Method and system for performance optimization of low flow resistance micro-channel evaporator based on flow field simulation

Abstract

The application discloses a low-flow-resistance micro-channel evaporator performance optimization method and system based on flow field simulation, and relates to the technical field of data processing.The method comprises the following steps: constructing a micro-channel evaporator fluid calculation domain and generating multiple sets of geometric candidate schemes; collecting convergence trajectory data containing residual variation and vorticity evolution through flow field simulation; establishing a convergence stability criterion based on the convergence trajectory data as the only data source; and screening by taking the convergence stability criterion as a preconditioning constraint in combination with pressure drop and heat exchange performance, and outputting a target geometric structure with low flow resistance, high heat exchange and stable operation.The application solves the technical problem that the existing micro-channel evaporator optimization method cannot identify pseudo-optimal structures with unstable convergence, resulting in deviation between simulation optimization results and actual performance, and achieves the technical effects of establishing a stability criterion based on convergence trajectory data, screening out unstable structures, and improving the reliability of simulation optimization results and the consistency of physical performance.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for optimizing the performance of low-flow-resistance microchannel evaporators based on flow field simulation. Background Technology

[0002] Current optimization of microchannel evaporator performance is largely based on steady-state flow field simulation results, directly targeting the lowest pressure drop and strongest heat transfer as the sole optimization objectives. However, spurious convergence often occurs in two-phase flow simulations: some candidate schemes, while exhibiting excellent steady-state indices, suffer from long-tailed oscillations or vortex hysteresis during their iteration process, reflecting instabilities such as flow separation or high shear dissipation within the flow field. Such structures are extremely sensitive to manufacturing tolerances and operating condition fluctuations, leading to significant deviations between simulation data and actual performance. Currently, there is a lack of effective means to use the solver convergence process itself as a basis for structural selection, making it difficult to predict and eliminate such high-flow-resistance structures that appear numerically promising but are not robust enough for construction during the design phase. Summary of the Invention

[0003] This application provides a method and system for optimizing the performance of low flow resistance microchannel evaporators based on flow field simulation, which solves the technical problem that existing microchannel evaporator optimization methods cannot identify pseudo-optimal structures with unstable convergence, resulting in deviations between simulation optimization results and actual performance.

[0004] The first aspect of this application provides a method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation, the method comprising: A fluid computational domain for a microchannel evaporator is constructed, and multiple candidate schemes containing channel cross-sectional geometric features are generated within this domain. Flow field simulation is performed on each candidate scheme, and convergence trajectory data is collected and stored synchronously during the iterative solution process. This convergence trajectory data includes residual change information and vortex evolution information. The convergence trajectory data is used as the sole data source for evaluating the flow stability of the candidate schemes, and a convergence stability criterion is constructed. With pressure drop reduction and heat transfer enhancement as optimization objectives, and the convergence stability criterion as a prerequisite hard constraint, the candidate schemes are screened, and only those schemes that meet the convergence stability criterion and have a smooth flow field convergence path are retained as the target geometry output.

[0005] The second aspect of this application provides a performance optimization system for low-flow-resistance microchannel evaporators based on flow field simulation. The system includes: a candidate scheme generation module for constructing the fluid computation domain of the microchannel evaporator and generating multiple sets of candidate schemes containing channel cross-sectional geometric features within the fluid computation domain; a flow field simulation calculation module for performing flow field simulation calculations for each set of candidate schemes, synchronously collecting and storing convergence trajectory data during iterative solution, the convergence trajectory data including residual change information and vorticity evolution information; a stability criterion construction module for using the convergence trajectory data as the sole data source for evaluating the flow stability of candidate schemes and constructing a convergence stability criterion; and a candidate scheme optimization module for using pressure drop reduction and heat transfer enhancement as optimization objectives, with the convergence stability criterion as a prerequisite hard constraint, to screen the candidate schemes, retaining only those that meet the convergence stability criterion and have a smooth flow field convergence path as the target geometry output.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application provides a method and system for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation, which relates to the field of data processing technology. By using the convergence trajectory data in the flow field simulation iteration as the sole data source for evaluating flow stability, a convergence stability criterion is constructed and set as a pre-imposed hard constraint. Under the optimization objectives of reducing pressure drop and enhancing heat transfer, unstable structures with residual oscillation and vortex hysteresis are pre-screened, so that the output target geometry has both low flow resistance, high heat transfer, and robustness in actual operation. This solves the technical problem that existing microchannel evaporator optimization methods cannot identify convergently unstable pseudo-optimal structures, leading to deviations between simulation optimization results and actual performance. It achieves the technical effect of constructing a stability criterion based on convergence trajectory data, screening out unstable structures, and improving the reliability of simulation optimization results and the consistency of actual performance. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0008] Figure 1 A schematic flowchart of a low-flow-resistance microchannel evaporator performance optimization method based on flow field simulation provided in this application embodiment; Figure 2 A schematic diagram of the performance optimization system for a low-flow-resistance microchannel evaporator based on flow field simulation provided in this application embodiment.

[0009] Figure labeling: Candidate scheme generation module 11, flow field simulation calculation module 12, stability criterion construction module 13, candidate scheme optimization module 14. Detailed Implementation

[0010] This application provides a method and system for optimizing the performance of low flow resistance microchannel evaporators based on flow field simulation, which solves the technical problem that existing microchannel evaporator optimization methods cannot identify pseudo-optimal structures with unstable convergence, resulting in deviations between simulation optimization results and actual performance.

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0012] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0013] Example 1, as Figure 1 As shown, this application provides a method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation. This method includes: P10: Construct the fluid computation domain of the microchannel evaporator and generate multiple sets of candidate schemes containing channel cross-sectional geometric features within the fluid computation domain.

[0014] Specifically, firstly, based on the structural form of the target microchannel evaporator and the flow characteristics of the working medium, a fluid computational domain for numerical flow field analysis is established. This domain describes the spatial region used to describe the flow, heat transfer, and energy exchange processes of the working fluid within the microchannel evaporator. It includes at least the fluid inlet region, the microchannel body region, the fluid outlet region, and the wall boundary regions related to the heat exchange process. When constructing the fluid computational domain, a three-dimensional geometric model is established based on the actual evaporator structural dimensions, and the solid and fluid regions are separated, retaining only the spatial region occupied by the actual flow of the working medium. This allows for subsequent solution of the flow state based on the governing equations. The boundary conditions in the fluid computational domain are used to define the fluid motion and heat exchange process, such as setting inlet flow rate, inlet pressure, inlet temperature, outlet pressure, and channel wall thermal boundary conditions, ensuring that the established computational domain reflects the flow and heat exchange state of the microchannel evaporator under actual operating conditions.

[0015] After constructing the fluid computational domain, multiple candidate schemes containing channel cross-sectional geometric features are generated within the domain. These are microchannel structure models to be evaluated, formed by combinations of different channel structural parameters, and used to construct the structural sample space in the subsequent performance optimization process. Specifically, by parametrically describing the shape and size parameters of the microchannel cross-section, geometric features affecting flow resistance and heat transfer performance are used as adjustable design variables, such as the width, height, hydraulic diameter, aspect ratio, cross-sectional shape, flow channel turning radius, and local expansion / contraction structure dimensions of the channel cross-section. The channel cross-sectional geometric features refer to the shape and size attributes of the channel cross-section perpendicular to the mainstream fluid direction. These characteristics determine the contact area between the fluid and the channel wall, the flow cross-sectional area, and the local velocity distribution, thereby affecting pressure loss, flow separation degree, and heat transfer enhancement effect.

[0016] Furthermore, based on the preset parameter value range and combination rules, the aforementioned geometric design variables are arranged and combined or sampled to generate multiple candidate structural schemes, so that different candidate schemes correspond to different combinations of channel cross-sectional geometric parameters. For example, multiple sets of candidate geometric structures can be generated under the conditions of meeting manufacturing constraints and fluid flow requirements through orthogonal experimental design, Latin hypercube sampling, or parameter traversal, to ensure that the candidate schemes cover different flow characteristic regions. Among them, the manufacturing constraints are used to limit the candidate structures to meet the actual processing feasibility conditions, including minimum channel size restrictions, structural continuity restrictions, and wall thickness safety requirements; the fluid flow requirements are used to avoid fluid blockage due to excessively small channel size or abnormal flow phenomena due to abrupt changes in cross-section.

