Axial reconfigurable non-contact flow field measurement method suitable for high-speed motor

Abstract

The invention discloses an axial reconfigurable non-contact flow field measurement method suitable for a high-speed motor, and particularly relates to the field of fluid mechanics measurement. A measuring device is adopted and comprises a laser and an imaging mechanism, the laser is arranged on the outer side of a stator of a motor through a moving mechanism and can move in the axial direction of the stator of the motor, and the imaging mechanism is used for collecting images of a plane illuminated by the laser; the method comprises the following steps: injecting tracer particles into an air gap area between a stator and a rotor of the motor; axially moving the laser through a moving mechanism, sequentially obtaining particle images of a plurality of radial planes of the air gap region, and obtaining a velocity vector of the corresponding radial plane according to each particle image; converting the velocity vector of each plane from a Cartesian coordinate system to a polar coordinate system; and splicing the converted velocity vectors of all the planes along the axial direction to obtain an average circumferential velocity field of the axial and radial planes. Based on the method, the flow field measurement of the high-speed motor can be realized.

Description

Technical Field

[0001] This application relates to the field of fluid dynamics measurement, and in particular to an axially reconfigurable non-contact flow field measurement method suitable for high-speed motors. Background Technology

[0002] During the operation of high-speed motors, the eddy current structure within the narrow air gap between the stator and rotor significantly impacts the motor's heat dissipation performance and operational stability. Traditional particle image velocimetry (PIV) systems primarily consist of tracer particles, a laser illumination system, an image acquisition system, and a data processing system. Although these methods are widely used in laminar and open flow channels, traditional PIV systems face the following significant limitations within the high-speed, small-gap, high-temperature, and enclosed environment of motors: 1. Severe optical interference: Reflected light from the rotor wall covers the near-wall region, making it impossible to resolve the boundary layer velocity field. Strong surface reflections from the high-speed rotor rotation reduce imaging contrast, particularly obscuring boundary layer particle images. 2. Limited measurement space: The light distribution and imaging window are extremely narrow under small air gaps (2~3 mm), making it difficult to achieve effective laser illumination and imaging system setup.

[0003] Therefore, there is an urgent need for a flow field measurement method suitable for high-speed, small-gap motor structures to solve the technical challenges of particle stabilization control and optical interference suppression. Summary of the Invention

[0004] The main objective of this application is to provide an axially reconfigurable non-contact flow field measurement method suitable for high-speed motors, aiming to solve the problem that the boundary layer velocity field cannot be analyzed in existing measurement methods.

[0005] To achieve the above objectives, this application provides an axially reconfigurable non-contact flow field measurement method suitable for high-speed motors. The method employs a measuring device including a laser and an imaging mechanism. The laser is positioned outside the stator of the motor via a moving mechanism and can move along the axial direction of the stator. The imaging mechanism is used to acquire images of the plane illuminated by the laser. The method includes: injecting tracer particles into the air gap region between the stator and rotor of the motor; moving the laser axially via the moving mechanism to sequentially acquire particle images of multiple radial planes in the air gap region; obtaining the velocity vector of the corresponding radial plane based on each particle image; converting the velocity vector of each plane from a Cartesian coordinate system to a polar coordinate system; and stitching the velocity vectors of all the converted planes along the axial direction to obtain the axial-radial plane average circumferential velocity field.

[0006] Optionally, the tracer particles are injected along the inlet axis into the air gap region between the stator and rotor of the motor.

[0007] Optionally, the velocity vector of each radial plane is measured using the PIV method.

[0008] Optionally, after obtaining the velocity vector of the corresponding radial plane, the method further includes: determining the time average field of each particle image to obtain grayscale distribution features; determining the light intensity threshold based on the grayscale distribution features; identifying the rotor wall in the particle image based on the light intensity threshold; extracting three points on the rotor wall to determine the position of the rotor center; determining the true physical distance of each pixel in the particle image based on the rotor radius and the number of pixels from the rotor wall to the rotor center; and multiplying the velocity vector of each radial plane by the true physical distance of the pixel to obtain the velocity vector of each radial plane in the Cartesian coordinate system.

