A method for optimizing the structure of a quenching oil tank based on CFD simulation

By pre-setting independent and controllable oil injection holes on the sidewall of the quenching oil bath and optimizing the switching mode of the oil injection holes using CFD simulation, the systematic problem of the quenching oil bath structure design was solved, achieving uniform cooling effect and production flexibility, and reducing cost and energy consumption.

CN123069841APending Publication Date: 2026-11-10CHONGQING UNIV +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611044691.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-14
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

The existing design of quenching oil tank structure lacks a systematic approach, the layout of oil nozzles cannot adapt to the dynamic changes in cooling demand, the multi-physics coupling degree of numerical simulation methods is insufficient, and the design of oil nozzle combination faces the problem of combination explosion, resulting in uneven cooling effect and high R&D cost.

Method used

Multiple independent and controllable oil injection holes are pre-set on the side wall of the quenching oil bath. A multiphysics coupling model is established by combining CFD simulation. The switching mode of the oil injection holes is optimized by sensitivity analysis, surrogate model and genetic algorithm to achieve targeted enhanced cooling.

Benefits of technology

It improves the uniformity of workpiece cooling, reduces R&D costs and energy consumption, and enhances the flexible production capacity of quenching equipment for multiple products.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123069841A_ABST
    Figure CN123069841A_ABST
Patent Text Reader

Abstract

This invention provides a method for optimizing the structure of a quenching oil bath based on CFD simulation. The method pre-defines N oil injection holes with independently switchable oil valves on the sidewall of an immersion quenching oil bath. A multiphysics coupled CFD simulation model is established, encompassing Eulerian multiphase flow, wall boiling heat transfer, and workpiece solid-state heat conduction. The maximum temperature difference on the workpiece surface, heat transfer coefficient uniformity, and vapor film discharge efficiency are extracted in real time as optimization objectives. Sensitivity analysis is used to screen candidate oil holes, Latin hypercube sampling is used to construct simulation samples, random forest is used to construct a discrete surrogate model, and a binary genetic algorithm is used for multi-objective optimization iteration to finally determine the optimal open / closed state for each oil injection hole. This method can improve quenching cooling uniformity, reduce workpiece deformation and residual stress, while reducing ineffective circulation flow, lowering pump power consumption, and supporting the storage of multiple switching modes to adapt to different workpieces, achieving flexible production.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of heat treatment equipment design and numerical simulation technology, and in particular to a method for optimizing the structure of a quenching oil tank based on CFD simulation. Background Technology

[0002] Quenching is a key process in metal heat treatment. Its principle involves heating the metal workpiece to its austenitizing temperature and then rapidly immersing it in a quenching medium. Rapid cooling is achieved through convective heat transfer, nucleation boiling, and film boiling on the workpiece surface, resulting in a hardened structure such as martensite or bainite to meet mechanical property requirements. The uniformity of the quenching cooling process is a crucial factor determining the final quality of the workpiece, directly affecting its residual stress distribution, dimensional deformation, and overall mechanical properties. In industrial production, immersion quenching oil baths are the most widely used quenching equipment. Workpieces are suspended in an array on a rack and immersed in the quenching oil bath to a predetermined depth until quenching is complete. The internal structural design significantly influences the flow path, turbulence intensity, and vapor film rupture behavior of the quenching medium.

[0003] In terms of quenching oil bath structural design, current engineering practices typically employ forced convection cooling of the workpiece by arranging fixed oil nozzles on the sidewalls or bottom of the oil bath. The layout of the oil nozzles is mainly determined based on engineering experience and production trial and error; the overall structure remains fixed once manufactured, and the oil flow path remains relatively constant throughout the quenching process. To further improve the cooling effect, some equipment also incorporates auxiliary structures such as agitators or guide plates to enhance the overall flow intensity of the quenching medium.

[0004] In the numerical simulation of quenching processes, with the development of computational fluid dynamics (CFD) technology, researchers have applied it to the simulation analysis of quenching processes. Existing studies typically model the temperature field or flow field independently during the quenching process: one approach establishes only the internal heat conduction model of the workpiece, using a preset heat transfer coefficient as boundary condition to predict the workpiece temperature evolution; another approach focuses on flow field analysis, evaluating the cooling effect by simulating the flow state of the quenching medium. Furthermore, for boiling heat transfer processes, some studies have used simplified models based on nucleus boiling or film boiling to describe the heat transfer boundary. However, the aforementioned existing techniques have the following significant shortcomings and defects:

[0005] A. The design of oil bath structures lacks a systematic approach and relies excessively on accumulated experience. The design of existing immersion quenching oil bath nozzle layouts and flow guiding structures primarily depends on production experience and repeated trials, lacking a systematic research method that matches the immersion trajectory of the hoisted workpiece. This approach not only has a long development cycle and high costs, but also makes it difficult to achieve optimal cooling structure configurations for different workpieces, failing to meet the needs of flexible, multi-variety production.