[0017] By using the above method, a set of candidate schemes containing multiple different channel cross-sectional geometric features is established, so that each candidate scheme corresponds to an independent fluid computational domain model, and provides a basic data source for subsequent flow field simulation calculations for different structures, obtaining convergence trajectory data, and screening based on convergence stability criteria.

[0018] P20: Perform flow field simulation calculations for each group of candidate schemes. During the iterative solution process, collect and store convergence trajectory data simultaneously. The convergence trajectory data includes residual change information and vortex evolution information.

[0019] Furthermore, step P20 in this embodiment of the application also includes: P21: In each iteration step of the flow field simulation calculation, read the residual vector output by the solver, extract the absolute value sequence of the residuals of the continuity equation and the momentum equation, and use it as the residual change information; P22: In the same iteration step, extract the vorticity tensor of the entire flow field, calculate the spatial distribution of the local vorticity dissipation intensity, and record its evolution process with the iteration step size, as the vorticity evolution information; P23: Establish a data mapping relationship indexed by the iteration step size, align and store the residual change information and the vorticity evolution information with the same iteration step size, and encapsulate them into structured convergence trajectory data.

[0020] Optionally, for each generated group of microchannel evaporator candidate structures, a corresponding numerical calculation model of the flow field is established, and iterative solutions are performed on each candidate scheme based on preset fluid control equations. The flow field simulation calculation refers to the process of discretizing the fluid motion state inside the microchannel using computational fluid dynamics methods. By numerically solving the continuity equation, momentum conservation equation, and energy conservation equation, the velocity field, pressure field, temperature field, and local flow characteristics under different geometric conditions are obtained. Unlike existing technologies that only focus on the final output results of simulation calculations, such as pressure drop and heat transfer coefficient, this application simultaneously acquires solution state change data during the iterative calculation process of the solver to form dynamic convergence trajectory data reflecting the entire flow field solution process.

[0021] Furthermore, in each iteration step of the flow field simulation calculation, the residual vector output by the solver is read, and the absolute value sequence of the residuals of the continuity equation and the momentum equation is extracted as the residual change information. Specifically, the iteration step refers to the process by which the solver performs a complete update calculation on the discretized control equations. After each iteration calculation, the solver outputs a residual vector to characterize the degree of deviation between the current calculation state and the target convergence state. The residual vector is used to describe the quantitative information of the degree of non-satisfaction of each control equation during the numerical solution process, including the mass conservation residual corresponding to the continuity equation and the velocity field balance residual corresponding to the momentum equation. By reading the residual vector corresponding to each iteration step and extracting the absolute values ​​of the residuals of the continuity equation and the momentum equation respectively, a residual sequence that varies with the iteration step size is formed. This residual sequence can reflect the numerical stability trend of the candidate structure during the solution process. For example, a continuous decrease in residuals indicates that the flow field solution is gradually stabilizing, while repeated oscillations or abnormal increases in residuals indicate that the structure may have complex flow disturbances or numerical instability.

[0022] Furthermore, within the same iteration step, the vorticity tensor of the entire flow field is extracted, the spatial distribution of local vorticity dissipation intensity is calculated, and its evolution with the iteration step size is recorded as the vorticity evolution information. Specifically, the vorticity tensor is used to describe the spatial variation of the rotational motion characteristics of fluid micro-elements. It is obtained by calculating the spatial gradient of the velocity field and is used to characterize the vortex motion state inside the microchannel caused by cross-sectional changes, flow channel turning, and local structural disturbances. In each iteration step, the vorticity tensor of each computational unit within the flow field region is calculated based on the velocity field data obtained in the current iteration, and the local vorticity dissipation intensity is further calculated according to the vorticity intensity, vorticity gradient change, or vorticity dissipation-related parameters to obtain the spatial distribution data under the corresponding iteration state.

[0023] Local vorticity dissipation intensity is an evaluation parameter characterizing the degree of energy loss caused by fluid rotation. It reflects the energy dissipation trend within the microchannel due to flow separation, secondary flow formation, and local high-speed shear regions. By continuously recording the distribution of local vorticity dissipation intensity at different iteration steps, the dynamic evolution of the vorticity structure from its initial calculated state to its steady state can be obtained, thus distinguishing potentially unstable structures that cannot be identified solely from the final flow field results. For example, although some candidate structures may ultimately achieve low pressure drop results, if they exhibit continuous vorticity enhancement or expansion of local dissipation regions during the iteration process, it indicates that the structure carries a risk of flow instability.

[0024] Furthermore, a data mapping relationship indexed by the iteration step size is established, and the residual change information and the vortex evolution information are correlated with each other according to the same iteration step size, and stored synchronously and aligned to form structured convergence trajectory data. Specifically, the iteration number in the solution process is used as a unified time index, and the residuals of the continuity equation, momentum equation, and vortex dissipation intensity distribution corresponding to the i-th iteration are mapped one-to-one, so that different types of solution state data are under the same iteration scale, thereby forming a multi-dimensional data structure containing iteration step, residual state, and vortex state.

[0025] The structured convergence trajectory data refers to the solution process data set organized and stored according to a preset data format. It includes at least iteration step numbers, residual change sequences, and vorticity evolution sequences, used to fully describe the dynamic changes of candidate schemes from the initial solution state to the convergence state. Through this method, this application no longer evaluates candidate structures solely based on the steady-state performance results after the flow field simulation, but retains the dynamic behavioral characteristics during the solution process. This provides a unique data source for subsequent screening of candidate schemes based on convergence stability criteria, thereby enabling the identification and elimination of pseudo-optimal structures that exhibit low flow resistance but lack actual flow stability.

[0026] P30: Using the convergence trajectory data as the sole data source for evaluating the flow stability of candidate schemes, a convergence stability criterion is constructed.

[0027] Furthermore, step P30 in this embodiment of the application also includes: P31: Traverse the convergence trajectory data and identify the continuous iteration interval after the residual value enters the preset error band in the residual change information; P32: Within the continuous iteration interval, statistically analyze the fluctuation amplitude and fluctuation frequency of the residual value to generate a first evaluation factor representing the stability of the convergence process; P33: Within the continuous iteration interval, extract the turning point of vortex dissipation intensity in the vortex evolution information, calculate the step size from the iteration start point to the turning point, and generate a second evaluation factor characterizing the hysteresis of the flow state switching; P34: Weight and fuse the first evaluation factor and the second evaluation factor to generate a quantitative score as a convergence stability criterion.

[0028] It should be understood that after collecting the convergence trajectory data for each candidate scheme, this application no longer uses the traditional method of evaluating candidate structures based on the final pressure drop, heat transfer coefficient, or a single steady-state flow field parameter. Instead, it uses the convergence trajectory data as the sole evaluation criterion, and establishes a convergence stability criterion to measure the flow stability of candidate schemes by analyzing the residual change trend and vortex dissipation evolution characteristics during the solution process. The convergence stability criterion refers to an index system that quantitatively evaluates whether the flow field state corresponding to the candidate structure has stable convergence characteristics based on the dynamic changes during the numerical solution process. It is used to identify structural schemes that meet the convergence conditions in the final calculation results but exhibit abnormal oscillations, delayed flow state switching, or continuously enhanced local dissipation during the solution process.

