Fusion phase-driven rotating flow field pressure pulsation multi-scale analysis method and system

Abstract

This invention discloses a multi-scale analysis method and system for pressure fluctuations in a rotating flow field using phase-driven fusion. First, unsteady pressure data of the rotating flow field is acquired and a spatiotemporal snapshot matrix is ​​constructed. Then, the dominant frequency and low-order coherent modes of the pressure fluctuations are extracted through dynamic mode decomposition. Subsequently, a global phase field of the entire flow field is constructed based on the temporal information of the modes. Finally, spatial phase gradient analysis is performed on the phase field to determine the propagation direction of the pressure wave. The global multi-scale phase field is traversed to identify characteristic points of phase changes, and the characteristic points of all dominant frequencies are spatially superimposed. Finally, the spatial modes of each characteristic frequency, the global multi-scale phase field, the propagation direction of the pressure wave, and the location of the core excitation source are superimposed and fused. The modes and phase fields at different times are combined according to a time sequence to synthesize a spatiotemporal evolution holographic map of pressure fluctuations in the rotating flow field. This invention achieves precise localization of the pressure fluctuation excitation source and frequency-space-time correlation analysis of pressure fluctuations in the rotating flow field.

Description

Technical Field

[0001] This invention relates to the field of fluid machinery flow analysis and optimization, specifically to a multi-scale analysis method and system for pressure pulsation in a rotating flow field that integrates phase-driven principles. Background Technology

[0002] Rotating flow fields are widely present in rotating fluid machinery such as centrifugal pumps, water pump turbines, and turbines. The dynamic and static interference effects between the rotor and stator inside the flow field can easily induce strong pressure pulsations. Such pressure pulsations can not only cause unit vibration, noise, and even fatigue damage to solid structures, but also reduce the energy conversion efficiency of the flow field and affect the long-term operational stability of the equipment. Therefore, analyzing the generation mechanism, spatiotemporal evolution law, and propagation path of pressure pulsations in rotating flow fields is the key to optimizing the structural design of rotating fluid machinery and improving operational reliability.

[0003] Currently, the mainstream methods for analyzing pressure fluctuations in rotating flow fields mainly include the Fast Fourier Transform (FFT) and Dynamic Mode Decomposition (DMD). The FFT is a traditional method for frequency domain analysis of pressure fluctuations, capable of identifying the dominant frequency components and corresponding amplitudes of pressure fluctuations in the flow field. DMD, as a data-driven flow field order reduction method, can extract the spatially coherent structure corresponding to specific frequencies from unsteady flow data of rotating flow fields, becoming an important technical means for analyzing the unsteady flow characteristics of flow fields.

[0004] However, the above two methods for analyzing pressure fluctuations in rotating flow fields have the following shortcomings:

[0005] On the one hand, the FFT method can only achieve correlation analysis between frequency and amplitude, but cannot establish the correspondence between frequency characteristics and the actual spatial flow structure of the flow field, let alone characterize the dynamic evolution of pressure pulsation in the rotating flow field with time and space, and it is difficult to explain the generation and propagation mechanism of pressure pulsation from the perspective of flow field structure.

[0006] On the other hand, while traditional DMD methods can extract spatially coherent modes corresponding to frequencies, the analysis only focuses on the static spatial distribution or energy ranking of modes, failing to fully explore the phase and temporal information implicit in the decomposition process. It lacks a mechanism for modal phase extraction and reconstruction, and cannot quantitatively characterize the relative lead and lag relationships of pressure pulsations in different spatial regions, making it difficult to reveal the spatiotemporal evolution of coherent structures. Furthermore, existing technologies have not yet established a system for determining pressure wave propagation characteristics based on DMD analysis. There is a lack of precise criteria to determine the direction of pressure wave propagation, and it is impossible to locate the core excitation source of pressure pulsations by identifying abrupt phase changes and gradient concentration regions in the flow field. It is also difficult to distinguish between energy accumulation and attenuation regions of pressure waves. This results in an inability to fully analyze the entire process of pressure pulsations in rotating flow fields from excitation and propagation to dissipation, making it difficult to provide accurate theoretical basis and technical support for vibration reduction, noise reduction, and flow stability optimization design in rotating flow fields. Summary of the Invention

