A fan sound source positioning analysis device and method

Abstract

The application relates to a fan sound source positioning analysis device and method. The technical problem to be solved by the application is to provide a fan sound source positioning analysis device and method. The technical scheme of the fan sound source positioning analysis device is as follows: a fan model installed on a wind tunnel rack, a far-field microphone device, a near-field microphone device, a particle image velocimeter, a wall surface pressure sensor device and a computer are arranged, the fan model comprises a stator support, a driving motor and a rotatable rotor impeller arranged on the stator support, and the far-field microphone device comprises a data acquisition card and a plurality of far-field microphones arranged at intervals along a circular arc and taking the fan model as the center. The fan sound source positioning analysis device and method have the following advantages: 1. wide applicability: four sub-test systems can be adjusted independently, and the noise forming reason can be accurately analyzed; 2. convenient maintenance and problem searching; and 3. high visualization degree.

Description

Technical Field

[0001] This invention relates to the field of axial flow fan technology, specifically to a fan sound source localization analysis device and method. Background Technology

[0002] Axial flow fans have wide applications in aerospace, transportation, and household appliances. While they solve ventilation and heat dissipation problems, they also bring significant noise issues. With increasing demands for comfortable working and living environments, identifying axial flow fan noise and implementing targeted noise reduction designs is essential. Traditional methods for detecting fan model noise and performance mainly involve simplified fan models and numerical simulation. The former focuses on interference noise between isolated airfoils or structures because these structures are relatively simple and easy to analyze, but it cannot accurately explore the specific noise reduction mechanism of new airfoils or structures applied to axial flow fans, nor can it accurately evaluate the noise reduction effect. The latter can obtain accurate results, but it consumes significant computational resources, takes a long time, and the authenticity and accuracy of the reflected flow and sound fields are difficult to verify. Summary of the Invention

[0003] To address the above problems, the technical problem to be solved by the present invention is to provide a device and method for locating and analyzing the sound source of a wind turbine.

[0004] The technical solution adopted by the wind turbine noise source localization and analysis device of the present invention is characterized by including a wind turbine model installed on a wind tunnel rig, a far-field microphone device, a near-field microphone device, a particle image velocimeter, a wall pressure sensor device, and a computer. The wind turbine model includes a stator support, a drive motor, and a rotor impeller rotatably mounted on the stator support. The far-field microphone device includes a data acquisition card and multiple far-field microphones arranged at arc intervals around the wind turbine model. The far-field microphones are connected to the data acquisition card, and the data acquisition card is connected to the computer via USB. The far-field microphones are used to collect noise signals from the wind turbine noise source side.

[0005] The near-field microphone device includes a group of micro-microphones and a camera arranged horizontally or vertically to the wind turbine model. The near-field microphone array is used to collect noise on a specified plane and visualize it. The particle image velocimeter includes a dual-pulse laser generator placed on the side of the wind turbine model, a CCD high-speed camera placed above the wind turbine model, a synchronization controller for controlling the dual-pulse laser generator and the CCD high-speed camera, and a PIV host. The particle image velocimeter is used to obtain the stall vortex and energy distribution in the flow field.

[0006] The wall pressure sensor device includes a wall pressure sensor fixed on the stator support and a pressure measuring microphone connected to the wall pressure sensor. The wall pressure sensor device is used to collect vibration signals on the stator blades of the wind turbine.

[0007] The computer is connected to the near-field microphone device, the drive motor, the PIV host, and the pressure measuring microphone, respectively.

[0008] The two far-field microphones are spaced 15° apart, and the micro-microphone group includes 112 micro-microphones, which are arranged in a circular structure with the camera as the center.

[0009] The technical solution adopted in the wind turbine sound source localization and analysis method of the present invention is characterized by the following specific steps:

[0010] The first step is to measure the noise frequency signals at different locations using the far-field microphone device and determine the frequency that needs to be analyzed.

[0011] The second step is to obtain the main location of the noise at the corresponding frequency using the near-field microphone device.

[0012] The third step involves using the particle image velocimeter to capture images of the main locations of the noise at that frequency to obtain the flow field velocity, vortex structure, and pressure signals.

[0013] The fourth step involves obtaining the frequency domain signal of the pressure through the wall pressure sensor device and combining it with the data from the third step to analyze the cause of noise formation.

[0014] In the first step, acoustic data is acquired using the far-field microphone device, the PSD is calculated using the Welch method, and the statistical confidence level is obtained using the Hanning window, based on the calculation formulas for linear sound pressure level and total sound pressure level.