[0029] For example, the convergence trajectory data is first traversed to identify the continuous iteration intervals after the residual values ​​in the residual change information enter the preset error band. Specifically, the convergence trajectory data corresponding to each group of candidate solutions is traversed and read. Based on the residual sequence of the continuity equation and the residual sequence of the momentum equation in the residual change information, it is determined whether the residual values ​​corresponding to each iteration step enter the preset error band. Here, the preset error band refers to the residual range used to determine whether the numerical solution is close to a stable convergence state. It can be preset according to the solution accuracy requirements, fluid model type, and computational resource conditions. For example, the region where the residual value is less than a preset threshold is set as a stable convergence candidate interval. When it is detected that the residual value remains within the preset error band for multiple consecutive iteration steps, the corresponding continuous iteration process is determined as the residual stable interval.

[0030] The continuous iteration interval refers to a set of continuous iteration steps in the residual change information that satisfy the stable convergence condition. It reflects the computational stage after the solver has approached a stable solution state. By identifying this interval, we can avoid mistaking drastic changes caused by initial condition disturbances, mesh adjustments, or flow field initialization in the initial solution stage for structural stability characteristics, and make subsequent evaluations focus more on the true dynamic behavior after the flow field tends to stabilize.

[0031] Next, within the continuous iteration interval, the fluctuation amplitude and frequency of the residual values ​​are statistically analyzed to generate the first evaluation factor for the stationarity of the aggregation convergence process. Specifically, after determining the residual stability interval, the residual change information within this interval is statistically analyzed to calculate the amplitude and number of changes in the residual values ​​during the continuous iteration process. The fluctuation amplitude refers to the degree of difference between the maximum and minimum residual values ​​during continuous iteration, reflecting the degree to which the residual deviates from the stable trend; the fluctuation frequency refers to the number of times the residual increases, decreases, or repeatedly changes within a unit iteration step, reflecting the degree of oscillation during the solution process.

[0032] For example, if the residual sequence corresponding to a candidate scheme maintains a low amplitude change after entering the preset error band, and there is no frequent increase or decrease, it indicates that the flow field solution process corresponding to the scheme has good stability; conversely, if the residual reaches the convergence threshold, but continues to fluctuate significantly in subsequent iterations, it indicates that the candidate scheme may have problems with enhanced flow field disturbance or numerical solution instability.

[0033] Furthermore, the fluctuation amplitude and frequency are normalized, and a first evaluation factor is generated according to a preset calculation rule to quantify the stability of the residual convergence process. A higher first evaluation factor indicates a smoother residual change and a more stable solution process. Within the continuous iteration interval, the turning point of vortex dissipation intensity in the vortex evolution information is extracted, and the step size from the iteration start point to the turning point is calculated to generate a second evaluation factor characterizing the hysteresis of flow state switching. Specifically, within the iteration range corresponding to the residual stability interval, the vortex evolution information is analyzed to obtain the changing trend of local vortex dissipation intensity with the iteration step size, and the locations where significant changes in direction occur are identified. The turning point of vortex dissipation intensity refers to the location in the vortex dissipation intensity change curve where the trend changes from increasing to decreasing or vice versa, reflecting the key node in the transition of the internal vortex structure from an unstable to a stable state.

[0034] After determining the turning point, the difference between the iteration step corresponding to that point and the initial iteration step of the continuous iteration interval is calculated to obtain the step span. The step span characterizes the iteration time scale required for the flow structure within the flow field to complete its state adjustment. When a candidate structure causes strong separation, backflow, or local energy dissipation in the fluid, the eddy current dissipation intensity may require a longer iteration process to stabilize, corresponding to a larger step span; while structures with more reasonable flow organization can usually complete the flow state adjustment quickly, corresponding to a smaller step span.

[0035] Based on the above step size range, a normalization process is performed to generate a second evaluation factor, which is used to describe the degree of hysteresis during the flow state switching process corresponding to the candidate scheme. The second evaluation factor can supplement the problem that residual evaluation cannot directly reflect the changes in flow structure, so that the convergence stability criterion considers both the numerical solution stability and the evolution characteristics of the flow physical state.

[0036] Next, the first and second evaluation factors are weighted and fused to generate a quantitative score as a convergence stability criterion. Specifically, the first evaluation factor, which characterizes the stationarity of the solution process, and the second evaluation factor, which characterizes the hysteresis of flow state switching, are normalized and fused according to preset weights to obtain a comprehensive quantitative score for evaluating the stability of candidate schemes. Weighted fusion refers to assigning weight coefficients to the first and second evaluation factors based on the degree of influence of different evaluation factors on flow field stability, and generating a single evaluation index through mathematical combination.

[0037] By employing the above method, the numerical convergence stationarity exhibited during residual change and the flow physical stability exhibited during vorticity evolution are jointly evaluated to form a comprehensive stability criterion based on convergence trajectory data. This criterion does not depend on the final pressure drop or heat transfer performance results obtained by the candidate structure, but only on the dynamic trajectory data during the solution process to judge the structural reliability. This allows for the early screening of candidate schemes with flow separation tendencies, abnormal vorticity dissipation, and unstable solution paths before performance optimization, providing stable and reliable screening constraints for subsequent optimization of low flow resistance microchannel evaporator structures.

[0038] Furthermore, step P30 in this embodiment of the application also includes: P35: If the convergence trajectory data exhibits long-tailed oscillations or non-monotonic convergence characteristics, then the candidate scheme is determined to have an inherent tendency to induce flow separation in actual operation.

[0039] Optionally, based on the constructed convergence stability criterion, further feature identification is performed on the overall change trend in the convergence trajectory data to determine whether there is a potential unstable flow mechanism in the flow field corresponding to the candidate scheme. Here, long-tailed oscillation refers to the phenomenon that, after the residual value has reached the preset error range, the residual change curve continues to show periodic increases and decreases within a long iteration interval. That is, although the solution process meets the apparent convergence condition, the system state fails to quickly stabilize to an equilibrium state. The non-monotonic convergence feature refers to the residual change information not showing a continuous downward trend, but rather exhibiting multiple rebounds, repeated fluctuations, or staged sudden increases during the iteration process. This feature reflects that the flow field state is subject to continuous disturbance or continuous adjustment of the local flow structure during the solution process.

[0040] Furthermore, when performing trend analysis on the convergence trajectory data, the residual sequences of the continuity equation and momentum equation in the residual change information, as well as the vortex dissipation intensity change sequence in the vortex evolution information, can be jointly judged. If it is detected that the residual sequence still has an oscillation interval exceeding a preset duration after entering the preset error band, or if the vortex dissipation intensity still undergoes periodic enhancement and decay changes during the stable convergence phase, then the flow field state of the corresponding candidate scheme is considered not to have reached a stable equilibrium state.

[0041] Flow separation refers to the phenomenon where, during fluid flow along the wall of a microchannel, changes in local pressure gradients, abrupt changes in channel cross-section, or alterations in flow direction cause a decrease in fluid velocity in the near-wall region and separation from the mainstream region. Flow separation in a microchannel evaporator is typically accompanied by the formation of local backflow regions, increased vorticity, and increased energy dissipation, leading to increased pressure loss, fluctuating heat transfer performance, and uneven flow distribution during actual operation. Therefore, this application identifies long-tailed oscillations and non-monotonic convergence characteristics in the convergence trajectory as indirect characterizations of flow separation tendency.

[0042] Furthermore, when the convergence trajectory data corresponding to a candidate scheme meets the aforementioned abnormal characteristic conditions, the candidate scheme is marked as a scheme with unqualified flow stability, and it is determined that it has an inherent tendency to induce flow separation under actual operating conditions. Here, the inherent tendency refers to the flow instability characteristics caused by the candidate scheme's own geometry, rather than accidental phenomena caused by single simulation parameter settings or random calculation errors. This judgment method utilizes dynamic evolution information during the solution process to identify potential unstable flow mechanisms within the structure in advance, avoiding the misjudgment of structures with low pressure drop appearances but insufficient actual operational reliability as preferred schemes when selecting solely based on final steady-state performance parameters.

[0043] P40: With pressure drop reduction and heat transfer enhancement as the optimization objectives, the convergence stability criterion is used as a prerequisite hard constraint to screen the candidate schemes. Only the candidate schemes that meet the convergence stability criterion and have a smooth flow field convergence path are retained as the target geometry output.