[0007] To address the shortcomings of existing methods for analyzing pressure pulsations in rotating flow fields, such as Fast Fourier Transform (FFT) which can only identify frequency and amplitude characteristics and cannot correlate with the actual spatial flow structure, and Traditional Dynamic Mode Decomposition (DMD) which only focuses on the static spatial distribution of modes and does not explore the implicit phase and temporal information, and lacks effective criteria to accurately determine the direction of pressure wave propagation and locate the core excitation source of pressure pulsations, thus failing to fully analyze the entire process mechanism of pressure pulsation "excitation-propagation-dissipation", this invention proposes a phase-driven multi-scale analysis method and system for pressure pulsations in rotating flow fields, solving the technical problem of the difficulty in quantitatively analyzing the spatiotemporal evolution law of pressure pulsations in rotating flow fields.

[0008] The objective of this invention is achieved through the following technical solution:

[0009] A multi-scale analysis method for pressure fluctuations in a rotating flow field based on phase-driven principles, comprising:

[0010] S1: Acquire unsteady pressure data, sample to obtain a continuous and stable flow field pressure snapshot sequence, and construct a pressure field spatiotemporal snapshot matrix;

[0011] S2: Perform dynamic mode decomposition on the spatiotemporal snapshot matrix of the pressure field to extract the characteristic frequencies, low-order coherent spatial modes and corresponding DMD complex modes of the dominant pressure fluctuations;

[0012] S3: Construct the flow signal matrix of grid nodes with the same frequency, extract the initial spatial phase of the entire flow field directly through argument operation based on DMD complex modes, calculate the phase spatial gradient, generate a global phase field with a single characteristic frequency, and combine the global phase fields of all dominant frequencies to obtain the global multi-scale phase field.

[0013] S4: Define the analysis path in the key region of the rotating flow field, extract the phase distribution on each analysis path and calculate the one-dimensional phase gradient of the path direction, and determine the propagation direction of the pressure wave based on the phase gradient criterion; traverse the global multi-scale phase field and gradient field, identify phase change feature points, and spatially superimpose all dominant frequency feature points to achieve pressure pulsation source tracing; at the same time, divide the pressure wave energy region by combining the phase gradient distribution characteristics.

[0014] S5: The spatial modes of each characteristic frequency, the global multi-scale phase field, the propagation direction of the pressure wave, and the location of the core excitation source are superimposed and fused on a unified rotating flow field geometric model. The mode and phase field data at different times are combined according to the time series to synthesize a holographic map of the spatiotemporal evolution of pressure pulsation in the rotating flow field.

[0015] Further, step S3 includes the following sub-steps:

[0016] S3.1: For each dominant characteristic frequency corresponding to the DMD complex mode, the initial spatial phase of each grid point in the entire flow field is directly extracted by complex argument calculation;

[0017] S3.2: Calculate the spatial gradient of the phase field to obtain the vector field characterizing the evolution direction of the coherent structure;

[0018] S3.3: Assign the phase angles of all extracted grid points to the spatial grid nodes of the DMD mode one by one to generate a global phase field covering the entire computational domain at a single characteristic frequency;

[0019] S3.4: Repeat the phase extraction, phase difference calculation and phase assignment operations of S3.1~S3.3 for all dominant characteristic frequencies to construct a global multi-scale phase field that can reflect the phase information of perturbation waves at different scales.

[0020] Furthermore, in S4, the direction of pressure wave propagation is determined based on the positive and negative characteristics of the one-dimensional phase gradient of the path:

[0021] When the one-dimensional phase gradient along the path direction is greater than 0, the phase increases along the path direction, and the pressure wave propagation direction is opposite to the analysis path direction.

[0022] When the one-dimensional phase gradient along the path direction is less than 0, the phase increases along the path direction, and the pressure wave propagation direction is the same as the analysis path direction.

[0023] Furthermore, in S4, the characteristic points of phase change include phase gradient reversal points, phase abrupt change points, and phase local extrema points.

[0024] Furthermore, the key regions of the rotating flow field include the rotor-stator interface, the bladeless region, and the flow channel centerline.

[0025] Furthermore, the analysis path includes a circumferential path and an axial flow path that runs through the rotor and stator.

