CN1472514A - Sound field reconfiguration method for rapid noise diagnosis - Google Patents

Sound field reconfiguration method for rapid noise diagnosis Download PDF

Info

Publication number
CN1472514A
CN1472514A CNA031293816A CN03129381A CN1472514A CN 1472514 A CN1472514 A CN 1472514A CN A031293816 A CNA031293816 A CN A031293816A CN 03129381 A CN03129381 A CN 03129381A CN 1472514 A CN1472514 A CN 1472514A
Authority
CN
China
Prior art keywords
sound field
acoustic pressure
sound
node
measurement point
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CNA031293816A
Other languages
Chinese (zh)
Other versions
CN1219194C (en
Inventor
杨瑞梁
汪鸿振
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN 03129381 priority Critical patent/CN1219194C/en
Publication of CN1472514A publication Critical patent/CN1472514A/en
Application granted granted Critical
Publication of CN1219194C publication Critical patent/CN1219194C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

The method divides sound field into a numbers of regions and to present sound field at each region as linear superposition of region node sound pressure according to sound pressure at several known measuring points in certain sound field of known sound source shape. The regional node sound pressure can be obtained according to sound pressure of measuring point through least square method and singular value decomposition, and then sound pressure for the complete sound field can be rebuilt through sound pressure of regional node. The method is more accurate and efficient than HELS method in describing sound field.