[0044] Furthermore, before applying the aforementioned prerequisite hard constraints, step P40 in this embodiment of the application further includes: P41: Apply a small geometric perturbation to the geometric model of the candidate scheme to simulate manufacturing tolerances, generating a perturbed geometric variant; P42: Re-execute the flow field simulation calculation for the perturbed geometric variant to obtain the perturbed convergence trajectory data; P43: Compare the perturbed convergence trajectory data with the baseline convergence trajectory data to extract the residual fluctuation pattern difference and vorticity turning point difference; P44: If the residual fluctuation pattern difference and vorticity turning point difference exceed the preset tolerance limit, the candidate scheme is determined to be unsuitable for manufacturing due to high flow field response sensitivity, and the candidate scheme is eliminated.

[0045] Specifically, after constructing the convergence stability criterion, the stability of each candidate scheme is first pre-screened using the criterion. Candidate schemes exhibiting abnormal oscillations, non-monotonic convergence, continuously increasing vorticity dissipation, or significant hysteresis in flow state switching during the flow field solution process are eliminated. Only candidate schemes that meet the stability constraints are retained for the performance optimization stage. The prerequisite hard constraints refer to the screening conditions that must be met before performing pressure drop reduction and heat transfer enhancement performance evaluation. That is, candidate schemes must first meet the flow field convergence stability requirements. If the threshold corresponding to the convergence stability criterion is not met, regardless of whether the final calculated pressure drop or heat transfer performance indicators are excellent, it will not be considered as an optimization candidate structure.

[0046] Furthermore, before applying the aforementioned pre-existing hard constraints, to further evaluate the adaptability of candidate solutions to actual processing errors, small geometric perturbations conforming to manufacturing process characteristics are introduced into the geometric model of the benchmark candidate solution to simulate structural dimensional deviations caused by factors such as processing accuracy limitations, material deformation, and surface processing errors during actual processing. These geometric perturbations refer to small-range random or regular changes to local geometric parameters affecting flow characteristics without altering the overall flow channel layout of the candidate solution. Examples of such changes include adjusting microchannel width, height, wall dimensions, corner radius, or local cross-sectional dimensions.

[0047] In practice, based on a preset manufacturing tolerance range, key geometric parameters in candidate designs can be positively offset, negatively offset, or subjected to combined perturbations, thereby generating multiple perturbed geometric variants. These geometric variants represent the dimensional changes that the same design structure may undergo during actual manufacturing. By re-simulating and analyzing the perturbed structure, it can be determined whether the candidate design possesses strong flow field robustness. Flow field robustness refers to the ability of a microchannel structure to maintain stable flow even with minor geometric deviations.

[0048] Next, for each set of perturbed geometric variants, the fluid computational domain is re-established under the same simulation conditions as the baseline candidate scheme, and the flow field iterative solution process is executed. During the solution process, residual change information and vorticity evolution information corresponding to the perturbed structure are collected simultaneously, and the corresponding perturbed convergence trajectory data are generated according to the data acquisition method in the above steps.

[0049] The converged trajectory data after perturbation shares the same data structure as the baseline converged trajectory data, including the same iteration step index, residual parameter type, and vorticity evolution description method. This enables direct comparison of flow field response behavior under different geometric conditions. In this way, the dynamic response characteristics of candidate schemes under perturbation conditions can be obtained, rather than simply evaluating their theoretical performance under ideal geometric conditions.

[0050] Next, the convergence trajectory data after perturbation is compared with the baseline convergence trajectory data to extract the differences in residual fluctuation morphology and vorticity turning point positions. This involves aligning the two sets of convergence trajectory data before and after perturbation in time series, using the iteration step size as a unified comparison benchmark, and performing a difference analysis on the residual change trend and vorticity dissipation evolution process. The difference in residual fluctuation morphology refers to the degree of difference in fluctuation amplitude, oscillation frequency, convergence speed, and local change trend of the residual change curves before and after perturbation. It reflects the impact of geometric perturbation on the stability of the numerical solution. For example, if a small geometric change causes the residual to change from a smooth decline to periodic oscillation, or significantly delays the time it takes for the residual to enter the stable region, it indicates that the structure is highly sensitive to manufacturing errors. The difference in vorticity turning point positions refers to the offset between iteration steps corresponding to the key turning points in the vorticity dissipation intensity change curves before and after perturbation. It characterizes the sensitivity of the flow structure adjustment process to geometric changes. For example, if a small size change causes the vorticity dissipation enhancement stage to appear earlier, or significantly shifts the vorticity stabilization position later, it indicates that the internal flow state of the structure is easily affected by manufacturing deviations. By simultaneously extracting the two types of differences mentioned above, the manufacturing adaptability of candidate schemes can be evaluated from two aspects: numerical convergence behavior and flow physics evolution behavior.

[0051] Next, based on the actual processing accuracy requirements and flow stability needs, the allowable variation ranges corresponding to the differences in residual fluctuation patterns and vorticity turning points are pre-set as manufacturing adaptability evaluation thresholds. When the convergence trajectory after disturbance shows a significant change compared to the baseline convergence trajectory, such as the residual oscillation amplitude exceeding the preset range, the convergence process changing from a smooth state to an unstable state, or a large shift in the vorticity dissipation turning point, it is considered that although the candidate scheme may have better performance under ideal geometric conditions, its flow field response is too sensitive to manufacturing errors and it is difficult to maintain stable performance under actual processing and operating conditions. Therefore, it is determined that it does not have manufacturing adaptability and is eliminated.

[0052] Furthermore, for candidate schemes that pass the manufacturing adaptability evaluation, the convergence stability criterion is used as a prerequisite hard constraint for screening, retaining only structural schemes that meet the following conditions: first, the convergence stability criterion reaches a preset threshold requirement; second, the flow field convergence trajectory remains stable under disturbance conditions; and third, it has low pressure drop and high heat transfer enhancement capability while satisfying stability. Finally, the candidate schemes that meet the above conditions are determined as the target geometric structure output, thereby avoiding the selection of pseudo-optimal structures that are highly sensitive to geometric errors and improving the consistency between the final microchannel evaporator structural design results and the actual product performance.

[0053] Furthermore, step P40 in this embodiment of the application also includes: P45: The channel inlet section of the candidate scheme is set as a variable cross-section structure with a gradually decreasing cross-sectional area along the fluid flow direction; P46: The variable cross-section structure is configured to extend the length of the flow development section, so that the vorticity evolution information in the convergence trajectory data exhibits delayed convergence characteristics.

[0054] Specifically, among the candidate schemes selected after manufacturing adaptability evaluation and convergence stability screening, the inlet flow region of the microchannel evaporator was structurally optimized by designing the inlet section as a variable cross-section structure with a flow cross-sectional area that gradually decreases along the mainstream fluid direction. The variable cross-section structure refers to a flow channel form where the cross-sectional area continuously changes along the axial direction. Unlike traditional constant cross-section inlet structures, this allows for active regulation of the flow development state in the inlet region by altering the velocity distribution and pressure gradient changes after the fluid enters the microchannel.

[0055] Specifically, the variable cross-section structure can be formed by gradually decreasing the inlet width, gradually decreasing the inlet height, or a combination of width and height variations, allowing the fluid to undergo a continuous, gradual transition as it enters the main body of the microchannel from the initial inlet region. Because the flow cross-sectional area gradually decreases along the flow direction, according to the law of conservation of mass, the fluid velocity continuously adjusts with the change in flow area, avoiding the rapid formation of localized high-speed jets, strong pressure gradients, and large-scale vortices caused by sudden diameter reduction. The channel inlet section refers to the transition region from when the fluid enters from the inlet boundary until the flow is fully developed; this region directly affects the uniformity of velocity distribution and the stability of the flow in the subsequent main channel.

[0056] Furthermore, by setting a variable cross-section structure, the inlet region is transformed from an abrupt flow state to a gradual flow state, giving the fluid more room for flow adjustment. During flow field simulation, this structure alters the variation law of vorticity evolution information in the convergence trajectory data, causing the vorticity dissipation intensity to gradually transition from a rapidly changing state to a stable state, thus forming a delayed convergence characteristic that differs from ordinary constant cross-section structures.

