Underwater structure noise control method based on parametric secondary sound source

Abstract

The application provides an underwater structure noise control method based on a parametric secondary sound source, which comprises the following steps: firstly, using the sound pressure and normal velocity information of a structure surface, a far-field noise prediction method is adopted to obtain the radiation noise at a virtual reference point in the far field, thereby providing an input basis for parametric secondary sound source control; secondly, a multi-channel parametric sound field model is established to verify the controllability of the parametric secondary sound field, and a regularization least square algorithm is adopted to obtain multi-channel signal emission parameters by taking the sum of the sound pressures at multiple virtual reference points in the far field as the minimum target; finally, when the monitoring information of the underwater structure is missing, a local control strategy is adopted to obtain the underwater structure noise control effect. The application is directed to directional control in the accurate area of far-field noise prediction, solves the problem of poor traditional active noise reduction effect caused by missing monitoring information, improves the reliability and stability of the system, and lays a theoretical foundation for the application of underwater structure active noise control technology.

Description

Technical Field

[0001] This invention belongs to the field of noise control technology, specifically relating to a method for controlling underwater structure noise based on parametric secondary sound sources. Background Technology

[0002] Active control of vibration and noise in underwater vehicle structures is of great significance to my country's maritime national security and economic development. With the continuous improvement of sonar systems' ability to detect low-frequency line spectrum noise, underwater vehicle structure noise control technology faces higher requirements and challenges. Active noise control technology cancels radiated noise from underwater structures along the sound propagation path, exhibiting good low-frequency control effects and thus attracting considerable attention. However, in the absence of surface monitoring information for underwater structures, traditional active noise reduction methods suffer from inaccurate sound field predictions by error sensors, leading to poor control of secondary sound sources and potentially even increased noise.

[0003] In active noise control systems, traditional sound sources, acting as secondary sound sources, may increase the sound wave energy received by the near-field reference sensor, which will reduce the control effect of the active control system and affect the stability of the system (Wu Shuaibing, Wu Ming, Yang Jun. Research on parametric array loudspeakers in pipeline noise control [J]. Applied Acoustics, 2013(6):439-445.). Multi-channel control systems cannot solve the contradiction between control effect and system stability. In order to obtain better control effect, multi-channel active control inevitably increases the complexity and computation of the system, reducing system stability. Low system complexity will lead to limited active control effect (Pang Yanbin, Chen Ke'an. Stability analysis of free-field distributed active acoustic control system [J]. Electroacoustic Technology, 2010(2):4.). Far-field noise prediction methods mainly include analytical methods, numerical methods, and a hybrid numerical-analytical method. Traditional prediction methods may involve singular integrals, high-dimensional matrix inversion, and the existence of non-unique solutions (Gordiyenko, Jia Zhifu. Acoustic Vector-Phase Technology [M]. National Defense Industry Press, 2014.), leading to inaccurate prediction results. Near the surface of underwater structures, there are locations where near-field monitoring sensors cannot be deployed. When monitoring information is missing, far-field radiated noise prediction results become inaccurate, reducing control effectiveness.

[0004] This invention addresses the problem of active control of far-field radiated noise from underwater target structures. It employs a multi-channel parametric secondary sound source emission method to perform directional noise control in areas with accurate far-field noise prediction. This solves the problem of excessively high risk coefficients caused by traditional local control of sound sources under conditions of missing monitoring information. At the same time, the strong directional beam formed by the parametric secondary sound source reduces acoustic coupling between multiple channels, improving the stability of the multi-channel active control system and laying a theoretical foundation for the application of active noise control technology for underwater structures. Summary of the Invention

[0005] This invention addresses the noise reduction problem of underwater target structures using parametric sound sources as secondary sound sources, and proposes a noise control method for underwater structures based on parametric secondary sound sources.

[0006] This invention is achieved through the following technical solution:

[0007] (1.1) Multiple vector hydrophones are deployed on the surface of the underwater structure to obtain the sound pressure and normal vibration velocity information of the underwater structure radiated noise. The distance between adjacent vector hydrophones is no more than one-fifth of the wavelength of the sound wave in the water.

[0008] The selection of the location, number, and spacing of monitoring points on the underwater structure surface determines whether the sound pressure amplitude and phase at the far-field observation point can be accurately predicted. Vector hydrophones are uniformly deployed at the selected monitoring points on the underwater structure surface, with the spacing between adjacent vector hydrophones not exceeding one-fifth of the wavelength of the sound wave in water. These monitoring points form an envelope that includes the noise source. The vector hydrophones deployed on the structure surface must meet the above deployment principles to accurately predict the radiated noise at the far-field observation point.

[0009] (1.2) Using the Helmholtz integral prediction method, based on the sound pressure and normal vibration velocity information obtained by each vector hydrophone on the surface of the underwater structure, the sound pressure amplitude and phase at each virtual reference point in the far field are calculated, and the radiation sound pressure matrix B composed of the complex sound pressure values ​​of the radiated noise at the virtual reference point in the far field is obtained.

