Engine intake distortion evaluation method, device, equipment, medium and product

Abstract

The present disclosure relates to the technical field of engines, and particularly provides an engine intake distortion evaluation method, device, equipment, medium and product. In the present disclosure, pressure distribution data of an engine to be evaluated is obtained, the pressure distribution data being pressure data of an intake port outlet section of the engine to be evaluated; feature extraction is performed on the pressure distribution data to obtain intake pressure features; the intake pressure features are analyzed based on a pre-trained partition model to obtain a partition result of the intake port outlet section, the partition result representing a flow field physical partition strategy of the intake port outlet section; and distortion analysis is performed based on the partition result to obtain an intake distortion evaluation result of the engine to be evaluated. The partition result obtained in this way is determined by the physical law of the flow field itself, can accurately fit the real physical structure in the flow field, and can improve the accuracy of the distortion evaluation result when calculating the distortion of each sub-region. Therefore, the accuracy of engine intake distortion evaluation can be improved.

Description

Technical Field

[0001] This disclosure pertains to the field of engine technology, specifically relating to a method, apparatus, equipment, medium, and product for assessing engine intake distortion. Background Technology

[0002] An aero-engine is a highly complex and precise thermodynamic machine that directly provides the power required for aircraft flight. The air intake outlet is the interface section between the air intake and the engine compressor, providing the required airflow for various engine states. The total pressure distortion present at the air intake outlet section is a key factor inducing instability phenomena such as fan / compressor stall and surge in aero-engines. Therefore, the stability of an aero-engine is closely related to the uniformity of the airflow coming into its air intake.

[0003] In related technologies, the intake duct outlet section is typically divided into several sub-regions based on human experience. Distortion indices for each sub-region are then calculated to quantify flow inhomogeneity and assess intake distortion. However, this assessment method relies on personal experience for zoning, and the resulting sub-regions may not correspond to the actual physical structure of the flow field, leading to unreasonable zoning and reduced accuracy of the assessment results. Therefore, it lowers the accuracy of engine intake distortion assessment. Summary of the Invention

[0004] This disclosure addresses some deficiencies mentioned in the background art by providing a method, apparatus, device, medium, and product for assessing engine intake distortion, which can improve the accuracy of engine intake distortion assessment.

[0005] In a first aspect, embodiments of this disclosure provide a method for evaluating engine intake distortion, comprising: acquiring pressure distribution data of an engine to be evaluated, wherein the pressure distribution data is pressure data of the intake duct outlet section of the engine to be evaluated; extracting features from the pressure distribution data to obtain intake pressure features; performing zoning analysis on the intake pressure features based on a pre-trained zoning model to obtain zoning results of the intake duct outlet section, wherein the zoning results characterize the flow field physical zoning strategy of the intake duct outlet section; and performing distortion analysis based on the zoning results to obtain an intake distortion evaluation result for the engine to be evaluated.

[0006] Optionally, the step of extracting features from the pressure distribution data to obtain intake pressure features includes: determining the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point, wherein the grid point is obtained by gridding the intake duct outlet cross-section; determining the pressure statistical characteristic parameters within a preset neighborhood range of each grid point based on the pressure distribution data corresponding to each grid point, wherein the preset neighborhood range is the area range determined with each grid point as the center; determining the pressure extreme points of each grid point based on the pressure distribution data corresponding to each grid point; and processing each spatial gradient vector, each pressure statistical characteristic parameter, and each pressure extreme point to obtain the intake pressure features.

[0007] Optionally, determining the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point includes: performing edge detection on the pressure distribution data corresponding to each grid point to obtain the pressure gradient amplitude of each grid point; and using the pressure gradient amplitude as the spatial gradient vector.

[0008] Optionally, determining the pressure statistical characteristic parameters within a preset neighborhood range of each grid point based on the pressure distribution data corresponding to each grid point includes: acquiring the pressure distribution data within a preset neighborhood range of each grid point; determining the average pressure within the preset neighborhood range based on the pressure distribution data within the preset neighborhood range; determining the pressure standard deviation within the preset neighborhood range based on the pressure distribution data corresponding to each grid point, the average pressure within the preset neighborhood range, and the total number of grid points within the preset neighborhood range, wherein the pressure standard deviation is used to measure the local uniformity of the partition; and using the pressure standard deviation as the pressure statistical characteristic parameter.

[0009] Optionally, the step of performing distortion analysis based on the partitioning results to obtain the intake distortion assessment result of the engine to be evaluated includes: for each sub-region in the partitioning results, determining the sub-region distortion value based on the average total pressure recovery coefficient of the low-pressure area of ​​the sub-region and the overall average total pressure recovery coefficient of the sub-region, wherein the partitioning results include at least one sub-region; and performing synthesis processing based on the distortion values ​​of each sub-region to obtain the intake distortion assessment result, wherein the intake distortion assessment result includes at least one of the following: angular parameters, radial parameters, average pressure values, sub-region areas, and distortion values ​​of each sub-region.

[0010] Optionally, before obtaining the intake distortion assessment result of the engine to be evaluated, the method further includes: determining the total distortion value of the intake duct outlet section based on the average total pressure recovery coefficient of the low-pressure zone of the intake duct outlet section and the overall average total pressure recovery coefficient of the intake duct outlet section; determining the sub-region zoning inspection result of each sub-region based on the distortion value of each sub-region, the total distortion value, and the preset constraint conditions; when the sub-region zoning inspection result indicates that the target sub-region zoning does not meet the preset constraint conditions, adjusting the angle parameters and radial parameters of the target sub-region, and recalculating the sub-region distortion value until the target sub-region meets the preset constraint conditions.

[0011] Optionally, acquiring the pressure distribution data of the engine to be evaluated includes: acquiring the original pressure field matrix and operating parameters of the engine to be evaluated; performing gridding processing on the intake duct outlet section to obtain multiple grid points; determining the total pressure value of each grid point based on the original pressure field matrix; performing coordinate system transformation on the total pressure value of each grid point to obtain a transformed total pressure value; and determining the pressure distribution data based on the transformed total pressure value of each grid point and a reference incoming flow total pressure, wherein the reference incoming flow total pressure is determined based on the operating parameters, and the operating parameters include at least the incoming flow Mach number.