Description

The sound field reconstructing method of quick noise diagnostics
Affiliated technical field
What the present invention relates to is a kind of sound field reconstructing method, particularly a kind ofly can come the sound field reconstructing method of the quick noise diagnostics of other dot informations of reconstruct accurately according to the information of the less point of measuring in the sound field, belongs to the noise diagnostics technical field.
Background technology
In the noise failure diagnosis, may can only measure noise signal seldom, how in a short period of time, be similar to unknown signaling in the sound field, cause many scholars' attention according to these less known signals.
Adopt computing method such as near-field holography, boundary element to carry out noise diagnostics, generally expend time in more or cost higher.1997, people such as Wang proposed to use the HELS method, and used analog computation and experimental study that the HELS method is verified respectively 1998,2000, calendar year 2001 etc.HELS method reconstruct sound field efficient is very high, and simple, flexible, does not have Element BEM in the not unique shortcoming of feature wave number.Therefore, the HELS method is extremely suitable is applied to quick noise diagnostics.In the prior art, the patent No. is 5,712,805 United States Patent (USP) " noise diagnostics system " is to adopt the HELS method, acoustic pressure in the sound field is converted into the stack of spheric wave function, uses least square method then, come the acoustic pressure on reconstruct sound source surface according to known noise signal.The problem that this method exists is: when independent function is selected improper or selected the measurement point of different radii is arranged, use the sound field of HELS method reconstruct pulsation ball also to produce than mistake.Require the point of reconstruct very approaching from the point of measuring, and, use the HELS method to be difficult to the sound field of the common spherical sound source of reconstruct, for non-ball sound source, the error of calculating is very big.This method is inapplicable for high frequency, and very low to the efficient of non-ball sound source sound field reconstruct.
Summary of the invention
Be deficiency and the defective that overcomes prior art, expand the scope of application of traditional HELS method, and keep the high-level efficiency of traditional HELS method.The present invention has proposed a kind of sound field reconstructing method that is used for quick noise diagnostics on the basis of traditional HELS method and Infinite Element Method.This new sound field reconstructing method, wider than the scope of application of traditional HELS method, efficient is higher.
This method is that the acoustic pressure of known some spots (being called measurement point) is divided into several zones to sound field according in known certain sound field of sound source shape, and each regional sound field is expressed as the linear superposition of Area Node acoustic pressure.According to the acoustic pressure of measurement point,, can obtain the acoustic pressure of Area Node by least square method and svd.By the acoustic pressure of Area Node, can obtain the acoustic pressure of each regional every bit then.Concrete grammar is as follows:
1, sets up ellipsoidal coordinates
The sound source near surface draw an ellipsoid more approaching with the sound source shape (also can be drawn in the sound source by fraction, in a word will with the roughly similar ellipsoid of sound source shape), set up ellipsoidal coordinates.
2, zoning and node
According to the number and the distribution situation of measurement point in the sound field, sound field is divided into several zones along the change in coordinate axis direction of ellipsoidal coordinates.If outside sound field, just the radial direction along ellipsoidal coordinates is divided into several infinite regions, and each infinite region extends along radial direction is outwards unlimited; If internal acoustic field just is divided into limited area.
If the number of measurement point is more in the sound field, the number of regions in the sound field needs more; The dense place of measurement point distribution also needs more zone in the sound field.There is plurality of nodes in each zone.For the purpose of convenience of calculation was accurate, the node in zone was next to measurement point as much as possible.
3, calculate the interpolating function of each measurement point
Utilize the interpolating function formula in infinite element method zone: ψ = ψ μ ( r ) ψ v ( φ ) ψ w ( θ ) . . . . . . ( 1 ) Wherein (r, φ θ) are ellipsoidal coordinates, ψ v (φ), ψ w (θ)Be respectively along φ, traditional finite element interpolating function of θ direction is chosen as the Lagrange interpolation polynomial here, if the sound field of limited area, ψ μ (r)Be the interpolation polynomial of r direction, also can be chosen as the Lagrange interpolation polynomial here.Because for the sound field of infinite region, must consider the infinite distance radiation boundary condition, be chosen as here: ψ μ r ( r μ ) = e - ik ( r - r μ ) Σ μ . = 1 n h μ μ . ( kr ) μ . μ = 1,2 , Λ , n ( n ≥ 2 ) . . . . . . ( 2 ) H=S wherein -1, S 1m=(kr m) -1, r μThe radial coordinate of representing μ node, k represents wave number.(2) formula also may be thought of as shifted Legendre polynomial expression and shifted Chebyshev polynomial expression simultaneously, and its ultimate principle is identical, and (2) formula is simple a little but use both calculation procedures of writing of back.
4, calculate the node acoustic pressure
Come the computing node acoustic pressure according to the sound pressure level of measurement point and the interpolating function of measurement point:
{C}=[ζ] u{ξ} (3)
Wherein, can calculate by interpolating function and measurement point acoustic pressure respectively in matrix [ζ] and the vector { ξ }.Matrix element wherein is respectively ζ mn = ψ jm * ψ nj * , ξ 1 = P oi ψ ji * , [ζ] uBe the pseudoinverse of trying to achieve by svd, P () iExpression measurement point acoustic pressure.
In particular cases, can be chosen as node to measurement point, calculating can be omitted this process 3 and 4 like this, and the calculated amount of this moment will sharply descend, and counting yield will be greatly improved.