[0009] Optionally, the velocity vector of each plane is converted from the Cartesian coordinate system to the polar coordinate system, including: using a bilinear interpolation method to convert the velocity vector of each plane from the Cartesian coordinate system to the polar coordinate system.

[0010] Compared with the prior art, the beneficial effects of this application are as follows: This invention provides an axially reconfigurable non-contact flow field measurement method for high-speed motors. A laser is mounted on a horizontal displacement stage, allowing precise horizontal repositioning of the laser plate. The laser plate illuminates the tangential radial plane of the air gap, and axial movement of the laser illuminates different positions. Simultaneously, an imaging system is triggered to record particle image sequences from each measurement surface, achieving effective laser illumination and imaging system setup. Multiple particle image sequences from measurement surfaces are acquired, capturing instantaneous velocity vectors. Through stitching and reconstruction, the flow structure within the axial and radial planes of the gap is obtained, achieving accurate reconstruction of complex flow structures and avoiding the problem of boundary layer particle images being easily submerged, leading to unresolved velocity fields. Attached Figure Description

[0011] Figure 1 This is a diagram showing the optical distortion caused by the roughness of the existing TC outer cylinder; Figure 2 This is a schematic diagram of the process of the axially reconfigurable non-contact flow field measurement method applicable to high-speed motors in this application; Figure 3 This is a schematic flowchart of the axially reconfigurable non-contact flow field measurement method applicable to high-speed motors in this application; Figure 4 This is a wall view of the rotor in the particle image of the axially reconfigurable non-contact flow field measurement method for high-speed motors applicable to this application; Figure 5 This is a schematic diagram of the bilinear interpolation process in the axially reconfigurable non-contact flow field measurement method for high-speed motors applicable to this application; Figure 6The diagram shows the distribution of the circumferential radial (r-θ) plane-averaged circumferential velocity field and the axial radial (zr) plane-averaged circumferential velocity field in the embodiment. Figure 7 This is a diagram showing the average circumferential velocity distribution in an embodiment. Figure 8 The shear Reynolds stress diagram is shown in the example.

[0012] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] It is worth noting that studying the axial and radial planar velocity distribution in a TC flow system involves directly photographing the axial and radial planes of the gap using planar PIV technology. However, rough elements on the stator of a high-speed motor can cause optical distortion, which can interfere with the camera's imaging. Figure 1 As shown. Therefore, the classic experimental scheme is not applicable. Therefore, this invention obtains the axial-radial plane flow structure within the gap by reconstructing the circumferential radial plane flow field, as shown. Figure 2 As shown. The circumferential radial direction is measured using a planar PIV (pivot-permeability) method. Velocity on the plane, It is the circumferential velocity (tangential to the circumference). The radial velocity (pointing towards / away from the center of rotation). To observe the Taylor vortex structure in the axial-radial plane, planar PIV measurements are performed at multiple different heights. The laser is mounted on a horizontal displacement stage, allowing for precise changes in the position of the laser sheet along the horizontal direction. A single axial movement of the laser ensures that at least one pair of Taylor vortices is included across the axial span, as detailed below.

[0015] This invention provides an axially reconfigurable non-contact flow field measurement method suitable for high-speed motors, such as... Figure 3 As shown, the specific steps include: Step S1: Inject tracer particles into the air gap region between the stator and rotor of the motor; Specifically, the tracer particles are injected axially into the air gap region between the stator and rotor of the motor, along the air inlet of the fan. For example, axial injection can be used, with the particles evenly distributed into the air gap region between the stator and rotor via a dedicated air duct connected to the inlet. By adjusting the wind speed to match the centrifugal escape critical condition, the particles can remain stably suspended even under high-speed rotation. For example, the wind speed range can be 0-10 m / s; for instance, for a maximum achievable linear velocity of 135 m / s, the axial wind speed is 4 m / s.

