the wall curve form of the flow guiding structure, while The determination is based on the complexity of the processing technology and material consumption factors for different curve types; S8. The key geometric parameters of the guide structure wall curve, the external dimensions of the guide structure, the positional relationship between the guide structure and the cooling water inlet of the inlet water chamber, and the positional relationship between the guide structure and the tube sheet are used as optimization design variables to lay the foundation for subsequent optimization. S9. Scientifically sample within the design variable space to construct a representative set of simulation conditions, so as to cover as many and representative parameter combinations as possible. The Latin hypercube sampling method is used to sample the working conditions of the simulation calculation. The minimum number of sample points is used to achieve the maximum spatial coverage, so that the obtained flow guide structure parameter combination is as diverse and representative as possible. For example, when designing a flow guide structure, assume there are two key design variables: the inlet radius of the flow guide structure. and export radius ,Will The range of values ​​is divided into Equal intervals, similar The range of values ​​is divided, and then each... Randomly select a value within the range, each Within the range, a value is also randomly selected, and these values ​​are combined into different parameter combinations; for example, if =5, for The range of values The divided intervals are There are 5 intervals, and one is randomly selected from each interval. value, Similarly, this can yield 25 different results. Parameter combinations, compared to random sampling, can more comprehensively cover the parameter space; In the Latin hypercube sampling process, the space-filling curve technique is combined to map the multidimensional design variable space to a one-dimensional space, then uniform sampling is performed in the one-dimensional space, and finally the sampling points are mapped back to the multidimensional design variable space, thereby ensuring that the distribution of sampling points in the design variable space is more uniform and reasonable. like Figure 2 As shown, the parameters a, b, thickness h, distance L1 between the centerline of the guide structure and the centerline of the cooling water inlet of the inlet water chamber, and distance L2 between the guide structure and the panel of the inlet water chamber are selected as input variables for the hyperbolic model of the guide structure wall.

[0023] Divide the range of values ​​for each of the above variables into N non-overlapping intervals at equal intervals.

[0024] Random sampling is performed within each interval, meaning that for each variable, there is one and only one sample value in each interval. First, ensure that each parameter takes values ​​evenly, then shuffle their order and pair them. This ensures that the value of each parameter covers the entire range and that the sample points are scattered randomly and evenly. S10, such as Figure 3 As shown, 3a is a three-dimensional view of the cooling water inlet side, 3b is a three-dimensional view of the tube bundle area side, 3c is a three-dimensional view of the flow guiding structure, and 3d is a side view of the flow guiding structure. Based on the selected flow guiding structure type and the condenser inlet water chamber structure design scheme, a three-dimensional parametric model of the condenser inlet water chamber, including the flow guiding structure, is established. S11. Based on the tube bundle layout scheme, set the boundary conditions for the outlet of the three-dimensional simulation model of the condenser inlet water chamber, i.e., the inlet area of ​​the tube bundle. S12. Using computational fluid dynamics, a three-dimensional numerical simulation of the inlet water chamber flow field under different parameter combinations is performed to obtain the flow state in the flow field. The velocity distribution of the outlet section is extracted from the simulation results, and the flow resistance characteristics of the cooling water guide structure of the condenser are calculated. When performing three-dimensional numerical simulation of the inlet water chamber flow field under different parameter combinations using computational fluid dynamics, dynamic mesh technology is adopted to consider the impact of vibration and deformation of the condenser during actual operation on the flow field. By updating the mesh in real time, the actual flow field can be simulated more accurately, thereby improving the reliability of the simulation results and the adaptability to actual working conditions. If the obtained flow guide structure resistance meets the predetermined standard, then proceed with the subsequent related steps; if it does not meet the predetermined standard, then repeat the part about simulation calculation and resistance judgment in steps S9-S12 until the flow guide structure resistance meets the predetermined standard. Furthermore, a dynamic adjustment factor is introduced during repeated execution. The sampling parameters are dynamically adjusted, and when the difference in flow guide structure resistance obtained from two consecutive simulations is less than a preset threshold... Dynamic adjustment factor according to Attenuation adjustments are made to more accurately approximate the flow guide structure resistance that meets predetermined standards. This is the dynamic adjustment factor used in the previous sampling. The preset attenuation coefficient; For example, when optimizing the design of the flow guiding structure to reduce resistance, a preset threshold is set. The first simulation yielded the flow guide structure resistance as: The second simulation yielded the resistance as ,like According to the formula Adjusting the dynamic adjustment factor ,when =0.8, =0.1, then the adjusted =0.8×(1-0.1)=0.72, using the adjusted... Re-sampling and simulation were performed to gradually approximate the design that met the predetermined resistance standard; S13. Based on simulation sample data, use optimization algorithms to find the optimal combination of design variables, including the geometric parameters of the guide structure wall curve, the external dimensions of the guide structure, the positional relationship between the guide structure and the cooling water inlet of the inlet water chamber, and the positional relationship between the guide structure and the tube sheet. If the optimal combination of design variables for the uniformity-optimized flow guide structure wall curve geometry, flow guide structure dimensions, flow guide structure and inlet water chamber cooling water inlet position relationship, and flow guide structure and tube sheet position relationship is not on the sampling boundary, then proceed with the subsequent related steps. If the result is a local optimum on the sampling boundary, repeat steps S9-S13 regarding sampling, simulation calculation, and determination of optimization variables until the design variable combination with the best uniformity is no longer on the sampling boundary. Simultaneously, during repeated sampling, an intelligent sampling strategy based on a genetic algorithm is employed. Based on the distribution of optimized variable combinations obtained from the previous sampling, the distribution density of sampling points is dynamically adjusted, increasing the number of sampling points in regions where a global optimum might exist. For example, in the previous sampling, it was found that certain regions yielded better-performing flow structures from optimized variable combinations, potentially indicating a global optimum. Based on the idea of ​​a genetic algorithm, these better-performing variable combinations are considered "superior individuals," and the number of sampling points in their nearby parameter space is increased. For instance, in the previous sampling, in a certain… and If the flow-guiding structure has relatively low resistance in the selected region, then in subsequent sampling, samples should be selected more densely around this region. and The values ​​are combined to increase the probability of finding the global optimum.