[0057] The variable cross-section structure is configured to extend the length of the flow development section, causing the vortex evolution information in the convergence trajectory data to exhibit delayed convergence characteristics, thereby suppressing the peak value of the shear force at the two-phase flow interface. Specifically, by setting a tapered variable cross-section structure in the channel inlet region, the fluid has a longer flow development path before entering the main microchannel, allowing the velocity gradient, pressure distribution, and local vortex structure to be gradually adjusted over a longer spatial range. The flow development section refers to the spatial region where the fluid gradually transitions from the inlet disturbance state to a stable and fully developed flow state; its length directly affects the fluid velocity profile formation process and local flow stability.

[0058] In existing microchannel structures with uniform cross-sections, the abrupt change in the inlet cross-section easily generates local high-speed and strong shear regions after the fluid enters the channel. This causes the vortex dissipation intensity to rise rapidly within a few iterations and form a significant peak during the flow field stabilization process. However, this application, through its variable cross-section structure, allows the inlet velocity gradient to gradually establish along the flow direction, reducing the instantaneous disturbance intensity experienced by the fluid. This results in a time delay in the peak value of the vortex dissipation intensity variation curve, meaning the vortex evolution information in the convergence trajectory data exhibits delayed convergence characteristics.

[0059] The delayed convergence characteristic refers to the increased number of iterations required for the vorticity dissipation intensity to reach a stable range, i.e., the vorticity evolution curve changes from a rapid transition state to a slow transition state. This characteristic does not indicate a decrease in numerical solution efficiency, but rather that the energy exchange process within the flow field is more gradual, avoiding sudden changes in the flow structure in the inlet region.

[0060] When microchannel evaporators are used for two-phase flow heat transfer, the tapered cross-section structure reduces the abrupt velocity changes in the initial stage of the gas-liquid two-phase flow entering the channel, resulting in a more uniform local shear force at the gas-liquid interface. The interfacial shear force refers to the tangential force between the gas and liquid phases caused by the velocity difference. Excessive peak values ​​can lead to rapid bubble stretching, breakup, or local flow instability, resulting in pressure fluctuations and decreased heat transfer performance. By extending the flow development section and allowing the vortex evolution process to exhibit delayed convergence, the peak value of the interfacial shear force caused by the rapid generation of local vortices in the inlet region can be reduced, making the two-phase flow process more stable. Furthermore, the vortex evolution characteristics generated by this structure can be further captured by the collected convergence trajectory data, serving as an important basis for constructing convergence stability criteria.

[0061] Furthermore, before performing the screening using the aforementioned convergence stability criterion, step P40 in this embodiment of the application further includes: P47: Based on the principles of mass conservation and momentum conservation, construct inequality constraints that limit the relative motion between the gas and liquid phases in two-phase flow, and define them as physical legitimacy criteria; P48: Call the instantaneous flow field snapshot in the convergence trajectory data to extract the distribution data of gas phase velocity and liquid phase velocity; P49: Verify whether the distribution data satisfies the physical legitimacy criteria. If the relative difference between the gas phase velocity and the liquid phase velocity exceeds the extreme value limited by the inequality constraints, then determine that the flow field state corresponding to the current convergence trajectory data is a physically unreliable state.

[0062] It should be understood that, before evaluating candidate schemes using the convergence stability criterion, this application further introduces a legality verification mechanism based on the laws of flow physics to avoid obtaining flow field results that are formally convergent but do not conform to actual physical laws due to numerical calculation errors, model parameter setting deviations, or local mesh quality problems. The physical legality criterion refers to constraints established based on the laws of mass transfer and momentum exchange in two-phase flow, used to determine whether the simulated gas and liquid phase motion states are within a reasonable physical range.

[0063] Specifically, in the gas-liquid two-phase flow process of a microchannel evaporator, there exists a relative motion relationship between the gas and liquid phases, determined by interfacial shear force, pressure gradient, interphase momentum exchange, and local flow resistance. According to the principle of mass conservation, the changes in mass flow rate of the gas and liquid phases should satisfy continuity constraints; according to the principle of momentum conservation, the velocity difference between the two phases should be limited by interphase forces and the flow channel structure. Therefore, based on the above physical laws, a constraint relationship for the relative velocity difference between the gas and liquid phases is established, for example: ;in, Indicates gas phase velocity, Indicates the liquid phase velocity. This represents the maximum permissible relative velocity difference determined based on the current operating conditions, fluid properties, and channel structure. It is used to limit the degree of slippage between the gas and liquid phases, preventing excessive velocity separation from leading to enhanced interface disturbances, abnormal changes in flow patterns, or loss of physical meaning in the calculation results.

[0064] Furthermore, the physical legitimacy criterion is not only used to judge the velocity relationship at a single location, but can also be used to make an overall evaluation based on the velocity distribution of multiple computational units within the flow field region. That is, the gas phase velocity field and liquid phase velocity field at different locations inside the microchannel are synchronously constrained to ensure that the two-phase motion relationship in the entire flow region conforms to the actual physical process.

[0065] Specifically, in the process of generating convergence trajectory data, in addition to recording residual change information and vorticity evolution information, instantaneous flow field state data corresponding to some key iteration nodes are also saved simultaneously. An instantaneous flow field snapshot refers to the recorded data of the spatial distribution of various physical variables within the flow field calculation area under a specified iteration step, including but not limited to velocity field, pressure field, phase distribution information, and phase interface state information. This step extracts the velocity vector data corresponding to the gas phase region and the velocity vector data corresponding to the liquid phase region from the instantaneous flow field snapshot, forming gas phase velocity distribution data and liquid phase velocity distribution data respectively. Specifically, by identifying the phases of the gas-liquid two-phase calculation area, the gas phase region and liquid phase region corresponding to different calculation units are determined based on the phase volume fraction, phase proportion, or interface identification results, and the velocity variables within the corresponding regions are read respectively. The gas phase velocity and liquid phase velocity distribution data are used to describe the differences in motion state at different locations during the two-phase flow, providing a data basis for subsequent physical validity verification.

[0066] Next, the acquired gas phase velocity distribution data and liquid phase velocity distribution data are substituted into the physical validity criterion to verify the relative motion state between the gas and liquid phases at different spatial locations point by point. When the relative difference between the gas phase velocity and the liquid phase velocity in any calculation region exceeds the preset maximum allowable range, it indicates that the motion relationship between the gas and liquid phases in the current simulation state has deviated from normal physical laws. For example, without a reasonable interphase interaction mechanism, if the gas phase velocity is much higher than the liquid phase velocity and continues to increase, it may indicate an abnormally enhanced local interface shear effect; if the liquid phase velocity shows a significant lag, it may indicate the presence of non-physical stagnation or numerical artifacts in a local area. The physically unreliable state refers to a state where, although the current flow field calculation may meet the solver residual convergence requirements, the corresponding gas-liquid two-phase motion state does not conform to the actual flow physics. Convergence trajectory data in the physically unreliable state are not used as valid input for subsequent convergence stability evaluation, or their corresponding candidate schemes are directly judged as unqualified schemes.

[0067] Therefore, before constructing a stability evaluation system based on converged trajectory data, this application adds a physical validity pre-screening process based on mass and momentum conservation. This ensures that the evaluation of candidate schemes considers not only whether the numerical solution process is stable, but also whether the flow field state after convergence satisfies the actual two-phase flow law. This avoids mistaking mathematically converged but physically distorted flow field results as valid results due to simply relying on the residual decreasing trend, thereby improving the reliability and engineering applicability of the target geometry screening process.

[0068] Furthermore, step P47 in this embodiment of the application also includes: P47-1: The inequality constraint is configured to limit the relative motion between the gas phase and the liquid phase; P47-2: When cavitation accumulation or cavitation backflow is detected, the current convergence trajectory data is marked as a numerical pseudo-solution and disabled in the subsequent screening process.