[0016] Step S2: Move the laser axially to sequentially acquire particle images of multiple radial planes in the air gap region, and obtain the velocity vector of the corresponding radial plane based on each particle image; Specifically, the velocity vector of each radial plane is measured using the PIV method. By moving the laser axially, the laser sheet illuminates the tangential radial plane of the air gap, forming a series of r-θ plane measurement windows; moving the laser to multiple positions along the axis, the imaging system is simultaneously triggered to record the particle image sequence of each measurement plane, capturing the instantaneous velocity vector.

[0017] Furthermore, since PIV measurements are performed at different positions along the axial direction, there is a "nearer field of view, farther field of view" issue relative to the camera; that is, the field of view is larger when the measurement position is closer to the camera and smaller when the position is farther away. Therefore, corresponding masks and velocity calibrations are required for experiments at different positions to obtain the true physical size and coordinates of the velocity vectors measured at different positions. This embodiment processes the velocity vector of each radial plane using the following method.

[0018] First, determine the time-averaged field of each particle image. This yields grayscale features; Secondly, the light intensity threshold is determined based on grayscale features, and the rotor wall in the particle image is identified based on the light intensity threshold. Figure 4 The white arc in the upper right corner of the image represents the rotor wall. Three points are extracted from the rotor wall to determine the position of the rotor center. For example, first, set the threshold to 1500 (test estimation; after calculating the average field, the light intensity threshold of the rotor wall is the largest). Display the 1500 strip on the image, which is the rotor wall. Take the three brightest points in it and fit a circle to obtain the position of the rotor center.

[0019] Next, based on the rotor's radius and the number of pixels from the rotor wall to the rotor's center, the true physical distance of each pixel in the particle image is determined; specifically, due to the rotor's radius... It remains constant, therefore the number of pixels from the rotor wall to the rotor center measured at different locations is constant. They all correspond to the same physical distance. Using this relationship, the true physical distance for each pixel can be obtained. : (1); in, The unit of measurement is "meters per pixel," representing the actual physical distance corresponding to each pixel on the measurement plane. The pixels in images at different locations correspond to different actual physical distances. They are all different. In this embodiment, the PIV (Dynamic Studio 7.6) software is used for multiple iterations to calculate the velocity vector of each radial plane. The query domain sizes for the first and second iterations are respectively... and overlap is Therefore, a velocity vector occupies The pixel area, so the real physical distance between each velocity vector in the Cartesian frame is .

[0020] Finally, the velocity vector of the radial plane is multiplied by the actual physical distance of the pixel, i.e., pixel displacement / imaging time. The image-level velocity is converted into the physical velocity of the actual flow field (such as m / s, mm / s), and the velocity vector of each radial plane in the Cartesian coordinate system is obtained.

[0021] In this embodiment, oil fume particles are blown into the gap as tracer particles. Due to the high rotor speed, oil fume residue will remain on the rotor and stator, causing stray light. Therefore, after each set of images, a 2mm diameter test tube brush wrapped in oil-absorbing paper is needed to clean and wipe the air gap. For particles that cannot be cleaned and have a diameter less than 5mm... The interference of stray light is eliminated by subtracting the time-averaged field of the original image when calculating the velocity vector in subsequent calculations.

[0022] Step S3: Convert the velocity vector of each plane from the Cartesian coordinate system to the polar coordinate system; Specifically, bilinear interpolation is used to transform the velocity vector of each plane from the Cartesian coordinate system to the polar coordinate system. The specific process is as follows.

[0023] Assuming the point is known and Target point The interpolation formula is: The algorithm has a time complexity of O(n). While computationally efficient, linear interpolation is only applicable to one-dimensional data and is sensitive to nonlinear variations. Bilinear interpolation is an extension of linear interpolation in two-dimensional space, used to estimate the function value at a point in a rectangular grid. For example... Figure 5 Along the x-axis for two pairs of points , , , Perform linear interpolation separately to obtain the midpoint. and Along the y-axis and Interpolation to obtain the target point The estimated value.