5, calculate the interpolating function of reconstruction point
The computing method of reconstruction point interpolating function are as 3, and computing formula still is (1) (2).
6, reconstruct in the sound field acoustic pressure of any arbitrarily
The acoustic pressure of reconstruction point can be calculated according to following acoustic pressure equation: P ( x ) = Σ i = 1 N C i ψ i ( x ) . . . . . . ( 4 )
What N wherein represented is the number of this zone interior nodes, C I(x) acoustic pressure of i node of expression, ψ iInterpolating function for reconstruction point.
Because the acoustic pressure of use node is described the information in the sound field, be equivalent to use rational interpolating function to come the sound field of approximate region, do not attempt to find the solution the simple function that to describe whole sound field, therefore sound field can be described more accurate than HELS method, more efficiently, sound source, each frequency range sound field reconstruct problem that can the approximate solution different shape.Again because be quick noise diagnostics, and do not require the acoustic pressure of accurately describing each point, so the zone can be seldom.The present invention has avoided using the loaded down with trivial details equation integration of the unlimited unit of ellipsoid sound, thereby has significantly reduced the calculating time spent.This method had both kept the high-level efficiency of traditional HELS method, had widened the calculating suitable application area of this method again widely.This method can be generalized to unbounded fields such as magnetic field, wave stroke equally easily.
Description of drawings
Fig. 1 is dividing region and a node synoptic diagram in certain long rotation ellipsoid sound source sound field.Contain 20 infinite regions, 40 nodes among the figure altogether, each 4 node in zone, for example 1,20,21,40 4 node is contained in zone one, and 1,2,21,22 4 node is contained in zone two, or the like.Among the figure from inside to outside the radius of ellipsoid be respectively 1,2,10, infinity, represent the radius of sound source, the radius of node 1~20 place ellipsoid, the radius of node 21~40 place ellipsoids, infinite distant place successively.
The sound pressure amplitude of the reconstruct of the long rotation ellipsoid sound source of Fig. 2 is with the variation of θ, and solid line representation theory value, * add dotted line and represent reconstruction value of the present invention, and adds the reconstruction value that dotted line represents to use the HELS method of 6 independent functions.Wherein the longitudinal axis represent sound pressure amplitude (unit is: handkerchief), transverse axis denotation coordination θ.
Embodiment
Below in conjunction with accompanying drawing concrete enforcement of the present invention is further described.
As shown in Figure 1, one long rotation ellipsoid sound source, its major axis and minor axis are respectively 1 and 0.5 meter, radius is on 3 and 9 (coordinate the is ellipsoidal coordinates) ellipsoid far away in the sound field, select a measurement point along angular direction θ from 4.5 degree every 9 degree, have 80 measurement points, the acoustic pressure of 80 measurement points is known.
According to these 80 measurement point acoustic pressures, the process that reconstructs whole sound field acoustic pressure is as follows:
1, sets up ellipsoidal coordinates
Set up ellipsoidal coordinates (only having used two-dimensional coordinate here), its major axis is 1 meter, and minor axis is 0.5 meter.Wherein ellipsoidal coordinates (r, θ) and Cartesian coordinates (x, corresponding relation y) is:
Figure A0312938100061
2, zoning and node
By the distribution situation of measurement point, whole sound field is divided into 20 zones along unlimited direction, there are 4 nodes in each zone.Each node is all on radius is 10 and 2 ellipsoid.4 nodes are contained in each zone.
3, calculate the interpolating function of each measurement point
Here set: ψ 1 ( θ ) ( ξ ) = 1 2 ( ξ - 1 ) ψ 2 ( θ ) ( ξ ) = 1 2 ( ξ + 1 ) θ ( ξ ) = Σ v = 1 n ψ v ( φ ) ( ξ ) θ v . . . . . . ( 6 )
Coordinate figure by measurement point can calculate: ψ 1 r ( r ) = e - i × 11.088 × ( r - 2 ) [ - 1 2 r + 5 r 2 ] ψ 2 r ( r ) = e - i × 11.088 × ( r - 10 ) [ 25 2 r - 25 r 2 ] . . . . . . ( 7 )
According to formula (1), can calculate the interpolating function of measurement point.
4, calculate the node acoustic pressure
According to formula (3), the acoustic pressure that can calculate each node is planted.
5, calculate the interpolating function of reconstruction point
The interpolating function computing method of reconstruction point are identical with 3.
6, reconstruct in the sound field acoustic pressure of any arbitrarily
According to the interpolating function value of reconstruction point and the sound pressure level of node,, can calculate the acoustic pressure of reconstruction point by formula (4).Because the position of reconstruction point can be selected arbitrarily, thus sound pressure level that can the whole sound field of reconstruct.
Reconstruction point is that result of calculation is shown in * among Fig. 2 on 5 the ellipsoid at radius, and the * with dashed lines among the figure links up.
Adopt Element BEM to calculate theoretical approximate value, represent with solid line in the drawings.
For superiority of the present invention is described, use 6 independent functions of HELS method to calculate same example, result of calculation is shown in among Fig. 2, and the with dashed lines among the figure links up.
As can be seen, reconstruction result of the present invention overlaps with theoretical solution is approximate from the figure.
Superiority of the present invention can also display by following reconstruction result.
(1) computing time
The 0.04 second used time of the present invention, and the HELS method surpasses 0.15 second (not comprising the orthogonalized time of its independent function).
(2) computational accuracy
Reconstructed error of the present invention is less than 1%, and the reconstructed error of HELS method nearly 200%.
Therefore, the new sound field reconstructing method of the present invention's proposition is superior to the HELS method far away.