[0012] Optionally, the partitioning model includes: an input layer for receiving the intake pressure features; a convolutional layer for extracting the intake pressure features to obtain spatial features; a pooling layer for performing a first dimensionality reduction on the spatial features to obtain dimensionality-reduced features; a flattening layer for performing a second dimensionality reduction on the dimensionality-reduced features to obtain a one-dimensional vector; a fully connected layer for performing classification mapping on the one-dimensional vector to obtain partitioning results; and an output layer for outputting the partitioning results.

[0013] Optionally, after determining the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point, the method further includes: identifying a continuous region where the spatial gradient vector is greater than a preset threshold, the continuous region including at least one of the grid points; and marking the continuous region to obtain the spatial gradient vector carrying a label, the label including at least one of the following: shear layer, high gradient band.

[0014] In a second aspect, embodiments of this disclosure provide an engine intake distortion assessment device, comprising: an acquisition module for acquiring pressure distribution data of an engine to be assessed, wherein the pressure distribution data is pressure data of the intake duct outlet section of the engine to be assessed; an extraction module for extracting features from the pressure distribution data to obtain intake pressure features; a partitioning analysis module for performing partitioning analysis on the intake pressure features based on a pre-trained partitioning model to obtain partitioning results of the intake duct outlet section, wherein the partitioning results characterize the flow field physical partitioning strategy of the intake duct outlet section; and a distortion analysis module for performing distortion analysis based on the partitioning results to obtain an intake distortion assessment result of the engine to be assessed.

[0015] In a third aspect, embodiments of this disclosure provide an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described engine intake distortion assessment method.

[0016] In a fourth aspect, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the above-described engine intake distortion assessment method.

[0017] In a fifth aspect, embodiments of this disclosure provide a computer program product including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device executes the above-described engine intake distortion assessment method.

[0018] In this disclosure, pressure distribution data of the engine to be evaluated is obtained, specifically the pressure data at the inlet outlet section of the engine. Feature extraction is performed on the pressure distribution data to obtain inlet pressure features. Based on a pre-trained partitioning model, the inlet pressure features are partitioned to obtain the partitioning results of the inlet outlet section. These partitioning results characterize the physical partitioning strategy of the flow field at the inlet outlet section. Distortion analysis is then performed based on the partitioning results to obtain the inlet distortion assessment results for the engine. By extracting features from the pressure distribution data at the inlet outlet section, structured coded inlet pressure features are obtained, facilitating subsequent analysis of the correlation between section partitioning and pressure features. The resulting partitioning results are determined by the physical laws of the flow field itself, establishing a direct correspondence between the partitioning results and the physical sources of distortion. This accurately reflects the true physical structure of the flow field, thereby improving the accuracy of distortion assessment results when calculating the distortion of each sub-region. Therefore, the accuracy of engine inlet distortion assessment can be improved.

[0019] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further illustration of the claimed technology. Attached Figure Description

[0020] Figure 1 This is a flowchart of an engine intake distortion assessment method provided in this disclosure.

[0021] Figure 2 This is a diagram illustrating the partition comparison provided in this disclosure.

[0022] Figure 3 This is a schematic diagram of the partitioning model provided in this disclosure.

[0023] Figure 4 This is another flowchart of an engine intake distortion assessment method provided in this disclosure.

[0024] Figure 5 A flowchart for partition optimization provided in this disclosure.

[0025] Figure 6 This is a schematic diagram of the structure of an engine intake distortion assessment device provided in this disclosure.

[0026] Figure 7 This is a hardware block diagram of an electronic device provided in this disclosure.

[0027] Figure 8 This is a schematic diagram of a computer program product provided in this disclosure. Detailed Implementation

[0028] To enable those skilled in the art to better understand the technical solution of this application, the application scenario of this application will be described first below.

[0029] Aero-engines are highly complex and precise thermodynamic machines. As the core power unit of aircraft, their operational stability directly affects flight safety and mission execution capabilities. The inlet outlet is the interface section between the inlet and the engine compressor. Within the entire operating envelope of the engine, the quality of the incoming flow provided by the inlet has a decisive impact on the aerodynamic stability of the engine. It is a critical interface connecting the inlet and the engine compressor, and the flow field characteristics at this section directly determine the operating conditions of the compressor inlet. When there is significant non-uniformity in the total pressure distribution at the inlet outlet section, i.e., total pressure distortion exists, the compressor blades will operate under non-design conditions, with local angle of attack shifting. In severe cases, this can lead to fan or compressor blade stall, and consequently, instability phenomena such as surge in the entire engine. Therefore, the stability of aero-engines is closely related to the uniformity of the incoming flow in its inlet. Accurately quantifying the distortion intensity and spatial distribution at the inlet outlet section is an indispensable foundational step in determining engine stability margins, assessing safety boundaries, and designing control systems.

[0030] Currently, the intake duct outlet section is typically divided into several sub-regions based on manual experience. Distortion indices (such as circumferential and radial distortion indices) are calculated for each sub-region to quantify flow inhomogeneity and assess intake distortion. The zoning method primarily employs fixed geometric partitioning, such as the "equiangular sector partitioning method" or the "equal-area annular partitioning method." However, in this assessment method, the boundaries of the partitions are pre-defined based on simple geometric rules (e.g., one sector every 30 degrees), completely detached from the actual physical structure of the flow field (e.g., shear layers with drastically changing pressure gradients, boundaries of vortex cores). This results in the partitioning results failing to accurately reflect the physical root causes of distortion. Furthermore, due to unreasonable partitioning methods, the calculated sub-region distortion values ​​may fail to capture the worst flow conditions or artificially fragment physically continuous areas of strong distortion, severely underestimating or distorting the actual distortion threat and affecting the accuracy of engine stability boundary assessment. Therefore, this reduces the accuracy of engine intake distortion assessment. In addition, the entire assessment process relies heavily on personal experience for manual zoning adjustments or result interpretation, which is cumbersome, inefficient, and results obtained by different personnel are inconsistent, thus affecting the accuracy of engine intake distortion assessment.