[0069] Specifically, when constructing the physical legitimacy criteria, the physical constraints on the gas-liquid phase motion during the two-phase flow in the microchannel evaporator are further considered, limiting the velocity difference, momentum exchange intensity, and phase interface migration rate between the gas and liquid phases. The relative motion limit refers to the maximum allowable relative motion range between the gas and liquid phases under the current fluid medium properties, inlet operating parameters, and channel geometric constraints. This is used to avoid non-realistic separation of the gas and liquid phases due to phase interface capture errors, phase volume fraction diffusion, or insufficient local mesh resolution during numerical solution.

[0070] In two-phase flow numerical simulations, the gas and liquid phases are allowed a certain degree of slip velocity due to differences in density, viscosity, and inertial properties. However, this slip velocity must satisfy the laws of mass conservation and momentum exchange. When the relative velocity between the gas and liquid phases exceeds a reasonable range, it may cause the bubble phase distribution in the numerical calculation to deviate from the actual motion law, such as non-physical bubble drift, local aggregation, or abnormal interface migration direction. Therefore, this application establishes inequality constraints to limit the relative velocity between the gas and liquid phases.

[0071] For example, the gas-liquid relative velocity constraint can be expressed as: ;in, Indicates the local velocity in the gas phase. Indicates the local velocity of the liquid phase. This represents the maximum permissible slip velocity threshold determined based on two-phase flow conditions. The maximum permissible slip velocity threshold can be determined based on parameters such as fluid properties, channel size, inlet mass flow rate, and phase content, making it adaptable to different microchannel evaporator operating conditions.

[0072] Furthermore, by introducing inequality constraints into the validity judgment process of convergence trajectory data, the convergence trajectory data not only meets the convergence requirements at the numerical calculation level but also satisfies the requirements of the physical laws of two-phase flow. That is, if the flow field calculation corresponding to a candidate scheme obtains a low residual, but the relative motion relationship between the gas and liquid phases exceeds the constraint range, then the convergence trajectory data is not considered to truly reflect the actual flow state. Among them, the numerical distortion phenomenon of cavitation drift refers to the phenomenon in the two-phase flow simulation process where the bubble trajectory, residence position, or migration velocity deviates from the actual physical behavior due to numerical errors. For example, bubbles may continue to move in locations where there is no actual driving force, or abnormally accumulate in local areas of the flow channel, thereby causing the subsequently extracted eddy evolution information, interface shear characteristics, and stability evaluation results to lose their authenticity.

[0073] When verifying the convergence trajectory data using the aforementioned physical validity criteria, anomaly detection is further performed on the spatial distribution of the gas-liquid two-phase system to identify non-physical phase behaviors caused by numerical errors. Specifically, cavitation accumulation refers to the continuous abnormal aggregation of gas phase regions at local flow channel locations, resulting in a significant deviation of the gas phase volume distribution from the actual two-phase flow pattern. Backflow of cavitation refers to the phenomenon where the gas phase region moves in the opposite direction to the mainstream direction, or the bubble migration trajectory exhibits reverse motion that violates the laws of pressure gradient and momentum exchange. These phenomena are usually not caused by actual flow physics mechanisms but may be due to insufficient mesh size, interface capture errors, unreasonable time step settings, or numerical oscillations during the solution process.

[0074] In practice, the presence of cavitation anomalies can be determined by detecting whether the gas phase volume fraction distribution continuously increases in a local area and exceeds a preset threshold, whether the gas phase velocity direction is opposite to the mainstream direction, and whether there are non-physical migrations or periodic abnormal movements at the gas-liquid interface. When any of the above anomaly conditions are met, the current convergence trajectory data is marked as a numerical pseudo-solution, which is a simulation result generated by the numerical calculation process but cannot correspond to the real physical flow state. Although it may meet the solver residual convergence requirements, it has no engineering analysis value. For convergence trajectory data marked as numerical pseudo-solutions, they will not be included in the convergence stability criterion calculation in subsequent steps, will not be used as the basis for evaluating the performance of candidate schemes, and the corresponding candidate schemes will be directly disabled to avoid the impact of erroneous flow field data on the target geometry selection results. This avoids the problem of erroneous structure optimization caused by accepting numerical pseudo-solutions in the traditional CFD optimization process, and improves the reliability and practical application value of the optimization results for low flow resistance structures of microchannel evaporators.

[0075] Furthermore, in the screening of the candidate solutions, step P40 of this embodiment also includes: P48: Retrieve the convergence stability criteria corresponding to each group of candidate schemes, and include candidate schemes with quantitative scores higher than the preset qualified line into the preliminary selection set; P49: In the preliminary selection set, extract the pressure drop data and heat transfer coefficient data of each candidate scheme when it reaches steady state in the flow field simulation, and perform secondary sorting according to the optimization objectives of pressure drop reduction and heat transfer enhancement; P410: Identify the convergence trajectory data of each candidate scheme in the preliminary selection set, eliminate candidate schemes with non-smooth flow field convergence paths, determine the remaining schemes as qualified schemes that meet the pre-condition hard constraints, and solidify the best-ranked qualified scheme into the target geometry.

[0076] Optionally, after completing the convergence trajectory data analysis and convergence stability criterion calculation for each candidate scheme, the quantitative score results for each group of candidate schemes are obtained, and the candidate schemes are screened in the first stage according to the preset stability evaluation threshold. The quantitative score refers to the comprehensive evaluation result obtained by weighted fusion of the first evaluation factor and the second evaluation factor, used to characterize the stability of the candidate scheme and the reliability of the flow state evolution during the numerical solution process.

[0077] In practice, the quantitative scores of each candidate scheme are compared with a preset pass threshold. If the quantitative score of a candidate scheme is higher than or equal to the preset pass threshold, the candidate scheme is considered to meet the basic flow field stability requirements and is included in the preliminary selection set. Otherwise, the candidate scheme is considered to have problems such as unstable convergence process, abnormal vorticity evolution, or delayed flow state switching, and is removed from the subsequent optimization process. The preliminary selection set refers to the subset of candidate structures formed after convergence stability constraint screening, distinct from the entire candidate scheme set. Each candidate scheme in the preliminary selection set has met the basic requirements for flow field solution stability. This step allows for the early elimination of structures with only surface performance advantages but insufficient flow field reliability before performance evaluation, reducing the risk of invalid calculations and incorrect selections in subsequent optimization processes.

[0078] Next, for the candidate schemes in the initial selection set after passing the convergence stability screening, the performance parameters of their corresponding flow field simulation after reaching a stable state are further read, including the pressure difference between the fluid inlet and outlet and the heat transfer capacity parameters in the fluid-solid heat transfer process.

[0079] The pressure drop data refers to the pressure loss generated by the fluid passing through the microchannel evaporator, used to evaluate the impact of the structure on flow resistance; the heat transfer coefficient data refers to the heat transfer capacity parameter under unit temperature difference and unit heat transfer area conditions, used to evaluate the heat transfer enhancement capability of candidate structures. Since the microchannel evaporator needs to simultaneously meet the requirements of low flow resistance and high heat transfer performance, this application does not solely use the minimum pressure drop or the maximum heat transfer coefficient as the evaluation criterion, but rather ranks the candidate schemes based on comprehensive performance that meet the flow field stability constraints.

[0080] Specifically, a comprehensive performance evaluation index can be established based on a preset weighting relationship to integrate the evaluation of pressure drop reduction and heat transfer enhancement. For example, the pressure drop normalization index can be used as the evaluation metric for reducing flow resistance, and the heat transfer coefficient normalization index can be used as the evaluation metric for enhancing heat transfer capacity, and then they can be comprehensively ranked according to preset weights. In this way, the ranking results can reflect the comprehensive performance balance between low flow resistance and enhanced heat transfer of candidate schemes. Since all candidate schemes entering the secondary ranking stage have passed the convergence stability criterion, the high-performance illusion caused by unstable structures in traditional optimization methods can be avoided, ensuring that the performance ranking results correspond to the real and reliable flow field state.