[0010] By deploying vector hydrophones at various monitoring points, the sound pressure and normal vibration velocity information at each monitoring point on the underwater structure surface are obtained. The amplitude and phase of the sound pressure at the far-field virtual reference point are calculated using the Helmholtz integral theorem for sound pressure external field points, as shown in the following formula:

[0011]

[0012] in, The complex sound pressure level at the far-field virtual reference point P is... The complex sound pressure value at monitoring point Q on the surface of the underwater structure. For the free field Green's function, Let n be the normal sound pressure gradient at monitoring point Q, and n be the outward normal at monitoring point Q. for The infinitesimal element at the location, These represent the positions of the far-field virtual reference point and the monitoring point on the structural surface, respectively. PQ Let Q be the distance from the monitoring point Q on the underwater structure surface to the far-field virtual reference point P. Simplifying, we get:

[0013]

[0014] Where η' is n and The included angle, v is the normal complex vibration velocity at monitoring point Q, k and ω are the line spectrum noise frequency and wave number, respectively, and ρ is the medium density.

[0015] Substituting the sound pressure and normal velocity received by the vector hydrophone into the Helmholtz integral formula for the sound pressure at the outer field point, the sound pressure at each far-field virtual reference point is calculated. The complex sound pressure values ​​at all far-field virtual reference point locations are used as elements of the radiated noise sound pressure matrix at the far-field virtual reference point locations, resulting in the radiated noise sound pressure matrix B shown below:

[0016]

[0017] in, Let be the sound pressure radiated by the underwater target structure at the i-th virtual reference point in the far field. This represents the position of the i-th virtual reference point, where i = 1, 2, ..., M, and M is the number of far-field virtual reference points.

[0018] (1.3) The parametric secondary sound sources are evenly and uniformly distributed on the surface of the underwater structure, ensuring that their collimation direction is uniformly distributed across the entire solid angle range. The radiated sound field of the parametric array sound sources is derived based on the wave equation to obtain the radiated sound field of a single parametric secondary sound source. The radiated sound field generated by other parametric secondary sound sources is calculated through coordinate transformation to obtain the complex sound pressure value of each parametric secondary sound source at its virtual reference point in the far field. The obtained complex sound pressure values ​​are used as elements in the multi-channel sound transfer matrix to obtain the multi-channel sound transfer matrix C.

[0019] Parametric secondary sound sources are placed at equal intervals on the surface of the underwater structure to uniformly cover it. The collimation direction of the parametric secondary sound sources should be along the normal direction perpendicular to the tangent of the underwater structure surface, aiming for uniform distribution within the solid angle. The analytical expression for the radiated sound field of a single parametric secondary sound source is derived using the wave equation, as shown below:

[0020]

[0021] Where, p d (r,ξ) represents the complex sound pressure level at any point in the difference-frequency sound field, r is the distance from any point to the equivalent sound center of the parametric secondary sound source, ξ is the angle at which any point deviates from the collimation direction, and β f ω is the nonlinear coefficient of the water medium. d With k dLet P1 and P2 be the difference frequency and wave number, respectively; P1 and P2 be the amplitudes of the two pump waves, respectively; S0 be the radiation cross-sectional area of ​​the parametric secondary sound source; ρ be the density of the water medium; c0 be the speed of sound in the water medium; and D(ξ) and φ(ξ) be the amplitude directivity and phase directivity functions of the difference frequency beam of the parametric secondary sound source, respectively, as shown below:

[0022]

[0023]

[0024] Where, α s =α1+α2, where α1 and α2 are the attenuation coefficients of the two pump waves in the water medium, respectively.

[0025] Parametric secondary sound sources in Position in coordinate system is The position of the far-field virtual reference point in this spherical coordinate system is: First, deflect it by ψ in both the pitch and azimuth directions. θ , Then along The coordinates are translated in the direction, with point T as the location of the parametric secondary sound source. A spherical coordinate system is established with the equivalent sound center of the parametric secondary sound source on the structural surface as the reference. The position of the far-field virtual reference point in this spherical coordinate system is: ψ θ , The elevation and azimuth angles of the collimation direction of the secondary sound source are respectively measured, and R... P '、θ P '、 use The parameters are represented in the coordinate system, as shown in the following formula:

[0026]

[0027]

[0028]

[0029] in, All of these are intermediate variables, as detailed below:

[0030]

[0031]

[0032]

[0033] Using the above methods The position coordinates of each virtual reference point in the coordinate system are all transformed to In the coordinate system, using p d The expression (r,ξ) calculates the corresponding complex sound pressure value, since p d The sound field corresponding to the expression (r,ξ) is a two-dimensional axisymmetric sound field, and its input parameters are r and ξ. Therefore, The coordinates in the coordinate system are transformed into the form (r, ξ), and the specific transformation relationship is as follows:

[0034] r ij =R P '

[0035]

[0036] Where, r ij Let ξ be the distance from the j-th parameter secondary sound source point T to the i-th far-field virtual reference point P. ij Let the angle between the i-th far-field virtual reference point P and the collimation direction of the j-th parametric secondary sound source be the angle. Then, the complex sound pressure levels formed by each parametric secondary sound source at each virtual reference point are used as elements of the multi-channel sound transfer matrix, thus obtaining the multi-channel sound transfer matrix C:

[0037]

[0038] Where, p d (r ij ,ξ ij Let be the complex sound pressure value formed by the j-th parametric secondary sound source at the i-th far-field virtual reference point, i = 1, 2, ..., M, j = 1, 2, ..., N, where M and N are the total number of virtual reference points and the total number of parametric secondary sound sources, respectively.