[0025] In S13, if the optimal combination of design variables with the best uniformity is not on the sampling boundary and the uniformity meets the minimum preset requirement, then the design of the cooling water guide structure of the compact condenser is completed. If the optimal design variable combination with the best uniformity is not on the sampling boundary and the overall uniformity does not meet the minimum preset requirement, but only the velocity amplitude in a local area is too large, then for the local flow non-uniform area at the inlet of the condenser cooling water tube bundle, after fine-tuning the tube bundle arrangement scheme to avoid the area with large flow non-uniformity, repeat the part about setting boundary conditions, simulation calculation and uniformity judgment in steps S11-S13, and re-perform the simulation calculation to ensure that the local adjustment of the tube bundle area has a small enough impact on the overall flow field until the uniformity meets the minimum preset requirement; If the optimal combination of design variables with the best uniformity is not on the sampling boundary, the overall uniformity does not meet the minimum preset requirement, and the uniformity cannot be made to meet the preset requirement by adjusting the tube bundle arrangement locally, then repeat the parts of steps S7-S13 regarding the selection of wall curve form, sampling, simulation calculation and optimization until the design requirements are met. Furthermore, during the repeated execution of the above steps, a multi-objective optimization model is introduced to comprehensively consider multiple objectives, including the resistance of the guide structure, the uniformity of the inlet velocity of the cooling water tube bundle, and the manufacturing cost of the guide structure. The weights of each objective are determined by weighted summation or analytic hierarchy process to seek the optimal guide structure design scheme under the comprehensive consideration of multiple objectives.

[0026] A surrogate model is established using Kriging, radial basis function, or multinomial regression methods to characterize the mapping relationship between design variables and the uniformity of the condenser cooling water tube bundle inlet velocity, and the optimal combination of design variables is sought using optimization algorithms. In the process of establishing the surrogate model, cross-validation is used to evaluate its accuracy. The sample dataset is divided into training and validation sets. Through multiple random partitioning and model training and validation, the surrogate model with the smallest prediction error is selected as the final mapping model to improve the reliability of the optimization results. For example, assuming there are 100 sets of sample data on the design parameters of the flow guide structure and the corresponding uniformity of the cooling water tube bundle inlet velocity, these 100 sets of data are randomly divided into 80 training sets and 20 validation sets. A Kriging surrogate model is built using the training set data, and then the prediction error of the model is verified using the validation set data. This process is repeated multiple times, with the training and validation sets randomly partitioned each time. Finally, the Kriging surrogate model with the smallest prediction error is selected as the final model to predict the uniformity of the cooling water tube bundle inlet velocity under different design parameters.