[0006] B. Fixed oil spraying structures cannot adapt to the dynamic changes in cooling requirements during quenching. In existing oil baths, the position and opening state of the spray nozzles remain fixed, and the oil flow path remains constant throughout the quenching process. However, during the immersion of the workpiece in quenching oil, the distribution of heat load and the vapor film coverage in different areas dynamically evolve over time. A fixed flow field structure cannot effectively target and enhance cooling of the workpiece's hot zones and vapor film retention areas. This easily leads to vapor film retention in complex geometric features such as grooves and pores, resulting in severely low local cooling rates, which in turn can cause localized soft spots, cracking, or severe deformation of the workpiece.

[0007] C. Existing numerical simulation methods suffer from insufficient multiphysics coupling, resulting in limited prediction accuracy. Most existing quenching numerical simulation methods employ independent analysis of the temperature or flow field, typically neglecting the strong coupling effect between the intense two-phase flow generated when the workpiece is immersed in the quenching medium and the workpiece's thermal conductivity. These methods cannot accurately capture the dynamic evolution of the oil-vapor interface, the formation and mechanical stripping of the vapor film, and the non-uniform distribution of the transient heat transfer coefficient on the workpiece surface. Therefore, they struggle to provide reliable quantitative data for optimizing oil bath structures.

[0008] D. The design of combined injection holes faces the problem of combination explosion, lacking efficient optimization strategies. If multiple independently controllable injection holes are set on the sidewall of the oil sump, the total number of combinations of their on / off states increases exponentially with the number of holes N, making it impractical to rely on full CFD simulation for enumeration.

[0009] Therefore, developing a method for optimizing the structure of a quenching oil tank based on CFD simulation is of great significance. Summary of the Invention

[0010] The purpose of this invention is to provide a method for optimizing the structure of a quenching oil bath based on CFD simulation, so as to solve the problems existing in the prior art.

[0011] The technical solution adopted to achieve the purpose of this invention is as follows: a method for optimizing the structure of a quenching oil tank based on CFD simulation. The quenching oil tank is an immersion quenching oil tank, including an oil tank body, a plurality of oil spray holes disposed on the side wall of the oil tank body, an oil delivery pipeline independently connected to each oil spray hole, and an oil valve with an independent opening / closing function installed on each oil delivery pipeline. The plurality of oil spray holes are arranged along the normal direction of the side wall of the oil tank in a regular array, and each oil spray hole has an independent number i=1,2,3,…,N. The opening / closing state of the oil spray hole is determined by the state variable s. i ={0,1} describes the state, where 0 represents off and 1 represents on. The method includes the following steps:

[0012] S1) Based on the actual dimensions of the workpiece and the quenching oil bath, establish a three-dimensional parametric geometric model of the workpiece and the quenching oil bath, and pre-determine the normal direction of the oil bath sidewall. The oil injection holes are arranged in a regular array to achieve an improved design of the quenching oil tank.

[0013] S2) Based on the improved design of the quenching oil tank, a CFD simulation calculation model was established based on the principles of computational fluid dynamics, which includes the Euler multiphase flow model, the wall boiling heat transfer model, and the workpiece immersion motion.

[0014] S3) During the workpiece quenching process, acquire temperature change information of multiple characteristic points on the workpiece surface in real time, and calculate the highest temperature on the workpiece surface at the current moment. With the lowest temperature The maximum temperature difference of the workpiece was obtained. .

[0015] S4) During the workpiece quenching process, acquire real-time information on the heat transfer coefficient changes at multiple characteristic points on the workpiece surface, and calculate the heat transfer coefficient uniformity index. ,in The spatial standard deviation of the heat transfer coefficient on the workpiece surface. This represents the spatial average value of the heat transfer coefficient on the workpiece surface.

[0016] S5) Output the vapor film discharge volume and pressure drop power consumption coefficient in real time during the workpiece quenching process, and calculate the vapor film discharge efficiency. ,in The total volume of the vapor membrane. This represents the volume of the retained vapor film.