[0026] Furthermore, the pressure wave energy region includes an energy accumulation region and an energy attenuation region.

[0027] Furthermore, the unsteady pressure data is obtained through unsteady numerical simulation or particle image velocimetry experiments.

[0028] A phase-driven multi-scale analysis system for pressure fluctuations in rotating flow fields includes one or more processors for implementing a phase-driven multi-scale analysis method for pressure fluctuations in rotating flow fields.

[0029] An electronic device, comprising:

[0030] One or more processors;

[0031] A storage device for storing one or more programs that, when executed by the electronic device, enable the electronic device to implement a multi-scale analysis method for pressure pulsation in a fused phase-driven rotating flow field.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] 1. Breaking through the limitations of traditional DMD analysis, this paper is the first to organically combine the DMD modal spatial structure with phase temporal information. Through the construction of a global multi-scale phase field, the paper realizes the correlation analysis of the frequency-space-time three dimensions of pressure pulsation in the rotating flow field, and establishes a direct correspondence between frequency characteristics and the actual flow structure of the flow field.

[0034] 2. An innovative pressure wave propagation criterion based on phase gradient was established, which can identify the propagation direction of pressure waves and locate the excitation source. At the same time, the characteristics of pressure wave energy accumulation and attenuation were accurately analyzed, and the entire mechanism of pressure pulsation from excitation, propagation to dissipation was completely analyzed, filling the technical gap in the quantitative analysis of pressure pulsation propagation characteristics in rotating flow fields.

[0035] 3. The generated spatiotemporal evolution holographic map intuitively displays the dynamic evolution process of multi-frequency pressure pulsation, providing direct theoretical basis and engineering guidance for flow field optimization, structural improvement, vibration reduction and noise reduction design of rotating fluid machinery such as centrifugal pumps, water pump turbines, and turbines. It is applicable to multiple fields such as energy, chemical industry, agriculture, and nuclear energy. Attached Figure Description

[0036] Figure 1 This is a flowchart of a multi-scale analysis method for pressure fluctuations in a rotating flow field driven by fusion phase, according to an embodiment of the present invention.

[0037] Figure 2 This is a schematic diagram of the centrifugal pump structure used in an embodiment of the present invention.

[0038] Figure 3 This is a modal contour plot of the centrifugal pump used in the embodiments of the present invention at typical characteristic frequencies, wherein, Figure 3 In the diagram, (a) is the mean flow field modal contour plot, and (b) to (e) are f BPF 2f BPF f DPF 2f DPF The corresponding modal contour plot.

[0039] Figure 4 The phase distribution of the centrifugal pump used in the embodiments of the present invention at typical characteristic frequencies is shown below. Figure 4 In the example, (a) to (d) represent f. BPF 2f BPF f DPF 2f DPF The corresponding phase distribution cloud map.

[0040] Figure 5 This refers to the phase change within the impeller-diffuser channel of the centrifugal pump used in this embodiment of the invention at a typical characteristic frequency, wherein... Figure 5 In the example, (a) to (d) represent f. BPF 2f BPF f DPF 2f DPF The corresponding channel phase change curve.

[0041] Figure 6 This is a schematic diagram illustrating the principle of determining the propagation direction of pressure waves based on phase gradient in an embodiment of the present invention. Figure 6 In the example, (a) to (d) represent f. BPF 2f BPF f DPF 2f DPF The corresponding phase change diagram of the bladeless region of the impeller, (e)~(h) are f BPF 2f BPF f DPF 2f DPF A schematic diagram of the phase change in the bladeless region of the corresponding diffuser.

[0042] Figure 7 This is a spatiotemporal evolution diagram of the pressure pulsation excitation source location and multi-frequency propagation path in an embodiment of the present invention. Detailed Implementation

[0043] The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. The purpose and effects of the present invention will become clearer. It should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0044] The core innovation of the phase-driven multi-scale analysis method and system for pressure pulsation in rotating flow fields of this invention lies in its use of phase information as the core driver throughout the entire process. It breaks through the limitations of traditional pressure pulsation analysis, which only focuses on amplitude, frequency, and spatial modes. Through global phase extraction, global multi-scale phase field construction, and quantitative analysis of phase gradient, a complete system for determining pressure wave propagation and locating excitation sources is established, enabling the physical, visual, and quantitative analysis of pressure pulsation in rotating flow fields from excitation and propagation to dissipation.