[0015] The spectrum diagrams and total sound pressure levels for different frequency bands at the monitoring points are as follows:

[0016]

[0017] Where SPLi represents the sound pressure level at different frequencies, and p represents the sound pressure value at different frequencies. ref For reference sound pressure level;

[0018] The pwelch function in MATLAB software is used to process the acquired wall pulsating pressure time-domain signal and convert it into a frequency-domain signal.

[0019] The third step involves processing multiple images captured by the particle image velocimeter. The specific steps are as follows:

[0020] (1) Determine the image magnification of the analysis section based on the captured scale image in MicroVec software;

[0021] (2) Divide the analysis section into smaller interrogation domains (32×32) and use the intercorrelation algorithm to analyze the particle images in each interrogation domain;

[0022] (3) Perform batch calculations on all particle image pairs with good results to obtain the average velocity vector field of the measured cross section.

[0023] (4) Import the results of the batch calculation into Tecplot software for further analysis to obtain additional information about the flow field of the photographed section.

[0024] The advantages of the wind turbine sound source localization analysis device and method of the present invention are as follows: 1. Wide applicability: It can independently adjust the frequency range of the four sub-test systems: near-field microphone device, far-field microphone device test system, particle image velocimeter, and wall pressure sensor device. Moreover, the test object is not limited to axial flow fans, which is very much in line with the system design goal of high cohesion and low coupling, and can accurately analyze the cause of noise formation; 2. Convenient maintenance and problem finding: Each test module is independent of each other. If one module fails, only the corresponding part needs to be replaced, without the need for maintenance and adjustment of the entire system; 3. High degree of visualization: The results generated by the analysis and processing of each module can be displayed in the form of pictures, which is convenient for understanding. Attached Figure Description

[0025] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0026] Figure 1 This is a physical image of the near-field and far-field noise measurement device of the present invention;

[0027] Figure 2 This is a top view of the near-field and far-field noise measurement device of the present invention;

[0028] Figure 3 This is an exploded view of the far-field microphone device of the present invention;

[0029] Figure 4 This is an exploded view of the particle image velocimeter of the present invention;

[0030] Figure 5 This is a schematic diagram of the wind turbine model of the present invention;

[0031] Figure 6 This is a schematic diagram of the far-field microphone noise measurement results of the present invention;

[0032] Figure 7 This invention relates to the location of the near-field microphone device corresponding to different frequencies of the sound source.

[0033] Figure 8 This is a physical layout diagram of the particle image velocimeter of the present invention;

[0034] Figure 9 This is a velocity distribution diagram of the experimental flow field of the particle image velocimeter of the present invention;

[0035] Figure 10 This is an experimental vortex distribution diagram of the particle image velocimeter of the present invention;

[0036] Figure 11 This is a schematic diagram of the wall pressure sensor installation of the present invention;

[0037] Figure 12 This is the spectrum diagram of the wall pressure sensor of the present invention. Detailed Implementation

[0038] like Figure 1-5 As shown, the wind turbine noise source localization and analysis device of the present invention includes a wind turbine model mounted on a wind tunnel rig, a far-field microphone device, a near-field microphone device, a particle image velocimeter, a wall pressure sensor device, and a computer. The wind turbine model includes a stator support, a drive motor, and a rotor impeller rotatably mounted on the stator support. The far-field microphone device includes a data acquisition card and multiple far-field microphones spaced at intervals around the wind turbine model. The far-field microphones are connected to the data acquisition card, and the data acquisition card is connected to the computer via USB. The far-field microphones are used to collect noise signals from the wind turbine noise source side.

[0039] The near-field microphone device includes a group of micro-microphones and a camera arranged horizontally or vertically with the wind turbine model. The near-field microphone array is used to collect noise on a specified plane and visualize it.

[0040] The particle image velocimeter includes a dual-pulse laser generator placed on the side of the wind turbine model, a CCD high-speed camera placed above the wind turbine model, a synchronous controller for controlling the dual-pulse laser generator and the CCD high-speed camera, and a PIV host. The particle image velocimeter is used to obtain the stall vortex and energy distribution in the flow field.

[0041] The wall pressure sensor device includes a wall pressure sensor fixed on the stator support and a pressure measuring microphone connected to the wall pressure sensor. The wall pressure sensor device is used to collect vibration signals on the stator blades of the wind turbine.

[0042] The computer is connected to the near-field microphone device, the drive motor, the PIV host, and the pressure measuring microphone, respectively.

[0043] The two far-field microphones are spaced 15° apart. The model of the far-field microphone is BSWA MA231. The model of the data acquisition card is NI-USB 6529. The micro-microphone group includes 112 micro-microphones and the micro-microphone group is arranged in a circular structure with the camera as the center.