[0081] After completing the secondary sorting based on pressure drop and heat transfer performance, the convergence trajectory data corresponding to each candidate scheme is further retrieved to re-verify the continuity and stable change characteristics of their solution paths. This is to identify candidate schemes that, although meeting the quantitative score requirements, still exhibit abnormal fluctuations in the local iteration process. The flow field convergence path refers to the dynamic trajectory formed by the changes in residual changes, vorticity evolution, and related flow state parameters with the iteration step size from the initial flow field state to the final steady state. Smooth flow field convergence path means that during continuous iteration, the residual change trend is continuous without significant abrupt changes, the vorticity dissipation intensity change process does not exhibit abnormal jumps, and the flow field state gradually approaches the final state according to a stable trend.

[0082] In practice, by analyzing the convergence trajectory data corresponding to the candidate schemes, it is possible to identify situations where the residual suddenly increases during a steady decline, the vortex dissipation intensity exhibits discontinuous jumps, or the flow field state repeatedly switches between multiple stable states. If the above non-smooth convergence characteristics are detected, it is determined that although the corresponding candidate scheme may meet the basic stability score requirements, its flow field solution path still has potential uncertainties, and therefore the candidate scheme is eliminated.

[0083] The remaining candidate schemes after the above screening are determined as qualified schemes that meet the prerequisite hard constraints. These prerequisite hard constraints include: the convergence stability criterion meets the preset requirements, the flow field response remains stable under disturbance conditions, the physical legality criterion is passed, and the convergence path remains smooth.

[0084] Furthermore, among all qualified schemes, based on the obtained comprehensive performance ranking results, the candidate scheme with the best overall performance in terms of pressure drop reduction and heat transfer enhancement is selected, and its corresponding geometric parameter combination is solidified to form the final output target geometric structure. Here, solidification refers to determining the complete geometric parameters of the target microchannel evaporator structure, including channel cross-sectional dimensions, inlet structure form, flow development section length, and other structural parameters affecting flow heat transfer performance, making it the target structure for subsequent manufacturing design or experimental verification.

[0085] Through the above steps, compared with the method of optimization based solely on steady-state pressure drop and heat transfer coefficient, this application can effectively avoid selecting pseudo-optimal structures that have excellent simulation results but unstable actual operation, thereby improving the flow field reliability, manufacturing adaptability, and consistency of actual operating performance of the target microchannel evaporator structure.

[0086] In summary, the embodiments of this application have at least the following technical effects: This application uses convergence trajectory data as the sole data source for evaluating flow stability, achieving a causal mapping from numerical convergence behavior to physical structural defects. This effectively solves the problem in existing technologies where pseudo-optimal structures with excellent steady-state indices but unstable numerical convergence are difficult to identify. By setting the convergence stability criterion as a prerequisite hard constraint, candidate schemes with flow separation tendencies and high shear dissipation risks are pre-screened during the design phase, significantly reducing the risk of flow resistance drift caused by manufacturing tolerances or operating condition fluctuations after physical processing. By introducing geometric disturbance tests simulating manufacturing tolerances and physical legitimacy verification of gas-liquid relative motion, the application avoids the mixing of numerical pseudo-solutions such as cavitation drift into the optimization results. By prioritizing the elimination of schemes with abnormal convergence trajectories in the screening process, the application reduces the amount of invalid CFD calculations, improves optimization efficiency, and ensures that the final output target geometry has high operational stability and physical reliability while reducing pressure drop and enhancing heat transfer.

[0087] The technology achieves the effect of constructing stability criteria based on convergent trajectory data, filtering out unstable structures, and improving the reliability of simulation optimization results and the consistency of physical performance.

[0088] Example 2 is based on the same inventive concept as the low flow resistance microchannel evaporator performance optimization method based on flow field simulation in the previous examples, such as... Figure 2 As shown, this application provides a performance optimization system for low-flow-resistance microchannel evaporators based on flow field simulation. The system and method embodiments in this application are based on the same inventive concept. The system includes: The candidate scheme generation module 11 is used to construct the fluid computation domain of the microchannel evaporator and generate multiple sets of candidate schemes containing channel cross-sectional geometric features within the fluid computation domain.

[0089] The flow field simulation calculation module 12 is used to perform flow field simulation calculations for each group of candidate schemes. During the iterative solution process, it synchronously collects and stores convergence trajectory data, which includes residual change information and vortex evolution information.

[0090] The stability criterion construction module 13 is used to construct a convergence stability criterion by using the convergence trajectory data as the sole data source for evaluating the flow stability of candidate schemes.

[0091] The candidate scheme optimization module 14 is used to select the candidate schemes with the goal of reducing pressure drop and enhancing heat transfer. The convergence stability criterion is used as a prerequisite hard constraint to screen the candidate schemes and retain only the candidate schemes that meet the convergence stability criterion and have a smooth flow field convergence path as the target geometry output.

[0092] Furthermore, the flow field simulation calculation module 12 is also used to perform the following steps: In each iteration step of the flow field simulation calculation, the residual vector output by the solver is read, and the absolute value sequence of the residuals of the continuity equation and the momentum equation is extracted as the residual change information. In the same iteration step, the vorticity tensor of the entire flow field is extracted, the spatial distribution of the local vorticity dissipation intensity is calculated, and its evolution process with the iteration step size is recorded as the vorticity evolution information. A data mapping relationship indexed by the iteration step size is established, and the residual change information and the vorticity evolution information are aligned and stored according to the same iteration step size, and then encapsulated into structured convergence trajectory data.

[0093] Furthermore, the stability criterion construction module 13 is also used to perform the following steps: The convergence trajectory data is traversed to identify the continuous iteration interval after the residual value enters the preset error band in the residual change information. Within the continuous iteration interval, the fluctuation amplitude and fluctuation frequency of the residual value are statistically analyzed to generate a first evaluation factor representing the stability of the convergence process. Within the continuous iteration interval, the turning point of vortex dissipation intensity in the vortex evolution information is extracted, and the step size from the iteration start point to the turning point is calculated to generate a second evaluation factor characterizing the hysteresis of the flow state switching. The first evaluation factor and the second evaluation factor are weighted and fused to generate a quantitative score as a convergence stability criterion.

[0094] Furthermore, the stability criterion construction module 13 is also used to perform the following steps: If the convergence trajectory data exhibits long-tailed oscillations or non-monotonic convergence characteristics, it is determined that the candidate scheme has an inherent tendency to induce flow separation in actual operation.

[0095] Furthermore, the candidate solution optimization module 14 is also used to perform the following steps: Before applying the aforementioned pre-existing hard constraints, a small geometric perturbation simulating manufacturing tolerances is applied to the geometric model of the candidate scheme to generate a perturbed geometric variant. The flow field simulation is then re-executed for the perturbed geometric variant to obtain the perturbed convergence trajectory data. The perturbed convergence trajectory data is compared with the baseline convergence trajectory data to extract the residual fluctuation morphology difference and the vortex inflection point difference. If the residual fluctuation morphology difference and the vortex inflection point difference exceed a preset tolerance limit, the candidate scheme is determined to lack manufacturing adaptability due to its high flow field response sensitivity, and is thus eliminated.

[0096] Furthermore, the candidate solution optimization module 14 is also used to perform the following steps: The candidate scheme has a channel inlet section with a variable cross-section structure that gradually decreases in cross-sectional area along the fluid flow direction; the variable cross-section structure is configured to extend the length of the flow development section, so that the vorticity evolution information in the convergence trajectory data exhibits delayed convergence characteristics.

[0097] Furthermore, the candidate solution optimization module 14 is also used to perform the following steps: Based on the principles of mass and momentum conservation, inequality constraints are constructed to limit the relative motion between the gas and liquid phases in two-phase flow, and these are defined as physical legitimacy criteria. Before using the convergence stability criteria for screening, instantaneous flow field snapshots from the convergence trajectory data are retrieved to extract the distribution data of gas and liquid phase velocities. The distribution data is then checked to see if it satisfies the physical legitimacy criteria. If the relative difference between the gas and liquid phase velocities exceeds the extreme value limited by the inequality constraints, the flow field state corresponding to the current convergence trajectory data is determined to be physically unreliable.