[0031] To address the aforementioned technical problems, this disclosure provides an inventive concept: by extracting features from the pressure distribution data of the inlet outlet section, structured coded inlet pressure features are obtained, facilitating subsequent analysis of the correlation between section partitioning and pressure features. The resulting partitioning is determined by the physical laws of the flow field itself, establishing a direct correspondence between the partitioning results and the physical sources of distortion. This accurately reflects the true physical structure of the flow field, thereby improving the accuracy of distortion assessment results when calculating the distortion of each sub-region. This overcomes the fundamental shortcomings of existing fixed geometric partitioning methods in aero-engine inlet distortion assessment, resolving issues such as low assessment accuracy, unclear physical meaning of results, and high reliance on human experience due to the disconnect between partition boundaries and flow field physical features. It achieves a shift in inlet distortion assessment from "geometric division" to "physical feature-driven" methods, improving the accuracy of engine inlet distortion assessment.

[0032] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present disclosure and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the drawings, not the entire structure.

[0033] Figure 1 This is a flowchart of a method for evaluating engine intake distortion provided in this disclosure. Figure 1 As shown, the method includes: S101: Obtain pressure distribution data of the engine to be evaluated.

[0034] Specifically, pressure distribution data refers to the pressure data at the inlet outlet section of the engine to be evaluated. In other words, pressure distribution data is the set of total pressure values ​​recorded at various locations on a specific spatial plane—the inlet outlet section—in the form of a discrete lattice or a continuous field. It is used to comprehensively describe the spatial distribution of airflow pressure on this section. In the field of aero-engine inlet evaluation, pressure distribution data carries all pressure-related physical information in the flow field, including but not limited to the magnitude and direction of pressure gradients, the location and value of pressure extrema, and the range and intensity of pressure abrupt change regions. This information serves as the data source for subsequent distortion analysis and stability margin assessment. By obtaining complete pressure distribution data at the inlet outlet section, a realistic and comprehensive physical data foundation is provided for subsequent feature extraction and model partitioning, ensuring the physical reliability of the entire evaluation process from the outset.

[0035] S102: Extract features from the pressure distribution data to obtain the intake pressure features.

[0036] Specifically, feature extraction can remove noise and redundant information from pressure distribution data while retaining and enhancing the physical features most relevant to distortion assessment. This achieves information transformation from raw data to physical features, significantly reducing data dimensionality while preserving the core physical information related to distortion assessment to the greatest extent possible. This facilitates subsequent input into the partitioning model for partitioning analysis, reducing the computational burden of the subsequent model while ensuring that the core physical information most relevant to distortion assessment is fully preserved, providing high-quality input for the accurate operation of the partitioning model.

[0037] S103: Based on the pre-trained partitioning model, the intake pressure characteristics are partitioned to obtain the partitioning results of the intake duct outlet section.

[0038] Specifically, the pre-trained partitioning model automatically outputs the optimal partitioning result for the inlet outlet section based on the input inlet pressure characteristics. This partitioning result is based on the real physical structure of the flow field, rather than artificial geometric rules. The partitioning model can quickly infer the location and extent of the physical structure in the current flow field based on learned mapping relationships, and then output the corresponding partitioning result. This process is completely automated and requires no manual intervention. Furthermore, the output partitioning result is no longer a geometric division with equal angles or areas, but a physical partition determined by the physical laws of the flow field itself. It is dynamically generated and can adaptively adapt to changes in the flow field structure under different flight conditions. The pressure distribution within each sub-region is relatively uniform, while there are significant pressure gradient abrupt changes or changes in flow field structure at the boundaries between sub-regions. In this embodiment, the partitioning model can be a convolutional neural network (CNN) model. The partitioning model performs forward propagation calculations and feature classification on the input inlet pressure characteristics and outputs the optimal partitioning result for the current pressure field region. The partitioning result represents the physical partitioning strategy of the flow field at the inlet outlet section.

[0039] By using a pre-trained partitioning model to automatically partition and analyze the intake pressure characteristics, adaptive partitioning results based on the physical structure of the flow field are output. This fundamentally solves the problems of lack of physical meaning and unreliable evaluation accuracy in traditional fixed geometric partitioning. At the same time, it realizes full-process automation, which greatly improves engineering efficiency and result consistency.

[0040] S104: Based on the partitioning results, perform distortion analysis to obtain the intake distortion assessment results of the engine to be evaluated.

[0041] Specifically, distortion analysis refers to the process of calculating distortion indices within each sub-region after physical partitioning, and assessing the engine's stable operating requirements. It is used to quantitatively evaluate the impact of the inlet outlet flow field on engine stability. In this embodiment, the partitioning results ensure that each sub-region physically corresponds to a relatively uniform internal flow field region. The calculation results accurately reflect the distortion intensity of that region, avoiding the problems of dispersed peak distortion and underestimated overall distortion caused by partitioning misalignment in related technologies. Furthermore, the evaluation results are based on actual physical partitioning, accurately capturing adverse conditions in the flow field, making the evaluation conclusions more reliable.

[0042] In this embodiment, partitioning is driven by the physical characteristics of the flow field, ensuring that the evaluation basis aligns with the actual flow field structure and fundamentally improving the accuracy of distortion characterization. The introduction of a CNN model for strategy optimization enables adaptive and intelligent partitioning logic, reducing reliance on human experience. Automated processing from data to evaluation results is achieved, significantly improving engineering analysis efficiency and process standardization, and strongly supporting the digital and intelligent R&D system of modern aero-engines.

[0043] In this disclosure, pressure distribution data of the engine to be evaluated is obtained, specifically the pressure data at the inlet outlet section of the engine. Feature extraction is performed on the pressure distribution data to obtain inlet pressure features. Based on a pre-trained partitioning model, the inlet pressure features are partitioned to obtain the partitioning results of the inlet outlet section. These partitioning results characterize the physical partitioning strategy of the flow field at the inlet outlet section. Distortion analysis is then performed based on the partitioning results to obtain the inlet distortion assessment results for the engine. By extracting features from the pressure distribution data at the inlet outlet section, structured coded inlet pressure features are obtained, facilitating subsequent analysis of the correlation between section partitioning and pressure features. The resulting partitioning results are determined by the physical laws of the flow field itself, establishing a direct correspondence between the partitioning results and the physical sources of distortion. This accurately reflects the true physical structure of the flow field, thereby improving the accuracy of distortion assessment results when calculating the distortion of each sub-region. Therefore, the accuracy of engine inlet distortion assessment can be improved.