[0045] like Figure 1 As shown, the multi-scale analysis method for pressure pulsation in a rotating flow field with fusion phase-driven flow according to the present invention includes the following steps one through five.

[0046] Step 1: Obtain unsteady pressure data, sample to obtain a continuous and stable flow field pressure snapshot sequence, and construct a pressure field spatiotemporal snapshot matrix.

[0047] Unsteady pressure time-series data across the entire computational domain of the rotating flow field were acquired through high-precision unsteady numerical simulations or particle image velocimetry (PIV) experiments. The raw pressure data underwent filtering, detrending, and dimensionless preprocessing to remove random noise interference and standardize data dimensions, eliminating the impact of scale differences on subsequent analysis. Then, the preprocessed pressure data was uniformly sampled at equal time intervals to form a continuous and stable flow field pressure snapshot sequence. This sequence was used to construct a spatiotemporal snapshot matrix of the pressure field, providing a standardized, high-quality time-series data source for subsequent mode decomposition and phase extraction.

[0048] Step 2: Perform dynamic mode decomposition on the spatiotemporal snapshot matrix of the pressure field to extract the characteristic frequencies, low-order coherent spatial modes, and corresponding DMD complex modes of the dominant pressure fluctuations. Specifically:

[0049] Dynamic mode decomposition (DMD) was performed on the pressure snapshot sequence. Dimensionality reduction of the flow field was achieved through singular value decomposition and linear mapping approximation, yielding the dynamic modal structure of the rotating flow field, including dynamic modes and their growth / decay rates and natural oscillation frequencies. Based on the modal energy proportions, characteristic frequencies of pressure pulsations dominated by rotor-stator dynamic-static interference were selected. Low-order coherent spatial modes and DMD complex modes corresponding to each dominant characteristic frequency were extracted. The decomposition accuracy was verified by reconstructing residuals, ensuring data validity.

[0050] The m-th order DMD complex mode at any node in space The expression at the location is:

[0051]

[0052] In the formula: Modal amplitude; For modal at node The initial spatial phase at that location.

[0053] This step is used to obtain complex mode information that can support phase analysis, providing basic input for subsequent global phase extraction and gradient calculation.

[0054] Step 3: Construct the flow signal matrix of grid nodes with the same frequency, extract the initial spatial phase of the entire flow field directly through argument operation based on DMD complex modes, calculate the phase spatial gradient, generate a global phase field with a single characteristic frequency, and combine the global phase fields of all dominant frequencies to obtain the global multi-scale phase field.

[0055] This step is the core innovation of the present invention. It deeply mines and completely extracts the phase information of the entire flow field from the modal time-series signal, and expands the traditional static spatial structure of DMD into a dynamic physical field containing time-series phase characteristics, providing a basis for determining the propagation direction of pressure waves, locating excitation sources, and analyzing energy evolution.

[0056] Step 3 includes the following sub-steps:

[0057] S3.1: For the complex mode of DMD corresponding to each dominant characteristic frequency, the initial spatial phase of each grid point in the entire flow field is directly extracted through complex argument calculation:

[0058]

[0059] In the formula, For the imaginary part of a complex number, It represents the real part of the complex number.

[0060] This step directly obtains the temporal phase distribution of pressure pulsations in space from the complex modes of the DMD.

[0061] S3.2: Calculate the spatial gradient of the phase field to obtain the vector field characterizing the direction of coherent structure evolution:

[0062]

[0063] The direction of the phase gradient vector represents the direction of the fastest phase change, i.e., the main direction of pressure wave propagation and evolution; the magnitude of the phase gradient represents the intensity of the phase change, directly reflecting the energy strength and attenuation rate of the pressure wave. The phase gradient can transform the abstract phase distribution into a quantitatively interpretable vector characteristic, providing a core basis for determining the propagation direction and locating the excitation source.

[0064] S3.3: Assign the phase angles of all extracted grid points to the spatial grid nodes of the DMD mode one by one to generate a global phase field covering the entire computational domain at a single characteristic frequency.