[0017] S6) The results obtained from steps S3) to S5) , and To optimize the objective, a discrete optimization strategy assisted by a surrogate model is adopted to determine the optimal opening / closing state of each injection hole. The optimal fuel injection port switching mode is obtained.

[0018] Furthermore, the diameter of the injection hole is 10mm to 100mm, and the cross-sectional shape of the injection hole is circular.

[0019] Furthermore, step S2) specifically includes the following sub-steps:

[0020] S2.1) The Euler multiphase flow control equations are solved using the finite volume method, including the continuity equation, momentum equation, and energy equation for each phase. The interface momentum migration term in the momentum equation considers the coupling effect of liquid phase drag, lift, and wall lubrication force.

[0021] S2.2) Using the RPI wall boiling model or the Leidenfrost temperature model, the working surface is divided into film boiling zone, transition boiling zone, nucleus boiling zone, and natural convection zone based on the wall superheat, and the heat flux density and heat transfer coefficient are determined. The model also considers the upward migration of the vapor film under buoyancy and the mechanical stripping coupling effect of the oil injection jet on the vapor film.

[0022] S2.3) Establish the heat conduction equation of the solid domain of the workpiece, and take the continuity of heat flux density between the workpiece and the quenching medium as the coupling condition to realize the conjugate heat transfer solution of the fluid domain and the solid domain.

[0023] S2.4) Overlapping mesh or dynamic mesh technology is used to simulate the transient immersion process of the workpiece from air into the quenching oil. Local mesh refinement is performed on the contact interface between the workpiece and the quenching medium, as well as the area near the oil injection holes, and settings are set to meet the requirements. The boundary layer mesh was used to resolve the viscous sublayer. Simulation calculations were submitted after mesh independence verification was completed.

[0024] Furthermore, step S6) specifically includes the following sub-steps:

[0025] S6.1) Sensitivity Analysis: Perform independent CFD simulations of each injection orifice in both open and closed states, and calculate the impact of each orifice's opening / closing state on the following parameters. The contribution of changes in the value of the injection hole was used to select a set of candidate injection holes that had a significant impact on uniformity.

[0026] S6.2) Sample Acquisition: A Latin hypercube sampling method is used on the candidate injection hole set, treating each hole's open / closed state as either 0 or 1, generating several on / off mode samples. CFD simulation is then performed on each sample to obtain the corresponding... , and .

[0027] S6.3) Proxy Model Construction: Using the fuel injection port switch state vector as input features, a discrete proxy model is constructed using methods such as random forest for fast prediction. , and .

[0028] S6.4) Optimization Solution: Use a binary genetic algorithm to perform multi-objective optimization iterations on the surrogate model to minimize and The goal is to constrain the upper limit of the number of open fuel injection holes until the iteration converges and the optimal fuel injection hole switching mode is obtained.

[0029] S6.5) CFD Verification and Storage: Perform a complete CFD simulation verification on the obtained optimal switching mode. After successful verification, store the optimal list of open injection port numbers in the controller for use in setting and recalling the oil valve assembly before actual production.

[0030] Furthermore, the controller stores multiple optimal oil injection hole switching modes corresponding to different workpiece types, and can call the corresponding switching mode according to the workpiece type when switching production, so as to realize flexible production.

[0031] The technical effects of this invention are beyond doubt:

[0032] A. By pre-setting multiple independent and controllable oil injection holes on the sidewall of the quenching oil bath, and using CFD simulation to accurately identify the heat zone distribution and vapor film retention area on the workpiece surface, the heat transfer coefficient uniformity index is used. and maximum temperature difference To optimize the target, the optimal switching mode of the oil injection hole was determined to achieve targeted and enhanced cooling of the hot zone and vapor film retention zone of the workpiece, effectively eliminating the problem of insufficient local cooling caused by vapor film retention, thereby significantly reducing the residual stress and deformation of the workpiece after quenching and improving the comprehensive mechanical properties of the workpiece.

[0033] B. A multi-physics coupled CFD model covering Euler multiphase flow, wall boiling heat transfer and solid heat conduction of workpiece was established. The coupled effect of vapor film buoyancy migration and mechanical stripping of oil jet was also considered. This model can accurately predict the non-uniform distribution of transient heat transfer coefficient on the workpiece surface, providing a reliable quantitative basis for oil tank structure optimization. It completely replaces the experience-based trial production method, significantly shortens the design cycle and reduces R&D costs.