[0044] In one possible implementation, an exemplary method for extracting features from pressure distribution data to obtain intake pressure features includes: Based on the pressure distribution data corresponding to each grid point, the spatial gradient vector of each grid point is determined; based on the pressure distribution data corresponding to each grid point, the pressure statistical characteristic parameters within the preset neighborhood range of each grid point are determined; based on the pressure distribution data corresponding to each grid point, the pressure extreme points of each grid point are determined; and the spatial gradient vector, the pressure statistical characteristic parameters, and the pressure extreme points are processed to obtain the intake pressure characteristics.

[0045] Specifically, the spatial gradient vector refers to a two-dimensional vector, centered at a specific grid point, representing the rate of change of pressure values ​​in both the radial and circumferential directions within a gridded coordinate system of the inlet outlet section. It quantitatively describes the spatial direction and rate of change of the pressure field at that grid point. The grid point is obtained by gridding the inlet outlet section, and the preset neighborhood is the region defined by each grid point. In inlet distortion analysis, the spatial gradient vector directly points to the regions in the flow field where pressure changes are most dramatic, such as the location of shear layers or the boundaries of vortex cores. The inlet outlet section is a ring-shaped region, which needs to be discretized into several grid points, each with a pressure value. For each grid point on the inlet outlet section, a spatial gradient vector is calculated using the surrounding pressure distribution data. This provides a direct, efficient, and physically meaningful feature input for subsequent partitioning models to identify physical structures such as shear layers and vortex boundaries. In other words, calculating the spatial gradient vector of the pressure field can identify the regions where flow parameters change most dramatically. The larger the magnitude of the spatial gradient vector, the more drastic the pressure change at that point, and the more likely it is to be in a highly distorted region such as a shear layer or the edge of a vortex; the direction of the gradient vector indicates the direction of the fastest pressure increase or decrease, and can be used to identify structural boundaries in the flow field.

[0046] Pressure statistical parameters refer to a set of descriptive values ​​obtained by statistically calculating the pressure values ​​of all grid points within a predetermined neighborhood centered on a given grid point. These parameters characterize the overall statistical features of the pressure distribution within the local area containing that grid point. In intake distortion analysis, statistical parameters are used to identify whether the pressure is uniform within a local area, and the degree and form of any non-uniformity. For each grid point, a predetermined neighborhood is defined centered on it, and then the statistical parameters of all pressure values ​​within that neighborhood are calculated. Analyzing the statistical parameters of pressure within a local window quantifies the uniformity within the region and can determine whether a region should be divided into an independent sub-region. If the statistical parameters within a region show significant pressure non-uniformity, then that region is not suitable for merging into other regions and should be divided separately. Furthermore, statistical parameters and spatial gradient vectors form a multi-scale complementary description. The gradient vector describes the local rate of change at a single point, while the statistical parameters describe the overall distribution characteristics within a region.

[0047] Pressure extrema points refer to grid points in the gridded coordinate system of the inlet duct outlet section where the pressure value reaches a local maximum or minimum. They are used to identify key locations in the flow field where the pressure is highest or lowest, typically corresponding to the core or boundary of the most representative physical structure in the flow field. The total pressure distortion at the inlet duct outlet section is essentially caused by the existence of local high-pressure and low-pressure regions, and the extrema points are precisely the centers of these high-pressure and low-pressure regions. Accurately identifying the locations of these extrema points is crucial for determining the most severe flow regions in the flow field and assessing the maximum distortion threat faced by the engine. In this embodiment, pressure extrema points include, for example, saddle points and vortex core centers.

[0048] These feature analysis results can be structured into a multi-dimensional physical feature vector or visualized as a feature map, providing direct input for subsequent partitioning analysis of the partitioning model.

[0049] In one possible implementation, an exemplary method for determining the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point includes: Edge detection is performed on the pressure distribution data corresponding to each grid point to obtain the pressure gradient amplitude of each grid point; the pressure gradient amplitude is used as the spatial gradient vector.

[0050] Specifically, the pressure distribution data can be processed using the Sobel operator for edge detection, calculating the pressure gradient magnitude at each grid point to obtain the gradient field. In this embodiment, the pressure gradient magnitude is defined in polar coordinates.

[0051] The following is an example formula for calculating the magnitude of the pressure gradient:

[0052] in, This represents the pressure gradient amplitude. For the pressure distribution data of the grid points, For the radial parameters of the grid points, The angle parameter for the grid points.

[0053] In one possible implementation, an exemplary method for determining the pressure statistical characteristic parameters within a preset neighborhood range of each grid point based on the pressure distribution data corresponding to each grid point includes: Obtain pressure distribution data within a preset neighborhood range for each grid point; determine the average pressure within the preset neighborhood range based on the pressure distribution data within the preset neighborhood range; determine the pressure standard deviation within the preset neighborhood range based on the pressure distribution data corresponding to each grid point, the average pressure within the preset neighborhood range, and the total number of grid points within the preset neighborhood range; use the pressure standard deviation as a pressure statistical characteristic parameter.

[0054] Specifically, the pressure standard deviation within a preset neighborhood of each grid point is calculated as a pressure statistical characteristic parameter. The pressure standard deviation is used to measure the local uniformity of the partition. In this embodiment, the preset neighborhood can be set to a 5×5 range.

[0055]

[0056] in, Represented by grid points The surrounding area centered on, Neighborhood range The total number of grid points within, Neighborhood range The average pressure inside, This represents the standard deviation of pressure.

[0057] In addition, the pressure gradient magnitude and pressure standard deviation can together form a two-channel feature map F, which serves as one of the inlet pressure characteristics characterizing the flow field structure.

[0058] In one possible implementation, an exemplary method for obtaining the intake distortion assessment results of the engine to be evaluated by performing distortion analysis based on the partitioning results includes: For each sub-region in the zoning results, the sub-region distortion value is determined based on the average total pressure recovery coefficient of the low-pressure area of ​​the sub-region and the overall average total pressure recovery coefficient of the sub-region. The distortion value of each sub-region is then synthesized to obtain the intake distortion assessment result.