Claims (1)

1, a kind of sound field reconstructing method of quick noise diagnostics, it is characterized in that: this method is according in known certain sound field of sound source shape, the acoustic pressure of known some spots (being called measurement point), sound field is divided into several zones, each regional sound field is expressed as the linear superposition of Area Node acoustic pressure, acoustic pressure according to measurement point, by least square method and svd, can obtain the acoustic pressure of Area Node, acoustic pressure by Area Node then, can reconstruct the acoustic pressure of whole sound field, concrete grammar is:
(1) sets up ellipsoidal coordinates
Draw an ellipsoid more approaching at the sound source near surface, set up ellipsoidal coordinates based on this with the sound source shape;
(2) zoning and node
Number and distribution situation according to measurement point in the sound field, change in coordinate axis direction along ellipsoidal coordinates is divided into several zones to sound field, if outside sound field, just the radial direction along ellipsoidal coordinates is divided into several infinite regions, and each infinite region extends along radial direction is outwards unlimited; If internal acoustic field just is divided into limited area;
(3) calculate the interpolating function of each measurement point
Utilize the interpolating function formula in infinite element method zone to calculate: ψ = ψ μ ( r ) ψ v ( φ ) ψ w ( θ ) ;
(4) calculate the node acoustic pressure
According to the sound pressure level of measurement point and the interpolating function of measurement point, come the computing node acoustic pressure by least square method and svd:
{C}=[ζ] u{ξ};
(5) calculate the interpolating function of reconstruction point
The computing method of reconstruction point interpolating function are the same with 3;
(6) reconstruct in the sound field acoustic pressure of any arbitrarily
According to the acoustic pressure of the interpolating function and the node of reconstruction point, calculate by following acoustic pressure equation: P ( x ) = Σ i = 1 N C i ψ i ( x ) .
CN 03129381 2003-06-19 2003-06-19 Sound field reconfiguration method for rapid noise diagnosis Expired - Fee Related CN1219194C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 03129381 CN1219194C (en) 2003-06-19 2003-06-19 Sound field reconfiguration method for rapid noise diagnosis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 03129381 CN1219194C (en) 2003-06-19 2003-06-19 Sound field reconfiguration method for rapid noise diagnosis

Publications (2)

Publication Number Publication Date
CN1472514A true CN1472514A (en) 2004-02-04
CN1219194C CN1219194C (en) 2005-09-14

Family

ID=34153504

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 03129381 Expired - Fee Related CN1219194C (en) 2003-06-19 2003-06-19 Sound field reconfiguration method for rapid noise diagnosis

Country Status (1)

Country Link
CN (1) CN1219194C (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1333243C (en) * 2004-06-15 2007-08-22 东北师范大学 Dynamic measuring system with four sensors for acoustic field spacial distribution of acho signals of bat
CN102089634B (en) * 2008-07-08 2012-11-21 布鲁尔及凯尔声音及振动测量公司 Reconstructing an acoustic field
CN102089633B (en) * 2008-07-08 2013-01-02 布鲁尔及凯尔声音及振动测量公司 Method for reconstructing an acoustic field
CN104778321A (en) * 2015-04-14 2015-07-15 哈尔滨工程大学 Analyzing and forecasting method for acoustic space with complicated shape
CN111220267A (en) * 2020-01-09 2020-06-02 上海索辰信息科技有限公司 Airborne noise sound field reproduction method
CN112415882A (en) * 2020-12-15 2021-02-26 厦门华钰电子科技有限公司 Electronic wall clock with sound intensity measuring device
CN112782492A (en) * 2021-01-05 2021-05-11 华北电力大学(保定) Audible noise three-dimensional distribution calculation method for alternating-current transmission line
CN115655631A (en) * 2022-12-12 2023-01-31 杭州兆华电子股份有限公司 Method and device for detecting voiceprint in wind tunnel environment based on hydraulic generator