[0065] S3.4: Repeat the phase extraction, phase difference calculation and phase assignment operations of S3.1~S3.3 for all dominant characteristic frequencies to construct a global multi-scale phase field that can reflect the phase information of perturbation waves at different scales.

[0066] The global multi-scale phase field fully preserves the temporal phase characteristics of pressure pulsations at all frequencies in the entire flow field, and is the core carrier for realizing pressure wave propagation determination and excitation source localization.

[0067] Step 4: Define the analysis path in the key region of the rotating flow field, extract the phase distribution on each analysis path and calculate the one-dimensional phase gradient of the path direction, and determine the propagation direction of the pressure wave based on the phase gradient criterion; traverse the global multi-scale phase field and gradient field, identify the phase change feature points, and spatially superimpose the feature points of all dominant frequencies to achieve pressure pulsation source tracing; at the same time, divide the pressure wave energy region by combining the phase gradient distribution characteristics.

[0068] Key regions of the rotating flow field include the rotor-stator interface, the bladeless region, and the flow channel centerline; targeted analysis paths include circumferential paths and axial flow channel paths penetrating the rotor and stator; the spatial phase gradient of the global phase field is calculated for each characteristic frequency, and a unified pressure wave propagation criterion is established based on the positive and negative characteristics of the path phase gradient: Let the path coordinates be... The one-dimensional phase gradient along the path is: .

[0069] when At this time, the phase increases along the path direction, and the propagation direction of the pressure wave is opposite to the direction of the analysis path;

[0070] when At this time, the phase decreases along the path direction, and the pressure wave propagation direction is the same as the analysis path direction;

[0071] Phase gradient reversal points, phase abrupt change points, and local phase extrema points are the regions where the propagation direction of pressure waves changes, energy concentrates, and interference is strongest. Spatially superimposing these characteristic points of all dominant frequencies, the region where they converge is the core excitation source of pressure pulsations in the rotating flow field. Simultaneously, the pressure wave energy regions are divided based on the distribution characteristics of the phase gradient: regions with small and uniformly distributed phase gradients are energy accumulation regions, while regions with large and frequently abrupt phase gradients are energy attenuation regions.

[0072] Step 5: The spatial modes (energy distribution) of each characteristic frequency, the global multi-scale phase field (phase temporal characteristics), the propagation direction of the pressure wave, and the location of the core excitation source are superimposed and fused on a unified rotating flow field geometric model. The modal and phase field data at different times are combined according to the time sequence to synthesize a holographic map of the spatiotemporal evolution of pressure pulsation in the rotating flow field. This visually restores the entire process of the pressure wave from the generation of the core excitation source, its propagation along a specific path, to the interference, reflection, superposition, and dissipation inside the flow channel, thus realizing the visualization of the evolution law of pressure pulsation.

[0073] This step transforms the abstract and complex mechanism of pressure pulsation propagation into observable, quantifiable, and interpretable visualization results, providing intuitive and reliable theoretical basis and technical support for flow field optimization, vibration reduction and noise reduction design, and structural improvement of rotating fluid machinery such as centrifugal pumps, water pump turbines, and turbines.

[0074] Corresponding to the aforementioned embodiments of the multi-scale analysis method for pressure fluctuations in a rotating flow field driven by fusion phase, the present invention also provides embodiments of a multi-scale analysis system for pressure fluctuations in a rotating flow field driven by fusion phase.

[0075] The fusion phase-driven rotating flow field pressure pulsation multi-scale analysis system of this embodiment includes one or more processors for implementing the fusion phase-driven rotating flow field pressure pulsation multi-scale analysis method in the above embodiment.

[0076] The embodiments of the phase-driven rotating flow field pressure pulsation multi-scale analysis system of the present invention can be applied to any device with data processing capabilities, such as a computer. The device embodiments can be implemented in software, hardware, or a combination of both. Taking software implementation as an example, as a logical device, it is formed by the processor of the device with data processing capabilities loading the corresponding computer program instructions from non-volatile memory into memory for execution. From a hardware perspective, in addition to the processor, memory, network interface, and non-volatile memory, the device with data processing capabilities in the embodiments may also include other hardware depending on its actual functions, which will not be elaborated further.

[0077] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0078] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0079] The effectiveness of the method of the present invention will be further demonstrated through a specific application example below.