[0098] Furthermore, the candidate solution optimization module 14 is also used to perform the following steps: The inequality constraints are configured to limit the relative motion between the gas and liquid phases; when cavitation accumulation or cavitation backflow is detected, the current convergence trajectory data is marked as a numerical pseudo-solution and disabled in the subsequent screening process.

[0099] Furthermore, the candidate solution optimization module 14 is also used to perform the following steps: With the optimization objectives of reducing pressure drop and enhancing heat transfer, the convergence stability criterion serves as a prerequisite hard constraint for screening the candidate solutions, including: The convergence stability criteria corresponding to each group of candidate schemes are retrieved, and candidate schemes with quantitative scores higher than the preset qualified line are included in the preliminary selection set. In the preliminary selection set, the pressure drop data and heat transfer coefficient data of each candidate scheme when reaching steady state in the flow field simulation are extracted, and a secondary sorting is performed according to the optimization objectives of pressure drop reduction and heat transfer enhancement. The convergence trajectory data of each candidate scheme in the preliminary selection set are identified, and candidate schemes with non-smooth flow field convergence paths are eliminated. The remaining schemes are determined as qualified schemes that meet the pre-condition hard constraints, and the best-ranked qualified scheme is solidified into the target geometry.

[0100] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0101] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0102] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation, characterized in that, The method includes: A fluid computation domain for a microchannel evaporator is constructed, and multiple candidate schemes containing channel cross-sectional geometric features are generated within the fluid computation domain. For each group of candidate schemes, flow field simulation calculations are performed. During the iterative solution process, convergence trajectory data is collected and stored simultaneously. The convergence trajectory data includes residual change information and vortex evolution information. Using the convergence trajectory data as the sole data source for evaluating the flow stability of candidate schemes, a convergence stability criterion is constructed. With the optimization goals of reducing pressure drop and enhancing heat transfer, the convergence stability criterion is used as a prerequisite hard constraint to screen the candidate schemes. Only the candidate schemes that meet the convergence stability criterion and have a smooth flow field convergence path are retained as the target geometry output.

2. In the low flow resistance microchannel evaporator performance optimization method based on flow field simulation as described in claim 1, if the convergence trajectory data exhibits long-tailed oscillations or non-monotonic convergence characteristics, it is determined that the candidate scheme has an inherent tendency to induce flow separation in actual operation.

3. The method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation as described in claim 1, characterized in that, For each candidate scheme, flow field simulation calculations are performed. During the iterative solution process, convergence trajectory data is collected and stored simultaneously, including: In each iteration of the flow field simulation calculation, the residual vector output by the solver is read, and the absolute value sequence of the residuals of the continuity equation and the momentum equation is extracted as the residual change information. In the same iteration step, the vorticity tensor of the entire flow field is extracted, the spatial distribution of local vorticity dissipation intensity is calculated, and its evolution process with the iteration step size is recorded as the vorticity evolution information. A data mapping relationship indexed by the iteration step size is established. The residual change information and the eddy evolution information are aligned and stored with the same iteration step size, and then encapsulated into structured convergence trajectory data.

4. The method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation as described in claim 1, characterized in that, Using the convergence trajectory data as the sole data source for evaluating the flow stability of candidate schemes, a convergence stability criterion is constructed, including: Traverse the convergence trajectory data and identify the continuous iteration interval after the residual value enters the preset error band in the residual change information; Within the continuous iteration interval, the fluctuation amplitude and fluctuation frequency of the statistical residuals are used to generate the first evaluation factor for the stationarity of the table convergence process. Within the continuous iteration interval, the turning point of vortex dissipation intensity in the vortex evolution information is extracted, the step size span from the iteration start point to the turning point is calculated, and a second evaluation factor characterizing the hysteresis of flow state switching is generated. The first evaluation factor and the second evaluation factor are weighted and fused to generate a quantitative score, which serves as a criterion for convergence stability.

5. The method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation as described in claim 1, characterized in that, Before applying the aforementioned pre-existing hard constraints, the following is also included: On the geometric model of the candidate scheme, a small geometric perturbation simulating manufacturing tolerances is applied to generate a perturbed geometric variant; The flow field simulation calculation is re-executed for the disturbed geometric variant to obtain the converged trajectory data after the disturbance; By comparing the converged trajectory data after the disturbance with the baseline converged trajectory data, the differences in residual fluctuation morphology and vorticity turning point are extracted. If the difference in residual fluctuation morphology and the difference in vorticity turning point exceed the preset tolerance limit, the candidate scheme is determined to be unsuitable for manufacturing due to its high sensitivity to flow field response, and the candidate scheme is eliminated.

6. The method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation as described in claim 1, characterized in that, The candidate scheme has a channel inlet section configured as a variable cross-section structure with a gradually decreasing cross-sectional area along the fluid flow direction. The variable cross-section structure is configured to extend the length of the flow development section, so that the vorticity evolution information in the convergence trajectory data exhibits delayed convergence characteristics.

7. The method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation as described in claim 1, characterized in that, Before using the convergence stability criterion for screening, the process also includes: Based on the principles of conservation of mass and momentum, inequality constraints are constructed to restrict the relative motion between the gas and liquid phases in two-phase flow, and these are defined as physical legality criteria. The instantaneous flow field snapshot in the convergence trajectory data is called to extract the distribution data of gas phase velocity and liquid phase velocity; Verify whether the distribution data meets the physical legitimacy criterion. If the relative difference between the gas phase velocity and the liquid phase velocity exceeds the extreme value limited by the inequality constraint, then determine that the flow field state corresponding to the current convergence trajectory data is a physically unreliable state.

8. The method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation as described in claim 7, characterized in that, The inequality constraint is configured to limit the relative motion between the gas phase and the liquid phase. When cavitation accumulation or cavitation backflow is detected, the current convergence trajectory data is marked as a numerical pseudo-solution and disabled in the subsequent filtering process.

9. The method for performance optimization of low-flow-resistance microchannel evaporators based on flow field simulation as described in claim 8, characterized in that, With the optimization objectives of reducing pressure drop and enhancing heat transfer, the convergence stability criterion serves as a prerequisite hard constraint for screening the candidate solutions, including: Retrieve the convergence stability criteria for each group of candidate solutions, and include the candidate solutions with quantitative scores higher than the preset qualification line into the preliminary selection set; In the initial selection set, the pressure drop data and heat transfer coefficient data of each candidate scheme when it reaches steady state in the flow field simulation are extracted, and a second sorting is performed according to the optimization objectives of reducing pressure drop and enhancing heat transfer. The convergence trajectory data of each candidate scheme in the initial selection set are identified, and candidate schemes with non-smooth flow field convergence paths are eliminated. The remaining schemes are determined as qualified schemes that meet the pre-condition hard constraints, and the best among the qualified schemes is solidified into the target geometry.

10. A performance optimization system for low-flow-resistance microchannel evaporators based on flow field simulation, characterized in that, The system includes: The candidate scheme generation module is used to construct the fluid computation domain of the microchannel evaporator and generate multiple sets of candidate schemes containing channel cross-sectional geometric features within the fluid computation domain. The flow field simulation calculation module is used to perform flow field simulation calculations for each group of candidate schemes. During the iterative solution process, it synchronously collects and stores convergence trajectory data, which includes residual change information and vorticity evolution information. The stability criterion construction module is used to construct a convergence stability criterion by using the convergence trajectory data as the sole data source for evaluating the flow stability of candidate schemes. The candidate scheme optimization module is used to optimize the pressure drop reduction and heat transfer enhancement. The convergence stability criterion is used as a prerequisite hard constraint to screen the candidate schemes. Only the candidate schemes that meet the convergence stability criterion and have a smooth flow field convergence path are retained as the target geometry output.

Images