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1333243C (en) * 2004-06-15 2007-08-22 东北师范大学 Dynamic measuring system with four sensors for acoustic field spacial distribution of acho signals of bat
CN102089634B (en) * 2008-07-08 2012-11-21 布鲁尔及凯尔声音及振动测量公司 Reconstructing an acoustic field
CN102089633B (en) * 2008-07-08 2013-01-02 布鲁尔及凯尔声音及振动测量公司 Method for reconstructing an acoustic field
CN104778321A (en) * 2015-04-14 2015-07-15 哈尔滨工程大学 Analyzing and forecasting method for acoustic space with complicated shape
CN104778321B (en) * 2015-04-14 2017-11-17 哈尔滨工程大学 A kind of analysis and prediction method of complicated shape acoustic space
CN111220267A (en) * 2020-01-09 2020-06-02 上海索辰信息科技有限公司 Airborne noise sound field reproduction method
CN112415882A (en) * 2020-12-15 2021-02-26 厦门华钰电子科技有限公司 Electronic wall clock with sound intensity measuring device
CN112415882B (en) * 2020-12-15 2022-03-29 厦门欣盛航进出口有限公司 Electronic wall clock with sound intensity measuring device
CN112782492A (en) * 2021-01-05 2021-05-11 华北电力大学(保定) Audible noise three-dimensional distribution calculation method for alternating-current transmission line
CN115655631A (en) * 2022-12-12 2023-01-31 杭州兆华电子股份有限公司 Method and device for detecting voiceprint in wind tunnel environment based on hydraulic generator
CN115655631B (en) * 2022-12-12 2023-04-07 杭州兆华电子股份有限公司 Method and device for detecting voiceprint in wind tunnel environment based on hydraulic generator

Also Published As

Publication number Publication date
CN1219194C (en) 2005-09-14

Similar Documents

Publication Publication Date Title
CN1219194C (en) Sound field reconfiguration method for rapid noise diagnosis
CN1037036C (en) Method and apparatus for rotary actuator arc compensation correction in a direct access storage device
CN1891159A (en) Quick scanning conversion method for 3-D supersonic imaging
CN1265294C (en) Address mapping method and system for FFT processor with completely parallel data
Betthauser Topological reconstruction of grayscale images
CN113866718A (en) Matching field passive positioning method based on co-prime matrix
Belan et al. Extraction of characteristic features of images with the help of the radon transform and its hardware implementation in terms of cellular automata
Corrochano et al. Grassmann-Cayley algebra and robotics applications
Bolognesi et al. Birational geometry of moduli spaces of configurations of points on the line
CN1758164A (en) Construction method of rod knife space scanning body based on compression body mould
CN1045343C (en) Chaos processor
Piepenbrock et al. 3D microbubble localization with a convolutional neural network for super-resolution ultrasound imaging
Hadnagy et al. Efficient implementation of convolutional neural networks on FPGA
CN102141431A (en) Method for measuring and transforming sound field in double-layer medium space
Mezey Chirality measures and graph representations
Shen et al. Radar semantic segmentation based on U-Net using vision transformer
Zhao et al. Pruned distributed and parallel subarray beamforming for 3-D underwater imaging with fine-grid sparse arrays
CN114325817B (en) Depth domain seismic synthetic record manufacturing method and device
Chatterjee et al. A note on finding extreme points in multivariate space
CN113099728A (en) Method for constructing and decomposing accurate decomposition equation of finite signal
Alayrangues et al. Homology of cellular structures allowing multi-incidence
Alayrangues et al. Border operator for generalized maps
Haupt et al. Effect of d-Dimensional Re-orderings on Lossless Compression of Radio-Astronomy and Digital Elevation Data
Ahlswede et al. Orthogonal Polynomials in Information Theory
CN113791378A (en) Two-dimensional angle analysis method, system, equipment and medium for sparse L-shaped array

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C19 Lapse of patent right due to non-payment of the annual fee
CF01 Termination of patent right due to non-payment of annual fee