Claims

1. A design method for a cooling water flow guiding structure of a compact condenser, characterized in that, Includes the following steps: S1. Based on the structural design scheme of the condenser inlet water chamber, provide the type and size of the condenser inlet water chamber and the type and size of the cooling water inlet of the inlet water chamber; Based on the overall thermodynamic design of the condenser, the requirements for the uniformity of the inlet velocity of the condenser cooling water tube bundle and the resistance requirements of the condenser cooling water guide structure are given. S2. Based on the flow characteristics of fluid in a variable cross-section channel, it is proposed to assume that the static pressure on the streamline increases linearly within the fluid structure in order to suppress flow separation to the greatest extent and ensure the uniformity and stability of the fluid during the flow process. S3. Under the assumption of inviscid and incompressible fluid, the difference in static pressure between the inlet and outlet of the flow guiding structure is derived based on Bernoulli's equation theory, providing data input for subsequent velocity and cross-section design. S4. Based on Bernoulli's equation, the expression for the velocity distribution at any cross section along the axial direction within the flow guiding structure is theoretically derived, and a mathematical model of the velocity variation with position is established. S5. Based on the flow velocity distribution, the required cross-sectional radius of the guide structure at each axial position is further derived, and a radius variation function with the axial coordinate as the independent variable is established to provide a theoretical curve for the wall profile of the guide structure. S6. Based on the theoretical curve of the guide structure wall profile obtained from the above theoretical derivation, the ideal form of the guide structure under the assumption of inviscid and incompressible fluid is obtained. S7. Based on actual engineering conditions and processing requirements, select the wall curve form that most closely approximates the theoretical result from commonly used curve types. S8. The key geometric parameters of the guide structure wall curve, the external dimensions of the guide structure, the positional relationship between the guide structure and the cooling water inlet of the inlet water chamber, and the positional relationship between the guide structure and the tube sheet are used as optimization design variables to lay the foundation for subsequent optimization. S9. Scientifically sample within the design variable space to construct a representative set of simulation conditions, so as to cover as many and representative parameter combinations as possible. S10. Based on the selected flow guide structure type and condenser inlet water chamber structure design scheme, establish a three-dimensional parametric model of the condenser inlet water chamber, including the flow guide structure. S11. Based on the tube bundle layout scheme, set the boundary conditions for the outlet of the three-dimensional simulation model of the condenser inlet water chamber; S12. Using computational fluid dynamics, a three-dimensional numerical simulation of the inlet water chamber flow field under different parameter combinations is performed to obtain the flow state in the flow field. The velocity distribution of the outlet section is extracted from the simulation results, and the flow resistance characteristics of the cooling water guide structure of the condenser are calculated. S13. Based on simulation sample data, use optimization algorithms to find the optimal combination of design variables, including the geometric parameters of the guide structure wall curve, the external dimensions of the guide structure, the positional relationship between the guide structure and the cooling water inlet of the inlet water chamber, and the positional relationship between the guide structure and the tube sheet.

2. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: In S2, when it is assumed that the static pressure on the streamlines within the fluid structure increases linearly, the effect of fluid viscosity is considered, i.e., a viscosity correction term is introduced. The static pressure distribution obtained based on the inviscid assumption is corrected, and the corrected static pressure distribution expression is as follows: in, For fluid dynamic viscosity, For the velocity gradient, For reference pressure, For fluid density, It is the acceleration due to gravity. For height coordinates, The static pressure coefficient increases linearly.

3. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: The radius of the cross-section of the flow guiding structure described in S5 at various axial positions, expressed as a radius variation function with the axial coordinate as the independent variable, is as follows: in, The inlet radius of the flow guide structure. The outlet radius of the guide structure, It represents the distance from any cross-section along the axial direction to the inlet cross-section of the guide structure. This is the distance from the inlet section to the outlet section of the flow guide structure. Let be the cross-sectional radius of the flow guide structure at each axial position.

4. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: The commonly used curve types in S7 include hyperbolas, parabolas, and spline curves. The selection of curves takes into account the ease of processing and manufacturing costs, and a processing cost coefficient is introduced. Under the premise of meeting the flow field performance requirements, the processing cost coefficient should be selected first. Smaller curve types serve as the wall curve form of the flow guiding structure, while The determination is based on the complexity of the processing technology and material consumption factors of different curve types.

5. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: The S9 uses the Latin hypercube sampling method to sample the working conditions of the simulation calculation, using the fewest sample points to achieve the largest spatial coverage, so that the obtained flow guiding structure parameter combinations are as diverse and representative as possible. In the Latin hypercube sampling process, the space-filling curve technique is combined to map the multidimensional design variable space to a one-dimensional space, then perform uniform sampling in the one-dimensional space, and finally map it back to the multidimensional design variable space, thereby ensuring that the distribution of sampling points in the design variable space is more uniform and reasonable.

6. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: In S12, when performing three-dimensional numerical simulation of the inlet water chamber flow field under different parameter combinations using computational fluid dynamics, dynamic mesh technology is adopted to consider the influence of vibration and deformation that may occur in the condenser during actual operation on the flow field. By updating the mesh in real time, the actual flow field can be simulated more accurately, thereby improving the reliability of the simulation results and the adaptability to actual working conditions.

7. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: In step S12, if the obtained flow guide structure resistance meets the predetermined standard, then the subsequent related steps are executed; if it does not meet the predetermined standard, then the part about simulation calculation and resistance judgment in steps S9-S12 is repeated until the flow guide structure resistance meets the predetermined standard. Furthermore, a dynamic adjustment factor is introduced during repeated execution. The sampling parameters are dynamically adjusted, and when the difference in flow guide structure resistance obtained from two consecutive simulations is less than a preset threshold... Dynamic adjustment factor according to Attenuation adjustments are made to more accurately approximate the flow guide structure resistance that meets predetermined standards. This is the dynamic adjustment factor used in the previous sampling. This is the preset attenuation coefficient.

8. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: In step S13, if the optimal combination of design variables for the uniformity-optimized flow guide structure wall curve geometric parameters, flow guide structure external dimensions, flow guide structure and inlet water chamber cooling water inlet position relationship, and flow guide structure and tube sheet position relationship is not on the sampling boundary, then the subsequent related steps are executed. If the result is a local optimum on the sampling boundary, repeat steps S9-S13 regarding sampling, simulation calculation, and determination of optimization variables until the design variable combination with the best uniformity is no longer on the sampling boundary. Meanwhile, during the repeated sampling process, an intelligent sampling strategy based on genetic algorithms is adopted. According to the distribution of the optimized variable combination obtained from the previous sampling, the distribution density of sampling points is dynamically adjusted, and the number of sampling points is increased in areas where the global optimum may exist.

9. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: In step S13, if the optimal design variable combination with the best uniformity is not on the sampling boundary and the uniformity meets the minimum preset requirement, then the design of the cooling water guide structure of the compact condenser is completed. If the optimal design variable combination with the best uniformity is not on the sampling boundary and the overall uniformity does not meet the minimum preset requirement, but only the velocity amplitude in a local area is too large, then for the local flow non-uniform area at the inlet of the condenser cooling water tube bundle, after fine-tuning the tube bundle arrangement scheme to avoid the area with large flow non-uniformity, repeat the part about setting boundary conditions, simulation calculation and uniformity judgment in steps S11-S13, and re-perform the simulation calculation to ensure that the local adjustment of the tube bundle area has a small enough impact on the overall flow field until the uniformity meets the minimum preset requirement; If the optimal combination of design variables with the best uniformity is not on the sampling boundary, the overall uniformity does not meet the minimum preset requirement, and the uniformity cannot be made to meet the preset requirement by adjusting the tube bundle arrangement locally, then repeat the parts of steps S7-S13 regarding the selection of wall curve form, sampling, simulation calculation and optimization until the design requirements are met. Furthermore, during the repeated execution of the above steps, a multi-objective optimization model is introduced to comprehensively consider multiple objectives, including the resistance of the guide structure, the uniformity of the inlet velocity of the cooling water tube bundle, and the manufacturing cost of the guide structure. The weights of each objective are determined by weighted summation or analytic hierarchy process to seek the optimal guide structure design scheme under the comprehensive consideration of multiple objectives.

10. The design method of the cooling water guiding structure of a compact condenser according to claim 1, characterized in that: In S13, a surrogate model is established using Kriging, radial basis function, or multinomial regression methods to characterize the mapping relationship between design variables and the uniformity of the condenser cooling water tube bundle inlet velocity, and the optimal combination of design variables is sought using optimization algorithms. In the process of building the surrogate model, cross-validation is used to evaluate the accuracy of the surrogate model. The sample dataset is divided into training set and validation set. Through multiple random partitioning and model training and validation, the surrogate model with the smallest prediction error is selected as the final mapping model to improve the reliability of the optimization results.

Images