[0059] Specifically, the sub-region distortion value refers to the evaluation index calculated separately for each sub-region after the inlet outlet section is physically partitioned. It quantifies the degree of non-uniformity in total pressure recovery within that sub-region and is used to accurately characterize the local impact of that physical region on the quality of the engine compressor inlet flow field. The total pressure recovery coefficient is the ratio of the total pressure in a certain region of the inlet outlet section to the total pressure of the incoming flow at the inlet. A value closer to -1 indicates less energy loss and better flow field quality in that region, while a smaller value indicates greater energy loss and worse flow field quality. The partitioning result includes at least one sub-region.

[0060] Within each sub-region, the local area with the lowest total pressure is identified. The average total pressure recovery coefficient of this low-pressure area is calculated by comparing it with the average total pressure recovery coefficient of the entire sub-region. The degree of difference between the two is the distortion value of that sub-region. By comparing the average total pressure recovery coefficient of the low-pressure area with that of the whole, the most severe flow conditions within the sub-region are accurately captured, overcoming the problem of loss of evaluation accuracy caused by spatial averaging of the overall distortion index in related technologies.

[0061] The following is an example formula for calculating the distortion value of a sub-region:

[0062] in, For the first Sub-region distortion value of each sub-region This represents the average total pressure recovery coefficient of the low-pressure area within this sub-region. This is the overall average total pressure recovery coefficient for this sub-region.

[0063] Next, the intake distortion results corresponding to the final partitioning results are organized into a structured, directly usable format. In this embodiment, the generated intake distortion assessment report typically includes: the precise definition of each sub-partition (radial start and end radii and circumferential start and end angles in polar coordinates), geometric area, average total pressure within the region, and calculated distortion value. That is, the intake distortion assessment results include at least one of the following: angular parameters (e.g., start angle, end angle), radial parameters (e.g., maximum radial parameter, minimum radial parameter), average pressure value, sub-partition area, and distortion value for each sub-partition.

[0064] Furthermore, the report can be output in both text data files (e.g., txt format) and visual graphics, seamlessly integrating into subsequent design processes. Among these, visual graphics include... Figure 2 As shown, Figure 2 This is a diagram illustrating the partitioning comparison provided in this disclosure. The left side shows a diagram of intermediate-angle partitioning in related technologies, while the right side shows a diagram of adaptive partitioning provided in this disclosure.

[0065] Furthermore, in one possible implementation, if the partitioning model outputs a "circumferential subdivision" partitioning result for a certain sector region in the annular cross section, a new radial dividing line can be inserted at the central corner of this region to further refine the partitioning of the sub-region.

[0066] In one possible implementation, before obtaining the intake distortion assessment results of the engine to be evaluated, the method further includes: Based on the average total pressure recovery coefficient of the low-pressure zone of the inlet outlet section and the overall average total pressure recovery coefficient of the inlet outlet section, the total distortion value of the inlet outlet section is determined. Based on the distortion value of each sub-region, the total distortion value, and the preset constraints, the sub-region zoning inspection results of each sub-region are determined. When the sub-region zoning inspection results indicate that the target sub-region does not meet the preset constraints, the angular parameters and radial parameters of the target sub-region are adjusted, and the sub-region distortion value is recalculated until the target sub-region meets the preset constraints.

[0067] Specifically, the total distortion value refers to a comprehensive distortion index obtained by comparing the average total pressure recovery coefficient of the low-pressure area within the inlet outlet section with the overall average total pressure recovery coefficient of the entire section at the global level. It is used to assess the non-uniformity of total pressure recovery on the inlet outlet section from a global perspective and serves as a benchmark reference value for subsequent zoning inspections of each sub-section.

[0068] Before proceeding to the sub-region inspection and adjustment phase, a global reference baseline must be established. The zonal inspection needs to determine whether the distortion values ​​of each sub-region are within a reasonable range, and this determination cannot be divorced from the global context. It is understandable that the calculation method for the global total distortion value is the same as that for the sub-region distortion value; its calculation scope is the entire intake duct outlet cross-section, which will not be elaborated upon here.

[0069] The distortion value of each sub-region is compared with the total distortion value of the entire inlet section to check whether the preset constraints are met. The following is an example formula for the preset constraints:

[0070] in, The distortion value of the sub-region. This represents the total distortion value. This is a configurable scaling factor. In this embodiment, The default value is preferably 0.8, but it can be understood that it can be adjusted according to specific circumstances.

[0071] If a subregion does not meet the above preset constraints, a boundary optimization algorithm will be automatically initiated to optimize the target subregion. Based on the pressure gradient and pressure standard deviation within the subregion, one or more boundaries are moved radially or circumferentially with small iteration steps. For example, the gradient direction within the target subregion can be analyzed, and the boundary can be finely adjusted by a small angle increment along the gradient descent direction (i.e., from the low-pressure side to the high-pressure side). (e.g., 0.5 degrees) or radial increment Then, the distortion value is recalculated to reduce the distortion level within the sub-region. This process continues until all sub-regions meet the preset constraints, or the preset maximum number of iterations or minimum partition size (e.g., 5 degrees) is reached.

[0072] In one possible implementation, an exemplary method for obtaining pressure distribution data of the engine to be evaluated includes: Obtain the original pressure field matrix and operating parameters of the engine to be evaluated; perform gridding on the intake duct outlet section to obtain multiple grid points; determine the total pressure value of each grid point based on the original pressure field matrix; perform coordinate system transformation on the total pressure value of each grid point to obtain the transformed total pressure value; determine the pressure distribution data based on the transformed total pressure value of each grid point and the reference incoming flow total pressure.

[0073] Specifically, it can receive raw pressure field matrices and related operating parameters from external computational fluid dynamics (CFD) simulation results or experimental measurement devices. The simulation results characterize the pressure field data, and the raw pressure field matrix is ​​typically defined on a ring or circular cross-section. The total pressure value of each grid point.

[0074] The original pressure field matrix is ​​subjected to engineering-usable standardization processing, including but not limited to: coordinate system transformation (e.g., transformation from rectangular coordinates to polar coordinates to adapt to the annular section), identification and removal of invalid or abnormal data points, and finally outputting a clean, normalized two-dimensional transformed total pressure value.

[0075] The converted total pressure value is normalized to the reference incoming total pressure, and the final normalized pressure distribution data is output. The reference incoming total pressure is determined based on operating parameters, which include at least the incoming Mach number.

[0076] The following is an example formula for determining pressure distribution data:

[0077] in, For grid points Total pressure value at the conversion point, For reference, the total incoming pressure, For grid points Pressure distribution data at the location.

