CN-119959876-B - Asynchronous measurement method of cyclostationary sound source, storage medium and computer equipment

CN119959876BCN 119959876 BCN119959876 BCN 119959876BCN-119959876-B

Abstract

The invention provides an asynchronous measurement method of a cyclostationary sound source, a storage medium and computer equipment. The asynchronous measurement method of the cyclostationary sound source comprises the following steps of carrying out P times of measurement on the sound source to be measured at different positions by utilizing a microphone array with M channels, respectively calculating inter-circulating spectrum matrixes of sound signals obtained by the microphone array at different positions, stacking the inter-circulating spectrum matrixes according to diagonal lines to obtain a data-missing inter-circulating spectrum matrix with the dimension of MP multiplied by MP, and completing matrix complementation of the data-missing inter-circulating spectrum matrix based on a complementation target structure constraint problem of the data-missing inter-circulating spectrum matrix to form a complete inter-circulating spectrum matrix. The technical scheme of the invention can be widely applied to the technical field of the measurement of the cyclostationary sound source.

Inventors

XIAO YOUHONG
KUANG YI
HE JIANYUAN
WANG SIHAN
ZHANG CHENYU

Assignees

哈尔滨工程大学

Dates

Publication Date: 20260505
Application Date: 20250118

Claims (4)

1. The asynchronous measurement method of the cyclostationary sound source is characterized by comprising the following steps of: S1, measuring a sound source to be measured P times at different positions by using a microphone array with M channels; S2, selecting the circulating frequency of the sound source to be tested Spectral frequencies of imaging ; S3, respectively calculating a mutual cyclic spectrum matrix of the acoustic signals obtained by the microphone array at the different positions; S4, stacking the P inter-cyclic spectrum matrixes with the dimension of MxM according to diagonal lines to obtain a data-missing inter-cyclic spectrum matrix with the dimension of MxMP; S5, completing matrix complementation of the data-missing inter-cyclic spectrum matrix based on the constraint problem of the complement target construction of the data-missing inter-cyclic spectrum matrix to obtain a complement inter-cyclic spectrum matrix; The constraint condition for constructing the constraint problem in the step S5 comprises weak sparsity of a cross cyclic spectrum matrix eigenvalue spectrum; the constraint conditions for constructing the constraint problems in step S5 include spatial continuity of the cyclostationary sound field; Expression of the constraint problem in step S5: , Wherein, the The product of the Hadamard is represented, A cross-cyclic spectrum matrix is deleted for the data; To complement the whole with each other a cyclic spectrum matrix; Is a sampling matrix; at a frequency of Projection base of lower structure, abbreviated as ; At a frequency of Projection base of lower structure, abbreviated as 。
2. The method for unsynchronized measuring of a cyclostationary sound source according to claim 1, wherein the method for completing matrix completion of the data-missing inter-cyclic spectrum matrix in step S5 comprises FISTA method, the expression of FISTA method: , Wherein, the Is the step size; Representing a parameter introduced to accelerate the iteration; is an intermediate quantity of the iterative process; Is a soft threshold contraction operator, calculated as Diag is an operator that extracts or constructs diagonal elements of the matrix, The representation takes the sum of 0 Is a larger value of (a); Is that A eigenvector matrix for eigenvalue decomposition; Is that And decomposing the eigenvalues.
3. A storage medium having stored thereon a computer program, characterized in that said computer program, when executed by a processor, realizes step S5 in the method for unsynchronized measuring of cyclostationary sound sources according to any one of claims 1 to 2.
4. Computer device, characterized in that it comprises a memory, a processor and a computer program stored on said memory and executable on said processor, said processor executing said computer program to carry out step S5 of the method for unsynchronized measuring of cyclostationary sound sources according to any one of claims 1 to 2.

Description

Asynchronous measurement method of cyclostationary sound source, storage medium and computer equipment Technical Field The invention relates to the technical field of sound source positioning, in particular to the technical field of positioning of cyclostationary sound sources. Background Mechanical devices produce noise during operation, which is also known as a source of environmental pollution. Accordingly, there is an increasing concern about reducing the intensity of noise generated by mechanical devices, and related legal regulations regulate industrial devices that may be contaminated with noise. The precondition for noise reduction of mechanical devices requires the determination of a noise source. At present, acoustic imaging techniques based on microphone arrays are applied in the industrial field. The core idea of this technique is to infer the acoustic physical quantity of the region of interest using sound field data measured by a microphone array, and obtain an acoustic image. The obtained acoustic image can be used for noise source identification, acoustic-based fault diagnosis and the like. The currently accepted acoustic imaging methods can be divided into two categories, acoustic beamforming and near-field acoustic holography, according to the imaging principle. Cyclostationary signals are a special class of non-stationary signals whose statistics exhibit periodic variations (first and second order statistics are of general interest for rotating machinery). The cyclostationarity of the rotating mechanical radiated acoustic signal is created by the rotation and reciprocation of the plurality of mechanical components. The cycle frequencies corresponding to the different rotating parts are different. The existing cyclostationary acoustic source beamforming method (CSCBF) inverts the cyclic spectrum correlation of the acoustic source (reflects the cyclostationary nature of the acoustic source). The cyclostationary sound source beam forming method is limited by the aperture of a microphone array for acquiring sound signals and the density of microphones, and has the advantages of large main lobe at low frequency and serious side lobe at high frequency, and influences the recognition effect of the cyclostationary sound source. Disclosure of Invention In order to solve the problem that the identification effect of the cyclostationary sound source is affected in the conventional cyclostationary sound source positioning method, the invention provides an asynchronous measurement method, a storage medium and computer equipment of the cyclostationary sound source. The technical scheme of the invention is as follows: the asynchronous measurement method of the cyclostationary sound source comprises the following steps: S1, measuring a sound source to be measured P times at different positions by using a microphone array with M channels; s2, selecting the circulating frequency alpha of the sound source to be detected and the imaging frequency spectrum f; s3, respectively calculating a mutual cyclic spectrum matrix of the acoustic signals obtained by the microphone array at the different positions; S4, stacking the P inter-cyclic spectrum matrixes with the dimension of MxM according to diagonal lines to obtain a data-missing inter-cyclic spectrum matrix with the dimension of MxMP; s5, completing matrix complementation of the data-missing inter-cyclic spectrum matrix based on the constraint problem of the complement target construction of the data-missing inter-cyclic spectrum matrix, and obtaining the complement inter-cyclic spectrum matrix. Optionally, the constraint condition for constructing the constraint problem in step S5 includes weak sparsity of a cross-cyclic spectrum matrix eigenvalue spectrum. Optionally, constructing constraints for the constraining problem in step S5 includes cycling the spatial continuity of the sound field. Optionally, the expression of the constraint problem in step S5: wherein, as indicated by the terms, hadamard product, truncated error, S PP is the data-missing inter-cyclic spectrum matrix, S is the complement inter-cyclic spectrum matrix, and Ω is the sampling matrix; at a frequency of A projection base of lower structure, abbreviated as ψ L; at a frequency of The projection base of the lower structure is abbreviated as ψ R. Optionally, the method for completing matrix complement of the data-missing inter-cyclic spectrum matrix in step S5 includes a FISTA method, and an expression of the FISTA method is as follows: Where μ is the step size, λ k represents the regularization parameter, t k represents a parameter introduced to accelerate the iteration, G k is the intermediate quantity of the iteration process, and krnk (G k,λk μ) is the soft threshold contraction operator calculated as Diag is an operator that extracts or constructs diagonal elements of the matrix,The representation takes the sum of 0U is a eigenvector matrix decomposed by the eigenvalue of G k; The cha