[0034] C. A four-stage strategy is adopted, which includes sensitivity analysis to pre-screen candidate oil holes, Latin hypercube sampling to construct a training set, random forest surrogate model for fast prediction, and binary genetic algorithm for global optimization. This strategy achieves efficient search for the global optimal switching mode within a limited number of CFD simulations, significantly reducing computational cost.

[0035] D. The optimized oil injection hole switching mode only opens the oil holes that significantly contribute to the cooling uniformity and closes the oil holes that contribute little to the cooling of the target area. Under the premise of ensuring quenching quality, it effectively reduces the ineffective circulation flow, reduces the power consumption of the oil pump, and improves the overall system energy efficiency.

[0036] E. For workpieces with different geometries and heat treatment requirements, the corresponding optimal oil injection hole switching mode can be calculated offline and stored in the controller. When switching production, it is only necessary to call the corresponding mode and complete the oil valve group setting. There is no need to redesign the oil tank hardware structure, which significantly improves the adaptability of quenching equipment to multi-variety production. Attached Figure Description

[0037] Figure 1 The flowchart shows the method for optimizing the structure of a quenching oil tank based on CFD simulation.

[0038] Figure 2 A schematic diagram of an immersion quenching model for a certain workpiece;

[0039] Figure 3 This is a schematic diagram of an improved immersion quenching oil bath. Detailed Implementation

[0040] The present invention will be further described below with reference to embodiments, but it should not be construed that the scope of the present invention is limited to the following embodiments. Various substitutions and modifications made based on ordinary technical knowledge and common practices in the art without departing from the above-described technical concept of the present invention should be included within the scope of protection of the present invention.

[0041] Example 1:

[0042] See Figure 1 This embodiment provides a method for optimizing the structure of a quenching oil tank based on CFD simulation. The quenching oil tank is an immersion quenching oil tank, including a tank body, several oil spray holes disposed on the side wall of the tank body, an oil delivery pipeline independently connected to each oil spray hole, and an oil valve with independent opening / closing function installed on each oil delivery pipeline. The several oil spray holes are arranged along the normal direction of the side wall of the oil tank in a regular array, and each oil spray hole has an independent number i=1,2,3,…,N. The opening / closing state of the oil spray hole is determined by the state variable s. i ={0,1} describes the state, where 0 represents off and 1 represents on. The method includes the following steps:

[0043] S1) See also Figure 2 and Figure 3 Based on the actual dimensions of the workpiece and the quenching oil bath, a three-dimensional parametric geometric model of the workpiece and the quenching oil bath is established, and a pre-defined normal direction on the side wall of the oil bath is set. The oil injection holes are arranged in a regular array to achieve an improved design of the quenching oil tank.

[0044] S2) Based on the improved design of the quenching oil tank, a CFD simulation calculation model was established based on the principles of computational fluid dynamics, which includes the Euler multiphase flow model, the wall boiling heat transfer model, and the workpiece immersion motion.

[0045] S3) During the workpiece quenching process, acquire temperature change information of multiple characteristic points on the workpiece surface in real time, and calculate the highest temperature on the workpiece surface at the current moment. With the lowest temperature The maximum temperature difference of the workpiece was obtained. .

[0046] S4) During the workpiece quenching process, acquire real-time information on the heat transfer coefficient changes at multiple characteristic points on the workpiece surface, and calculate the heat transfer coefficient uniformity index. ,in The spatial standard deviation of the heat transfer coefficient on the workpiece surface. This represents the spatial average value of the heat transfer coefficient on the workpiece surface.

[0047] S5) Output the vapor film discharge volume and pressure drop power consumption coefficient in real time during the workpiece quenching process, and calculate the vapor film discharge efficiency. ,in The total volume of the vapor membrane. This represents the volume of the retained vapor film.

[0048] S6) The results obtained from steps S3) to S5) , and To optimize the objective, a discrete optimization strategy assisted by a surrogate model is adopted to determine the optimal opening / closing state of each injection hole. The optimal fuel injection port switching mode is obtained.

[0049] Thus, a CFD simulation-based method for optimizing the structure of a quenching oil bath was completed. This method pre-defines multiple independently switchable oil injection holes on the sidewall of the oil bath, establishes a CFD model incorporating VOF multiphase flow, boiling heat transfer, and workpiece motion, and uses a discrete optimization algorithm to determine the optimal opening / closing state of each oil hole, targeting the non-uniformity of the heat transfer coefficient distribution. The optimized oil bath has only some oil holes open, and the distribution of the open oil holes is targeted to enhance cooling in the hot zones and vapor film retention areas of the workpiece.