[0078] In one possible implementation, the partitioning model includes: The input layer receives intake pressure features; the convolutional layer extracts the intake pressure features to obtain spatial features; the pooling layer performs a first dimensionality reduction on the spatial features to obtain dimensionality-reduced features; the flattening layer performs a second dimensionality reduction on the dimensionality-reduced features to obtain a one-dimensional vector; the fully connected layer performs classification mapping on the one-dimensional vector to obtain the partitioning results; and the output layer outputs the partitioning results.

[0079] Specifically, in this embodiment, a partitioning model can be constructed based on a convolutional neural network (CNN). Intake pressure features are fed into this partitioning model and, after forward propagation, the category with the highest probability in the output layer is adopted as the optimal partitioning strategy for that region. The partitioning model automatically performs deep feature learning and abstraction on the feature map through its multi-layered structure (convolutional layers, activation function layers, and pooling layers), identifying the implicit complex patterns related to the ideal partitioning strategy.

[0080] Figure 3 This is a schematic diagram of the partitioning model provided in this disclosure, such as... Figure 3As shown, the partitioning model includes an input layer 301 for receiving intake pressure features; two convolutional-pooling layers, where convolutional layer 302 uses a 3×3 convolutional kernel to extract spatial features, followed by a ReLU activation function layer 303 and a 2×2 max pooling layer 304 for dimensionality reduction to obtain dimensionality-reduced features; then a flattening layer 305 is connected to obtain a one-dimensional vector, and a fully connected layer 306 is used for classification mapping to obtain the partitioning result; finally, the output layer 307 has three neurons, corresponding to the probabilities of three partitioning strategy categories. In this embodiment, the three partitioning strategy categories are preservation, radial subdivision, and circumferential subdivision.

[0081] For example, partitioning strategy category 0 indicates "the current area is uniform, it is recommended to keep it and not subdivide it"; partitioning strategy category 1 indicates "there is significant radial pressure stratification within the area, it is recommended to subdivide it radially"; partitioning strategy category 2 indicates "there is circumferential non-uniformity within the area, it is recommended to subdivide it circumferentially". In addition, the partitioning model was trained using a large amount of historical operating data, and the labels in the training data were generated by fluid mechanics experts using the optimal partitioning strategy based on the physical structure of the flow field.

[0082] In one possible implementation, after determining the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point, the method further includes: Identify continuous regions whose spatial gradient vectors are greater than a preset threshold; label these continuous regions to obtain spatial gradient vectors carrying labels.

[0083] Specifically, by identifying continuous regions where the spatial gradient vector exceeds a preset threshold, high-gradient connected regions that truly reflect strong physical structures are accurately extracted from the full gradient information. This effectively separates physical signals from noise signals in principle, providing high signal-to-noise ratio and high information density structural boundary location information for subsequent partitioning models. The continuous region includes at least one grid point, and the label includes at least one of the following: shear layer, high-gradient band.

[0084] Figure 4 This is another flowchart of an engine intake distortion assessment method provided in this disclosure. Figure 4 As shown, the method includes: S401: Data input and preprocessing.

[0085] Specifically, the original pressure field matrix and operating parameters (such as the incoming Mach number) of the engine intake duct outlet section are obtained; the original pressure field matrix is ​​standardized by coordinate system normalization and invalid data removal to obtain standardized two-dimensional pressure distribution data.

[0086] S402: Extraction of physical features of the flow field.

[0087] Specifically, spatial gradient calculation, local uniformity statistical analysis, and feature region identification are performed on the standardized pressure distribution data to extract vectors or feature maps that characterize the inlet pressure features of the non-uniform flow field structure.

[0088] S403: Intelligent partitioning strategy decision.

[0089] Specifically, the intake pressure features are input into a pre-trained classification model. The classification model performs forward propagation calculations and feature classification on the input feature map, and outputs the optimal partitioning strategy category for the current pressure field region. The strategy categories include radial subdivision, circumferential subdivision, or no subdivision.

[0090] S404: Perform partitioning and closed-loop optimization.

[0091] Figure 5 A flowchart for partition optimization provided in this disclosure. For example... Figure 5 As shown, the method includes: S501: Perform initial geometric partitioning according to the strategy.

[0092] Specifically, based on the output decision category, a specific geometric partitioning operation is performed at the corresponding location of the pressure field to form an initial set of sub-regions.

[0093] S502: Calculate the distortion value of each sub-region and the total distortion value.

[0094] Specifically, the distortion value of each sub-region in the initial set is calculated, and the total global distortion value is calculated. It is then determined whether the distortion value of each sub-region fully satisfies the preset constraint condition, i.e., the sub-region distortion value ≤ × Total distortion value.

[0095] S503: Determine whether all sub-regions satisfy the constraints.

[0096] Specifically, if yes, proceed to S405; otherwise, proceed to S504. That is, if there is a sub-region that does not meet the constraints, optimization is initiated, and its geometric boundary is automatically adjusted according to the flow field characteristics inside the sub-region.

[0097] S504: Adjust sub-region.

[0098] Specifically, for sub-regions that do not meet the conditions, the boundaries are fine-tuned along the direction of reducing internal distortion, and then returned to S502 for recalculation and evaluation.

[0099] Convergence and Exit: When all sub-partitions satisfy the constraints or other recursive termination conditions are met, the optimization loop ends and the final partitioning scheme is obtained.

[0100] S405: Integrate the final partition results.

[0101] Specifically, the partitioning results are integrated to obtain the intake distortion assessment results, and a report and visualization map in txt format are output.

[0102] Figure 6 This is a schematic diagram of the structure of an engine intake distortion assessment device provided in this disclosure. Figure 6 As shown, the device 600 includes: an acquisition module 610, an extraction module 620, a partition analysis module 630, and a distortion analysis module 640.

[0103] The acquisition module 610 is used to acquire pressure distribution data of the engine to be evaluated, wherein the pressure distribution data is the pressure data of the intake duct outlet section of the engine to be evaluated; the extraction module 620 is used to extract features from the pressure distribution data to obtain intake pressure features; the partitioning analysis module 630 is used to perform partitioning analysis on the intake pressure features based on a pre-trained partitioning model to obtain the partitioning results of the intake duct outlet section, wherein the partitioning results characterize the flow field physical partitioning strategy of the intake duct outlet section; and the distortion analysis module 640 is used to perform distortion analysis based on the partitioning results to obtain the intake distortion evaluation results of the engine to be evaluated.