[0080] This embodiment uses the rotating flow field of a centrifugal pump with radial guide vanes as a typical analysis object, such as... Figure 2 As shown, the pressure time series data of the entire flow field in the impeller and guide vane regions were obtained through unsteady numerical simulation, and the method of this invention was used to conduct a full-process analysis.

[0081] Figure 3 This is a modal contour plot at typical characteristic frequencies, where f BPF f is the frequency at which the impeller blades pass through, representing the excitation frequency generated by the impeller rotation on stationary components; DPF The frequency at which the diffuser blades pass through represents the frequency of the feedback excitation generated by the stationary component to the rotating flow field. Figure 3 It can be seen that f BPF The corresponding modal energies are mainly distributed within the diffuser channel, reflecting the periodic excitation effect of the impeller rotation on the diffuser; f DPF The corresponding modal energies are mainly distributed within the impeller flow channel, reflecting the periodic constraint effect of the diffuser on the internal flow of the impeller. The spatial distribution characteristics of different frequency modes are highly consistent with the dynamic-static interference mechanism.

[0082] Figure 4 This represents the phase distribution of a centrifugal pump at a typical characteristic frequency. Figure 4 It can be seen that f BPF The diffuser channel exhibits a continuous and regular phase progression characteristic, f DPF The impeller channel exhibits periodic phase jump characteristics, and the phase advancement, phase delay, and phase interference patterns of pressure pulsations at different frequencies in the impeller, bladeless region, and diffuser channel can be intuitively presented, realizing a visual expression of the spatiotemporal evolution of pressure pulsations.

[0083] Figure 5 This is the phase change curve within the impeller-diffuser channel at a typical characteristic frequency. Figure 5It can be seen that there is a phase maxima in the bladeless region. The turning point of the phase change trend in the bladeless region is the key area where the direction of pressure wave propagation changes, further confirming that the bladeless region is the core area for the generation and propagation of pressure pulsations.

[0084] Figure 6 This is a schematic diagram illustrating the principle of determining the propagation direction of pressure waves based on phase gradient. Figure 6 It can be seen that the positive or negative phase gradient along the analysis path can directly determine the propagation direction of the pressure wave. Using this criterion, the propagation direction of pressure waves of different frequencies can be accurately identified, which is the core basis of the phase-driven analysis system of this invention.

[0085] Figure 7 This provides a spatiotemporal evolution map of the pressure pulsation excitation source and its multi-frequency propagation path. Figure 7 It can be seen that the excitation location of pressure waves of all frequencies is concentrated in the bladeless region between the impeller outlet and the diffuser inlet, f BPF The pressure wave propagates downstream from the bladeless region to the guide vane channel, f DPF Pressure waves propagate from the bladeless region to the upstream of the impeller channel. The two types of pressure waves propagate in opposite directions and interfere strongly within the bladeless region, fully presenting the spatiotemporal evolution law of pressure pulsation from excitation, propagation, interference to dissipation.

[0086] By spatially superimposing multi-frequency phase feature points, the common excitation source of all pressure pulsation components is accurately located in the bladeless region between the impeller outlet and the guide vane inlet. This result is completely consistent with the physical mechanism of dynamic-static interference in fluid mechanics and numerical observation results. At the same time, the energy accumulation region and attenuation region are clearly delineated by combining the phase gradient distribution, and the entire process of pressure pulsation excitation-propagation-dissipation is completely restored.

[0087] The results of this embodiment show that the phase-based analysis system of the present invention can significantly improve the physical consistency and positioning accuracy of pressure pulsation analysis.