[0050] Example 2:

[0051] The main content of this embodiment is the same as that of embodiment 1, wherein the diameter of the oil injection hole is 10mm to 100mm and the cross-sectional shape of the oil injection hole is circular.

[0052] Example 3:

[0053] The main content of this embodiment is the same as that of embodiment 1 or 2, wherein step S2) specifically includes the following sub-steps:

[0054] S2.1) The Euler multiphase flow control equations are solved using the finite volume method, including the continuity equation, momentum equation, and energy equation for each phase. The interface momentum migration term in the momentum equation considers the coupling effect of liquid phase drag, lift, and wall lubrication force.

[0055] S2.2) Using the RPI wall boiling model or the Leidenfrost temperature model, the working surface is divided into film boiling zone, transition boiling zone, nucleus boiling zone, and natural convection zone based on the wall superheat, and the heat flux density and heat transfer coefficient are determined. The model also considers the upward migration of the vapor film under buoyancy and the mechanical stripping coupling effect of the oil injection jet on the vapor film.

[0056] S2.3) Establish the heat conduction equation of the solid domain of the workpiece, and take the continuity of heat flux density between the workpiece and the quenching medium as the coupling condition to realize the conjugate heat transfer solution of the fluid domain and the solid domain.

[0057] S2.4) Overlapping mesh or dynamic mesh technology is used to simulate the transient immersion process of the workpiece from air into the quenching oil. Local mesh refinement is performed on the contact interface between the workpiece and the quenching medium, as well as the area near the oil injection holes, and settings are set to meet the requirements. The boundary layer mesh was used to resolve the viscous sublayer. Simulation calculations were submitted after mesh independence verification was completed.

[0058] Example 4:

[0059] The main content of this embodiment is the same as any one of embodiments 1 to 3, wherein step S6) specifically includes the following sub-steps:

[0060] S6.1) Sensitivity Analysis: Perform independent CFD simulations of each injection orifice in both open and closed states, and calculate the impact of each orifice's opening / closing state on the following parameters. The contribution of changes in the value of the injection hole was used to select a set of candidate injection holes that had a significant impact on uniformity.

[0061] S6.2) Sample Acquisition: A Latin hypercube sampling method is used on the candidate injection hole set, treating each hole's open / closed state as either 0 or 1, generating several on / off mode samples. CFD simulation is then performed on each sample to obtain the corresponding... , and .

[0062] S6.3) Proxy Model Construction: Using the fuel injection port switch state vector as input features, a discrete proxy model is constructed using methods such as random forest for fast prediction. , and .

[0063] S6.4) Optimization Solution: Use a binary genetic algorithm to perform multi-objective optimization iterations on the surrogate model to minimize and The goal is to constrain the upper limit of the number of open fuel injection holes until the iteration converges and the optimal fuel injection hole switching mode is obtained.

[0064] S6.5) CFD Verification and Storage: Perform a complete CFD simulation verification on the obtained optimal switching mode. After successful verification, store the optimal list of open injection port numbers in the controller for use in setting and recalling the oil valve assembly before actual production.

[0065] Example 5:

[0066] The main content of this embodiment is the same as any one of embodiments 1 to 4. The controller stores multiple sets of optimal oil injection hole switching modes for different workpiece types. The corresponding switching mode can be called according to the workpiece type when switching production to achieve flexible production.