[0104] Optionally, the extraction module includes: a first determining submodule, used to determine the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point, wherein the grid point is obtained by gridding the intake duct outlet section; a second determining submodule, used to determine the pressure statistical characteristic parameters within a preset neighborhood range of each grid point based on the pressure distribution data corresponding to each grid point, wherein the preset neighborhood range is a region range determined with each grid point as the center; a third determining submodule, used to determine the pressure extreme points of each grid point based on the pressure distribution data corresponding to each grid point; and a processing submodule, used to process each spatial gradient vector, each pressure statistical characteristic parameter, and each pressure extreme point to obtain the intake pressure characteristics.

[0105] Optionally, the first determining submodule is configured to: perform edge detection on the pressure distribution data corresponding to each grid point to obtain the pressure gradient amplitude of each grid point; and use the pressure gradient amplitude as the spatial gradient vector.

[0106] Optionally, the second determining submodule is configured to: acquire pressure distribution data within a preset neighborhood range of each of the grid points; determine the average pressure within the preset neighborhood range based on the pressure distribution data within the preset neighborhood range; determine the pressure standard deviation within the preset neighborhood range based on the pressure distribution data corresponding to each of the grid points, the average pressure within the preset neighborhood range, and the total number of grid points within the preset neighborhood range, wherein the pressure standard deviation is used to measure the local uniformity of the partition; and use the pressure standard deviation as the pressure statistical characteristic parameter.

[0107] Optionally, the distortion analysis module is used to: determine the sub-region distortion value of each sub-region in the partitioning result based on the average total pressure recovery coefficient of the low-pressure area of ​​the sub-region and the overall average total pressure recovery coefficient of the sub-region, wherein the partitioning result includes at least one sub-region; and perform synthesis processing based on the distortion values ​​of each sub-region to obtain the intake distortion assessment result, wherein the intake distortion assessment result includes at least one of the following: angular parameters, radial parameters, average pressure values, sub-region areas, and distortion values ​​of each sub-region.

[0108] Optionally, the device further includes: before obtaining the intake distortion assessment result of the engine to be evaluated, determining the total distortion value of the intake duct outlet section based on the average total pressure recovery coefficient of the low-pressure zone of the intake duct outlet section and the overall average total pressure recovery coefficient of the intake duct outlet section; determining the sub-region zoning inspection result of each sub-region based on the distortion value of each sub-region, the total distortion value, and the preset constraint conditions; when the sub-region zoning inspection result indicates that the target sub-region zoning does not meet the preset constraint conditions, adjusting the angle parameters and radial parameters of the target sub-region, and recalculating the sub-region distortion value until the target sub-region meets the preset constraint conditions.

[0109] Optionally, the acquisition module is used to: acquire the original pressure field matrix and operating parameters of the engine to be evaluated; perform gridding processing on the intake duct outlet section to obtain multiple grid points; determine the total pressure value of each grid point based on the original pressure field matrix; perform coordinate system transformation on the total pressure value of each grid point to obtain a transformed total pressure value; and determine the pressure distribution data based on the transformed total pressure value of each grid point and the reference incoming flow total pressure, wherein the reference incoming flow total pressure is determined based on the operating parameters, and the operating parameters include at least the incoming flow Mach number.

[0110] Optionally, the partitioning model of the device includes: an input layer for receiving the intake pressure features; a convolutional layer for extracting the intake pressure features to obtain spatial features; a pooling layer for performing a first dimensionality reduction on the spatial features to obtain dimensionality-reduced features; a flattening layer for performing a second dimensionality reduction on the dimensionality-reduced features to obtain a one-dimensional vector; a fully connected layer for performing classification mapping on the one-dimensional vector to obtain partitioning results; and an output layer for outputting the partitioning results.

[0111] Optionally, the device further includes: after determining the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point, identifying a continuous region where the spatial gradient vector is greater than a preset threshold, the continuous region including at least one of the grid points; and performing a labeling process on the continuous region to obtain the spatial gradient vector carrying a label, the label including at least one of the following: shear layer, high gradient band.

[0112] This application also provides an electronic device for performing the above-described engine intake distortion assessment method. Please refer to... Figure 7 It illustrates a schematic diagram of an electronic device provided by some embodiments of this application. For example... Figure 7 As shown, the electronic device 70 includes: a processor 700, a memory 701, a bus 702, and a communication interface 703. The processor 700, the communication interface 703, and the memory 701 are connected via the bus 702. The memory 701 stores a computer program that can run on the processor 700. When the processor 700 runs the computer program, it executes the engine intake distortion assessment method provided in any of the foregoing embodiments of this application.

[0113] The memory 701 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between the device network element and at least one other network element is achieved through at least one communication interface 703 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0114] Bus 702 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. Memory 701 is used to store programs. After receiving an execution instruction, processor 700 executes the program. The engine intake distortion evaluation method disclosed in any of the foregoing embodiments of this application can be applied to processor 700, or implemented by processor 700.

[0115] The processor 700 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 700 or by instructions in software form. The processor 700 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 701. Processor 700 reads the information in memory 701 and, in conjunction with its hardware, completes the steps of the above method.

[0116] The electronic equipment provided in this application embodiment and the engine intake distortion assessment method provided in this application embodiment are based on the same inventive concept and have the same beneficial effects as the methods they adopt, operate or implement.

[0117] This application also provides a computer-readable storage medium corresponding to the engine intake distortion assessment method provided in the foregoing embodiments. The computer-readable storage medium shown can be an optical disc, on which a computer program is stored. When the computer program is run by a processor, it executes the engine intake distortion assessment method provided in any of the foregoing embodiments.

[0118] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0119] The computer-readable storage medium provided in the above embodiments of this application and the engine intake distortion assessment method provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0120] This application also provides a computer program product 800, such as... Figure 8 As shown. This computer program product carries a computer program 801, the instructions of which can be used to execute the steps of the engine intake distortion assessment method described in the above method embodiments, as detailed in the above method embodiments, which will not be repeated here.