[0024] Specifically, the velocity components in the Cartesian coordinate system ( , Transform to polar coordinates ( , To achieve this goal, the following steps must be performed in sequence: Establishing coordinate transformation relationships Based on the geometric relationship between polar coordinates and Cartesian coordinates: (2-1); The formula for velocity components in polar coordinates is: (2-2); Interpolation grid definition Let the points to be interpolated in the Cartesian coordinate system be... The coordinates of its four adjacent grid nodes are: (2-3); Where h is the grid spacing. Define normalized coordinates: (2-4); Bilinear interpolation formula

[0025] Through formula Calculate separately and Substitute the interpolation result into This will give you the polar coordinate system. , .

[0026] Step S4: The velocity vectors of all the transformed planes are spliced ​​along the axial direction to obtain the axial and radial plane average circumferential velocity field.

[0027] The instantaneous velocity field can be obtained by tracking the displacement of tracer particles using the methods described above. However, the instantaneous velocity field only reflects the flow state at a certain moment, while actual flows, especially turbulent flows, often exhibit high levels of unsteadiness and randomness. Therefore, it is usually necessary to extract the average velocity field by performing statistical analysis on multiple instantaneous velocity fields.

[0028] According to Reynolds' time-averaged theory, instantaneous velocity It can be decomposed into average velocity and pulsation velocity linear superposition, Representing position and time variables respectively, as in the formula As shown. Average velocity field The time-averaged steady-state component characterizing the flow reflects the persistent macroscopic structures in the flow field, such as the velocity gradient along the mainstream direction, the backflow structure in the separation region, and the location of the vortex core. By filtering out high-frequency random fluctuations, the average velocity field can clearly show the global characteristics of the flow, providing a verification benchmark for theoretical models and numerical simulations. Fluctuating velocity field Characterized by instantaneous random disturbances caused by turbulence, with a mean of zero ( However, second-order statistics (such as variance and covariance) contain key information about turbulent energy transport and dissipation.

[0029] (3-1); The average velocity field is calculated using either time averaging or ensemble averaging, depending on the method of selecting the statistical sample. Time averaging is suitable for steady-state flows (where the statistical characteristics of the flow do not change over time), and is obtained by averaging the instantaneous velocities at the same spatial location over a time series, as shown in the equation. As shown.

[0030] (3-2); In the formula, u i Indicates speed, i Indicates direction; x Representing space, t n Time represents time and together they describe spatiotemporal information; ensemble averaging is applicable to unsteady flows and requires averaging the instantaneous velocity fields of multiple independent experiments under the same phase or conditions. If the flow is steady-state, time averaging is equivalent to ensemble averaging (based on the ergodic assumption).

[0031] Example Through steps S1-3, the circumferential velocity field of each group is obtained via PIV. By performing time averaging, the average circumferential velocity in the circumferential radial (r-θ) plane at 18 axial positions is obtained. . Figure 6 (a) and (b) respectively show the motor speed as and Average circumferential velocity field at the same axial position That is, the average circumferential velocity field at a location, determined by the linear velocity of the inner cylinder of the rotor. Normalization (where The rotor angular velocity, (where is the rotor radius). Figure 6 (c) and (d) respectively demonstrate and Average circumferential velocity at the next 18 positions Axial radial plane (zr) mean circumferential velocity field obtained by splicing along the axial direction That is, the average circumferential velocity field of the reconstructed flow field. As can be seen from the figure, in Down, The length of the velocity vector arrow gradually decreases radially from the inner cylinder to the outer cylinder, indicating the absence of a clear flow structure, a result consistent with previous similar work; when Increase to , The velocity vector arrow length exhibits a significant gradient near the outer and inner cylinders, while remaining almost constant in the intermediate layer, indicating the absence of any obvious large-scale flow structures. This implies that the velocity vector varies with the inner cylinder angular velocity. The boundary layer between the outer and inner cylinders is gradually thinning, and the circumferential velocity of the intermediate layer is... There is a certain degree of homogeneity in the space (r-θ-z). These results show that... This increases the tendency of flow evolution. Therefore, it can be shown that the measurement method of this application can accurately measure the flow field velocity of a high-speed motor.