Claims

1. A method for optimizing the structure of a quenching oil tank based on CFD simulation, characterized in that, The quenching oil tank is an immersion quenching oil tank, comprising a tank body, several oil spray holes disposed on the side wall of the tank body, an oil supply pipeline independently connected to each oil spray hole, and an oil valve with an independent opening / closing function installed on each oil supply pipeline; the several oil spray holes are arranged along the normal direction of the side wall of the oil tank in a regular array, and each oil spray hole has an independent number i=1,2,3,…,N; the opening / closing state of the oil spray hole is determined by the state variable s. i ={0,1} describes a value where 0 indicates off and 1 indicates on; The method includes the following steps: S1) Based on the actual dimensions of the workpiece and the quenching oil bath, establish a three-dimensional parametric geometric model of the workpiece and the quenching oil bath, and pre-determine the normal direction of the oil bath sidewall. The oil injection holes are arranged in a regular array to achieve an improved design of the quenching oil tank; S2) Based on the improved design of the quenching oil tank, a CFD simulation calculation model including the Euler multiphase flow model, the wall boiling heat transfer model and the workpiece immersion motion is established based on the principles of computational fluid dynamics. S3) During the workpiece quenching process, acquire temperature change information of multiple characteristic points on the workpiece surface in real time, and calculate the highest temperature on the workpiece surface at the current moment. With the lowest temperature The maximum temperature difference of the workpiece was obtained. ; S4) During the workpiece quenching process, acquire real-time information on the heat transfer coefficient changes at multiple characteristic points on the workpiece surface, and calculate the heat transfer coefficient uniformity index. ,in The spatial standard deviation of the heat transfer coefficient on the workpiece surface. This represents the spatial average value of the heat transfer coefficient on the workpiece surface. S5) Output the vapor film discharge volume and pressure drop power consumption coefficient in real time during the workpiece quenching process, and calculate the vapor film discharge efficiency. ,in The total volume of the vapor membrane. This refers to the volume of the retained vapor film. S6) Based on the results obtained in steps S3) to S5) , and To optimize the objective, a discrete optimization strategy assisted by a surrogate model is adopted to determine the optimal opening / closing state of each injection hole. The optimal fuel injection port switching mode is obtained.

2. The method for optimizing the structure of a quenching oil tank based on CFD simulation according to claim 1, characterized in that: The diameter of the injection hole is 10mm to 100mm, and the cross-sectional shape of the injection hole is circular.

3. The method for optimizing the structure of a quenching oil tank based on CFD simulation according to claim 1, characterized in that, Step S2) specifically includes the following sub-steps: S2.1) The Euler multiphase flow control equations are solved using the finite volume method, including the continuity equation, momentum equation and energy equation for each phase; the interface momentum migration term of the momentum equation considers the coupling effect of liquid phase drag, lift and wall lubrication force. S2.2) The RPI wall boiling model or the Leidenfrost temperature model is used to divide the film boiling zone, transition boiling zone, nucleus boiling zone and natural convection zone according to the wall superheat, and to determine the heat flux density and heat transfer coefficient of the workpiece surface; the model also considers the upward migration of the vapor film under the action of buoyancy and the mechanical peeling coupling effect of the oil injection jet on the vapor film. S2.3) Establish the heat conduction equation of the solid domain of the workpiece, and take the continuity of heat flux density between the workpiece and the quenching medium as the coupling condition to realize the conjugate heat transfer solution of the fluid domain and the solid domain. S2.4) Overlapping mesh or moving mesh technology is used to simulate the transient immersion process of a workpiece from air into quenching oil; The mesh at the interface between the workpiece and the quenching medium, as well as the area near the oil spray holes, is locally refined, and settings are set to meet the requirements. Boundary layer mesh to resolve viscous sublayers; Submit the simulation calculation after completing the mesh independence verification.

4. The method for optimizing the structure of a quenching oil tank based on CFD simulation according to claim 1, characterized in that, Step S6) specifically includes the following sub-steps: S6.1) Sensitivity Analysis: Perform independent CFD simulations of each injection orifice in both open and closed states for comparison, and calculate the impact of each orifice's opening / closing state on... The contribution of the change was used to screen out the candidate set of injection holes that have a significant impact on uniformity; S6.2) Sample Acquisition: A Latin hypercube sampling method is used on the candidate injection hole set, treating each hole's open / closed state as either 0 or 1, generating several on / off mode samples. CFD simulation is then performed on each sample to obtain the corresponding... , and ; S6.3) Proxy Model Construction: Using the fuel injection port switch state vector as input features, a discrete proxy model is constructed using methods such as random forest for fast prediction. , and ; S6.4) Optimization Solution: Use a binary genetic algorithm to perform multi-objective optimization iterations on the surrogate model to minimize and The goal is to constrain the upper limit of the number of open fuel injection holes until the iteration converges and the optimal fuel injection hole switching mode is obtained. S6.5) CFD Verification and Storage: Perform a complete CFD simulation verification on the obtained optimal switching mode; after verification, store the optimal list of open injection port numbers in the controller for use in setting and calling the oil valve group before actual production.

5. The method for optimizing the structure of a quenching oil tank based on CFD simulation according to claim 4, characterized in that: The controller stores multiple optimal oil injection hole switching modes for different workpiece types. When switching production, the corresponding switching mode can be called according to the workpiece type to achieve flexible production.