[0088] It will be understood by those skilled in the art that the above descriptions are merely preferred examples of the invention and are not intended to limit the invention. Although the invention has been described in detail with reference to the foregoing examples, those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. All modifications and equivalent substitutions made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. A multi-scale analysis method for pressure fluctuations in a rotating flow field using phase-driven fusion, characterized in that, include: S1: Acquire unsteady pressure data, sample to obtain a continuous and stable flow field pressure snapshot sequence, and construct a pressure field spatiotemporal snapshot matrix; S2: Perform dynamic mode decomposition on the spatiotemporal snapshot matrix of the pressure field to extract the characteristic frequencies, low-order coherent spatial modes and corresponding DMD complex modes of the dominant pressure fluctuations; S3: Construct the flow signal matrix of grid nodes with the same frequency, extract the initial spatial phase of the entire flow field directly through argument operation based on DMD complex modes, calculate the phase spatial gradient, generate a global phase field with a single characteristic frequency, and combine the global phase fields of all dominant frequencies to obtain the global multi-scale phase field. S4: Define the analysis path in the key region of the rotating flow field, extract the phase distribution on each analysis path and calculate the one-dimensional phase gradient of the path direction, and determine the propagation direction of the pressure wave based on the phase gradient criterion. By traversing the global multi-scale phase field and gradient field, identifying phase change feature points, and spatially superimposing all dominant frequency feature points, pressure pulsation source tracing is achieved; at the same time, pressure wave energy regions are divided by combining phase gradient distribution characteristics. S5: The spatial modes of each characteristic frequency, the global multi-scale phase field, the propagation direction of the pressure wave, and the location of the core excitation source are superimposed and fused on a unified rotating flow field geometric model. The mode and phase field data at different times are combined according to the time series to synthesize a holographic map of the spatiotemporal evolution of pressure pulsation in the rotating flow field.

2. The multi-scale analysis method for pressure fluctuations in a rotating flow field with fused phase-driven propagation as described in claim 1, characterized in that, S3 includes the following sub-steps: S3.1: For each dominant characteristic frequency corresponding to the DMD complex mode, the initial spatial phase of each grid point in the entire flow field is directly extracted by complex argument calculation; S3.2: Calculate the spatial gradient of the phase field to obtain the vector field characterizing the evolution direction of the coherent structure; S3.3: Assign the phase angles of all extracted grid points to the spatial grid nodes of the DMD mode one by one to generate a global phase field covering the entire computational domain at a single characteristic frequency; S3.4: Repeat the phase extraction, phase difference calculation and phase assignment operations of S3.1~S3.3 for all dominant characteristic frequencies to construct a global multi-scale phase field that can reflect the phase information of perturbation waves at different scales.

3. The multi-scale analysis method for pressure fluctuations in a rotating flow field with fused phase-driven flow according to claim 2, characterized in that, In step S4, the direction of pressure wave propagation is determined based on the positive and negative characteristics of the one-dimensional phase gradient along the path. When the one-dimensional phase gradient along the path direction is greater than 0, the phase increases along the path direction, and the pressure wave propagation direction is opposite to the analysis path direction. When the one-dimensional phase gradient along the path direction is less than 0, the phase increases along the path direction, and the pressure wave propagation direction is the same as the analysis path direction.

4. The multi-scale analysis method for pressure fluctuations in a rotating flow field with fused phase-driven propagation as described in claim 1, characterized in that, In S4, the characteristic points of phase change include phase gradient reversal points, phase abrupt change points, and phase local extrema points.

5. The multi-scale analysis method for pressure fluctuations in a rotating flow field with fused phase-driven propagation as described in claim 1, characterized in that, The key regions of the rotating flow field include the rotor-stator interface, the bladeless region, and the flow channel centerline.

6. The multi-scale analysis method for pressure fluctuations in a rotating flow field with fused phase-driven propagation as described in claim 1, characterized in that, The analysis path includes a circumferential path and an axial flow path that runs through the rotor and stator.

7. The multi-scale analysis method for pressure fluctuations in a rotating flow field with fused phase-driven flow according to claim 1, characterized in that, The pressure wave energy region includes an energy accumulation region and an energy attenuation region.

8. The multi-scale analysis method for pressure fluctuations in a rotating flow field with fused phase-driven propagation as described in claim 1, characterized in that, The unsteady pressure data were obtained through unsteady numerical simulation or particle image velocimetry experiments.

9. A multi-scale analysis system for pressure fluctuations in a rotating flow field using phase-driven methods, characterized in that, It includes one or more processors for implementing the multi-scale analysis method for pressure pulsation in a rotating flow field with fusion phase drive as described in any one of claims 1 to 8.

10. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by the electronic device, cause the electronic device to implement the multi-scale analysis method for pressure pulsation in a rotating flow field with fused phase drive as described in any one of claims 1 to 8.

Images