[0032] The circumferential velocity field at the 18 axial positions after the above time averaging Time averaging was performed, and spatial averaging was applied in the circumferential and axial directions to obtain the normalized average circumferential velocity. The results are shown Figure 7 The figure shows the normalized average circumferential velocity. The velocity values ​​distributed radially are all expressed as the linear velocity of the inner cylinder wall surface. Normalization for reference. The radius ratio is 0.98 and average circumferential velocity Passing through the side near the inner cylinder ; The radius ratio is 0.98 and Below, normalized average circumferential velocity Along the middle layer The decrease was gradual; existing technology measured the radius ratio as 0.909 using LDA technology. average circumferential velocity It can be seen that the mean velocity in the intermediate layer decreases slowly from 0.48 to 0.44 radially, which is more gradual than all our results, and also more... The speed is even lower. This seems to suggest that for Taylor-Couette flow where only the inner cylinder rotates, the average circumferential velocity of the intermediate layer is lower. The radial velocity gradient increases as the radius ratio increases. It increases and then becomes smaller. This phenomenon has been observed in previous similar studies, see... Figure 7 [Monico, 2014]. Mean circumferential velocity. The velocity gradient near the inner and outer cylinders follows The increase in size means that the boundary layer between the inner and outer cylinders becomes thinner.

[0033] This embodiment calculates the radial Reynolds shear stress based on the measured velocity of the reconstructed flow field. The linear velocity of the inner cylinder Normalization, results are shown in Figure 8 , Reynolds shear stress near the inner cylinder It is higher than that at the outer cylinder; this is a residual effect of the Gortler vortex, because... hour Almost symmetrical. (This is related to) the linear velocity of the inner cylinder. Increase Reynolds shear stress The suppression indicates that the measurement method of this embodiment can realize the flow field measurement of high-speed motors and has reliability. The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for axially reconfigurable non-contact flow field measurement suitable for high-speed motors, characterized in that, The method employs a measuring device, including a laser and an imaging mechanism. The laser is mounted outside the stator of the motor via a moving mechanism and can move along the axial direction of the stator. The imaging mechanism is used to acquire images of the plane illuminated by the laser. Tracer particles are injected into the air gap region between the stator and rotor of the motor; The laser is moved axially by a moving mechanism to sequentially acquire particle images of multiple radial planes in the air gap region. Based on each particle image, the velocity vector of the corresponding radial plane is obtained. Transform the velocity vector of each plane from the Cartesian coordinate system to the polar coordinate system; By splicing the velocity vectors of all the transformed planes along the axial direction, we obtain the axial-radial plane average circumferential velocity field.

2. The axially reconfigurable non-contact flow field measurement method for high-speed motors according to claim 1, characterized in that, The tracer particles are injected along the inlet axis into the air gap region between the stator and rotor of the motor.

3. The axially reconfigurable non-contact flow field measurement method for high-speed motors according to claim 1, characterized in that, The velocity vector of each radial plane is obtained by measuring using the PIV method.

4. The axially reconfigurable non-contact flow field measurement method for high-speed motors according to claim 1, characterized in that, After obtaining the velocity vector corresponding to the radial plane, the method further includes: Determine the temporal mean field of each particle image to obtain grayscale distribution characteristics; The light intensity threshold is determined based on the grayscale distribution characteristics. The rotor wall in the particle image is identified based on the light intensity threshold. Three points are extracted from the rotor wall to determine the position of the rotor center. The true physical distance of each pixel in the particle image is determined based on the rotor radius and the number of pixels from the rotor wall to the rotor center. Multiplying the velocity vector of each radial plane by the actual physical distance of the pixel yields the velocity vector of each radial plane in Cartesian coordinates.

5. The axially reconfigurable non-contact flow field measurement method for high-speed motors according to claim 1, characterized in that, The process of converting the velocity vector of each plane from the Cartesian coordinate system to the polar coordinate system includes: The velocity vector of each plane is converted from Cartesian coordinates to polar coordinates using bilinear interpolation.

Images