[0044] The drive motor is a DC motor, and the voltage can be adjusted freely between 0-12V.

[0045] The method for locating and analyzing the sound source of a wind turbine involved in this invention comprises the following specific steps:

[0046] The first step is to measure the noise frequency signals at different locations using the far-field microphone device and determine the frequency that needs to be analyzed.

[0047] Acoustic data was acquired at a sampling frequency of 51200Hz and analyzed in 4096 blocks, yielding a frequency resolution of 12.5Hz. (A higher sampling frequency represents richer information, but also results in a larger data volume and higher computational complexity. According to the sampling theorem, when the sampling frequency is greater than twice the highest frequency in the signal, the sampled digital signal completely retains the information in the original signal. In practical applications, the sampling frequency is generally kept to be 5 to 10 times the highest frequency of the signal. Our experiment mainly studies the frequency range of the human ear, focusing on the frequency band from 100 to 10000Hz, hence the setting of 51200Hz. 51200Hz / 4096 blocks yields a frequency resolution of 12.5Hz. The more blocks, the higher the resolution.) The lower the resolution, the denser the digital signal, and the more realistic the obtained signal. The Welch method is used to calculate the PSD (the idea behind the Welch method is: first, divide the signal of length N into L segments, each segment being M in length, then N = LM; then apply a window function w to each segment to calculate the power spectrum (PSD) of each segment; finally, average the power spectra of each segment to obtain the power spectrum of the entire signal. There can be overlap between data segments, and the window function w is the Hanning window. The Welch method is the most commonly used method for calculating PSD in EEG. In Matlab software, the pwelch function can be used directly to perform power spectrum estimation using the Welch method, and its general calling format is as follows:

[0048] [Pxx,F]=pwelch(X,WINDOW,NOVERLAP,NFFT,Fs)

[0049] [Pxx,F]=pwelch(X,WINDOW,NOVERLAP,NFFT,Fs,REQRANGE)

[0050] For details on the meaning of the parameters in the function, you can enter "help pwelch" in the Matlab command window to view its usage (it will not be elaborated here). The statistical confidence level is obtained using the Hanning window. Overlap is set to the function's default value. A Fast Fourier Transform is performed on 51200 data points with a frequency resolution of 1Hz. The calculation formulas for linear sound pressure level and total sound pressure level are used.

[0051]

[0052] The spectrum diagrams and total sound pressure levels for different frequency bands at the monitoring points are as follows:

[0053]

[0054] Where SPLi represents the sound pressure level at different frequencies, and p represents the sound pressure value at different frequencies. ref For reference sound pressure level;

[0055] In MATLAB software, the pwelch function is used to process the acquired time-domain signal of wall pulsating pressure, converting it into a frequency-domain signal, such as... Figure 6 As shown.

[0056] The second step is to obtain the main location of the noise at the corresponding frequency using the near-field microphone device.

[0057] Acoustic signals from a near-field microphone array were recorded for 8 seconds at a sampling rate of 48 kHz and analyzed using a block size of 2048, resulting in a frequency resolution of approximately 24 Hz. Eigenvalue-optimized beamforming was used to post-process the acquired data to locate noise sources, obtaining the specific locations of the noise sources at the frequencies to be analyzed. Figure 7 As shown.

[0058] The third step is as follows: Figure 8 As shown, the flow field velocity, vortex structure, and pressure signals are obtained by capturing images of the main locations of the noise at this frequency using the particle image velocimeter.

[0059] Tracer particles with an average diameter of approximately 1 μm, generated by the combustion of sulfur powder, were distributed in the flow field for easy capture by a CCD camera. The host computer analyzed two consecutive flow field images to calculate the particle velocities in the flow field. Further analysis of these velocities revealed the distribution of stall vortices in the flow field. The thickness of the laser sheet in the experimental model area was approximately 1 mm. 500 images of the flow field were captured at each different angle of attack. Each set of images was processed using a 32×32 pixel cross-correlation region and a 50×50% overlay. First, the image magnification of the analysis section was determined in MicroVec software based on the captured scale images. Then, the analysis section was divided into smaller interrogation regions (32×32), and the particle images in each interrogation region were analyzed using a cross-correlation algorithm. Batch calculations were then performed on all particle image pairs with good results to obtain the average velocity vector field of the measured section. Finally, the batch calculation results were imported into Tecplot software for further analysis to obtain additional information about the flow field of the captured section, such as… Figure 9-10 As shown.

[0060] The fourth step involves obtaining the frequency domain signal of the pressure through the wall pressure sensor device and combining it with the data from the third step to analyze the cause of noise formation.