[0121] The aforementioned computer program product can be implemented through hardware, software, or a combination thereof. In one optional embodiment, the computer program product is specifically embodied in a computer storage medium; in another optional embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0122] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0123] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0124] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0125] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0126] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0127] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0128] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method for evaluating engine intake distortion, characterized in that, include: Obtain pressure distribution data of the engine to be evaluated, wherein the pressure distribution data is the pressure data of the intake duct outlet section of the engine to be evaluated; Feature extraction is performed on the pressure distribution data to obtain the intake pressure features; The intake pressure characteristics are analyzed by partitioning based on the pre-trained partitioning model to obtain the partitioning results of the intake duct outlet section. The partitioning results characterize the flow field physical partitioning strategy of the intake duct outlet section. Distortion analysis is performed based on the partitioning results to obtain the intake distortion assessment results of the engine to be evaluated.

2. The method according to claim 1, characterized in that, The step of extracting features from the pressure distribution data to obtain intake pressure features includes: Based on the pressure distribution data corresponding to each grid point, the spatial gradient vector of each grid point is determined. The grid point is obtained by meshing the inlet outlet section. Based on the pressure distribution data corresponding to each grid point, the pressure statistical characteristic parameters within a preset neighborhood range of each grid point are determined, wherein the preset neighborhood range is the area range determined with each grid point as the center; Based on the pressure distribution data corresponding to each grid point, the pressure extreme points of each grid point are determined; The intake pressure characteristics are obtained by processing each of the spatial gradient vectors, each of the pressure statistical characteristic parameters, and each of the pressure extreme points.

3. The method according to claim 2, characterized in that, The determination of the spatial gradient vector for each grid point based on the pressure distribution data corresponding to each grid point includes: Edge detection is performed on the pressure distribution data corresponding to each grid point to obtain the pressure gradient amplitude of each grid point; The pressure gradient magnitude is used as the spatial gradient vector.

4. The method according to claim 2, characterized in that, The step of determining the pressure statistical characteristic parameters within a preset neighborhood range of each grid point based on the pressure distribution data corresponding to each grid point includes: Obtain pressure distribution data within a preset neighborhood range of each grid point; Based on the pressure distribution data within the preset neighborhood range, the average pressure within the preset neighborhood range is determined; Based on the pressure distribution data corresponding to each grid point, the average pressure within the preset neighborhood range, and the total number of grid points within the preset neighborhood range, the pressure standard deviation within the preset neighborhood range is determined. The pressure standard deviation is used to measure the local uniformity of the partition. The pressure standard deviation is used as the pressure statistical characteristic parameter.

5. The method according to claim 1, characterized in that, The distortion analysis based on the partitioning results to obtain the intake distortion assessment result of the engine to be evaluated includes: For each sub-region in the partitioning result, the sub-region distortion value of the sub-region is determined based on the average total pressure recovery coefficient of the low-pressure area of ​​the sub-region and the overall average total pressure recovery coefficient of the sub-region. The partitioning result includes at least one sub-region. The intake distortion assessment result is obtained by synthesizing the distortion values ​​of each sub-region. The intake distortion assessment result includes at least one of the following: angular parameters, radial parameters, average pressure values, sub-region areas, and distortion values ​​of each sub-region.

6. The method according to claim 5, characterized in that, Before obtaining the intake distortion assessment result of the engine to be evaluated, the method further includes: The total distortion value of the intake duct outlet section is determined based on the average total pressure recovery coefficient of the low-pressure zone of the intake duct outlet section and the overall average total pressure recovery coefficient of the intake duct outlet section. Based on the distortion values ​​of each sub-region, the total distortion value, and preset constraints, the sub-region partitioning inspection results of each sub-region are determined; When the sub-region zoning inspection result indicates that the target sub-region does not meet the preset constraint conditions, the angular and radial parameters of the target sub-region are adjusted, and the sub-region distortion value is recalculated until the target sub-region meets the preset constraint conditions.

7. The method according to claim 1, characterized in that, The acquisition of pressure distribution data for the engine to be evaluated includes: Obtain the original pressure field matrix and operating parameters of the engine to be evaluated; The intake duct outlet section is meshed to obtain multiple grid points; Based on the original pressure field matrix, the total pressure value of each grid point is determined; The coordinate system is transformed for the total pressure value of each grid point to obtain the transformed total pressure value; The pressure distribution data is determined based on the converted total pressure value and the reference incoming total pressure at each of the grid points. The reference incoming total pressure is determined based on the operating parameters, which include at least the incoming Mach number.

8. The method according to claim 1, characterized in that, The partitioning model includes: The input layer is used to receive the intake pressure characteristics; A convolutional layer is used to extract the intake pressure features to obtain spatial features; A pooling layer is used to perform a first dimensionality reduction process on the spatial features to obtain dimensionality-reduced features. A flattening layer is used to perform a second dimensionality reduction process on the dimensionality-reduced features to obtain a one-dimensional vector. A fully connected layer is used to perform classification mapping on the one-dimensional vector to obtain the partitioning result; The output layer is used to output the partitioning results.

9. The method according to claim 2, characterized in that, After determining the spatial gradient vector of each grid point based on the pressure distribution data corresponding to each grid point, the method further includes: Identify a continuous region where the spatial gradient vector is greater than a preset threshold, wherein the continuous region includes at least one of the grid points; The continuous region is labeled to obtain the spatial gradient vector carrying the label, wherein the label includes at least one of the following: a shear layer and a high gradient band.

10. An engine intake distortion assessment device, characterized in that, include: The acquisition module is used to acquire pressure distribution data of the engine to be evaluated, wherein the pressure distribution data is the pressure data of the intake manifold outlet section of the engine to be evaluated. The extraction module is used to extract features from the pressure distribution data to obtain intake pressure features; The partitioning analysis module is used to perform partitioning analysis on the intake pressure characteristics based on a pre-trained partitioning model to obtain the partitioning results of the intake duct outlet section. The partitioning results characterize the flow field physical partitioning strategy of the intake duct outlet section. The distortion analysis module is used to perform distortion analysis based on the partitioning results to obtain the intake distortion evaluation results of the engine to be evaluated.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method as described in any one of claims 1-9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by a processor to implement the method as described in any one of claims 1-9.

13. A computer program product, characterized in that, Includes computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the method as described in any one of claims 1-9.

Images