[0061] Drill pressure measurement holes on the blades of the stator support of the wind turbine model, and install the wall pressure sensor shown in the figure on them. The pressure measurement holes are led out through a thin copper tube embedded inside the plate, as shown in the figure. Figure 11 As shown, a copper tube and a rubber tube are connected at the end of the flat plate to transmit the pressure signal to a pressure-measuring microphone of model BSWA MPA416, which converts the wall pressure signal into an electrical signal, such as... Figure 12 As shown, on the one hand, the frequency value at the signal peak can be obtained, and combined with the spectrum obtained from the far-field noise experiment, if there is a corresponding peak, then the source of the pressure pulsation can be obtained (because some peaks in the far-field noise spectrum can be directly identified as the main cause of their generation); on the other hand, the pressure pulsation changes at different locations are different, and the flow field information, velocity, pressure, and vortex structure in the PIV experiment can be combined to conduct a specific analysis and explain the pressure changes.

[0062] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention are included within the protection scope of the present invention.

Claims

1. A fan sound source localization and analysis device, characterized in that: The system includes a wind turbine model mounted on a wind tunnel rig, a far-field microphone device, a near-field microphone device, a particle image velocimeter, a wall pressure sensor device, and a computer. The wind turbine model includes a stator support, a drive motor, and a rotor impeller rotatably mounted on the stator support. The far-field microphone device includes a data acquisition card and multiple far-field microphones spaced at intervals around the wind turbine model. The far-field microphones are connected to the data acquisition card, which is connected to the computer via USB. The far-field microphones are used to acquire noise signals from the wind turbine noise source side. The near-field microphone device includes a group of micro-microphones and a camera arranged horizontally or vertically with the wind turbine model. The near-field microphone array is used to collect noise on a specified plane and visualize it. The particle image velocimeter includes a dual-pulse laser generator placed on the side of the wind turbine model, a CCD high-speed camera placed above the wind turbine model, a synchronous controller for controlling the dual-pulse laser generator and the CCD high-speed camera, and a PIV host. The particle image velocimeter is used to obtain the stall vortex and energy distribution in the flow field. The wall pressure sensor device includes a wall pressure sensor fixed on the stator support and a pressure measuring microphone connected to the wall pressure sensor. The wall pressure sensor device is used to collect vibration signals on the stator blades of the wind turbine. The computer is connected to the near-field microphone device, the drive motor, the PIV host, and the pressure measuring microphone, respectively.

2. The fan sound source localization and analysis device according to claim 1, characterized in that: The two far-field microphones are spaced 15° apart, and the micro-microphone group includes 112 micro-microphones, which are arranged in a circular structure with the camera as the center.

3. A method for locating and analyzing the sound source of a wind turbine, implemented using the wind turbine sound source location and analysis device according to any one of claims 1-2, characterized in that: The specific steps are as follows: The first step is to measure the noise frequency signals at different locations using the far-field microphone device and determine the frequency that needs to be analyzed. The second step is to obtain the main location of the noise at the corresponding frequency using the near-field microphone device. The third step involves using the particle image velocimeter to capture images of the main locations of the noise at that frequency to obtain the flow field velocity, vortex structure, and pressure signals. The fourth step involves obtaining the frequency domain signal of the pressure through the wall pressure sensor device and combining it with the data from the third step to analyze the cause of noise formation.

4. The method for locating and analyzing the sound source of a wind turbine according to claim 3, characterized in that: In the first step, acoustic data is acquired using the far-field microphone device, the PSD is calculated using the Welch method, and the statistical confidence level is obtained using the Hanning window, based on the calculation formulas for linear sound pressure level and total sound pressure level. The spectrum diagrams and total sound pressure levels for different frequency bands at the monitoring points are as follows: Where SPLi represents the sound pressure level at different frequencies, and p represents the sound pressure value at different frequencies. ref For reference sound pressure level; The pwelch function in MATLAB software is used to process the acquired wall pulsating pressure time-domain signal and convert it into a frequency-domain signal.

5. The method for locating and analyzing the sound source of a wind turbine according to claim 3, characterized in that: The third step involves processing multiple images captured by the particle image velocimeter. The specific steps are as follows: (1) Determine the image magnification of the analysis section in the MicroVec software based on the captured scale image; (2) Divide the analysis section into smaller query domains (32×32) and use the cross-correlation algorithm to analyze the particle images in each query domain; (3) Perform batch calculations on all particle image pairs with good results to obtain the average velocity vector field of the measured cross section; (4) Import the results of the batch calculation into Tecplot software for further analysis to obtain additional information about the flow field of the photographed section.

Images