[0039] (1.4) The number of far-field virtual reference points must not be less than the number of parametric secondary sound sources. When the number of far-field virtual reference points equals the number of parametric secondary sound sources, the emission weight coefficients X of the parametric secondary sound sources can be solved using the least squares method with the cost function J(X) minimized, based on elementary matrix operations. When the number of far-field virtual reference points exceeds the number of parametric secondary sound sources, it may cause poor stability of the noise control system and ill-posed problems in the matrix solution process. In this case, the emission signal weight coefficients X of the parametric secondary sound sources need to be obtained using the Tikhonov regularized least squares method.

[0040] When the number of far-field virtual reference points is less than the number of parametric secondary sound sources, the equation system with N unknowns and M variables obtained by elementary matrix simplification will produce an infinite number of solutions, and it will be impossible to find the optimal emission weight coefficients of the parametric secondary sound sources in the infinite number of solutions.

[0041] When the number of far-field virtual reference points equals the number of parametric secondary sound sources, the multi-channel sound transfer matrix C is a square matrix. Let the emission weighting coefficients of the parametric secondary sound sources be X, which is an N×1 column vector. The noise-controlled sound pressure matrix CX+B is obtained as shown below:

[0042]

[0043] in, For noise control at each far-field virtual reference point The complex sound pressure level at that location, This represents the position of the i-th virtual reference point, where i = 1, 2, ..., M, and M is the number of far-field virtual reference points.

[0044] The cost function J(X) is the sum of squares of all elements in CX+B, as shown below:

[0045] J(X) = ||CX + B|| 2 =(CX+B) H (CX+B)

[0046] =X H C H CX+X H C H B+B H CX+B H B

[0047] in,(·) H The conjugate transpose operation is represented by ||·||, and the 2-norm of the matrix-vector is represented by the square root of the sum of the squares of the elements in the vector. The least squares method is used to minimize the cost function J(X) to obtain the optimal emission weights X of the parametric secondary sound source. To minimize J(X), the derivative of J(X) with respect to X is set to 0, thus obtaining the optimal weight matrix X:

[0048] X = -C -1 B

[0049] When the number of far-field virtual reference points is greater than the number of parametric secondary sound sources, Tikhonov regularization is applied when using the least squares method, transforming the cost function J(X) into:

[0050] J(X) = ||CX + B|| 2 +ε 2 ||X 2

[0051] Where ε is the regularization parameter, the optimal parameter value ε is obtained by iterative weighting using the Bayesian criterion, and the derivative of J(X) with respect to X is set to 0, thus obtaining the optimal weight coefficient matrix X:

[0052] X = -(C H C+ε 2 ·I N×N ) -1 ·C H ·B

[0053] Among them, I N×N It is an N-row, N-column identity matrix.

[0054] (1.5) When monitoring information is missing, the parametric secondary sound source in the area where the monitoring information is inaccurate should be shut down. Utilizing the high directivity of the parametric secondary sound source, a local control strategy is adopted to accurately and effectively control noise in the local area where the monitoring information is accurate, achieving noise control throughout the entire space without causing noise leakage.

[0055] When monitoring information is missing, parametric secondary sound sources in areas with inaccurate predictions are shut down. The high directivity of the parametric sound sources is utilized to control noise in areas with accurate predictions by relying on parametric secondary sound sources within those areas. By performing precise local noise control only in areas with accurate noise predictions, good local noise control results are achieved, and the phenomenon of energy overflow and increased exposure caused by directional control after inaccurate noise predictions is avoided.

[0056] To measure the noise control effect across the entire space, the ratio of the sum of sound energy at each far-field virtual reference point before and after noise control is used to set the noise control effect evaluation index J in all directions. m The specific formula is as follows:

[0057]

[0058] Among them, J m This represents the ratio of sound field energy before and after noise control. Let i be the position of the i-th virtual reference point. and These represent the noise control before and after. The complex sound pressure level at point M, where M is the number of virtual reference points, and J is the number of virtual reference points. m It is related to the frequency of radiated noise within the envelope, the number of parametric secondary sound sources, and the number of far-field virtual reference points.

[0059] After noise control, energy enhancement may occur in certain directions, increasing the risk of underwater vehicle exposure in those directions. Therefore, a risk coefficient L is defined. m To measure the exposure risk of underwater structures in all spatial directions after noise control. Among them, L m As shown in the following formula:

[0060]

[0061] Among them, L mM represents the exposure risk coefficient. Top This represents the set of all virtual reference points that enhance energy after noise control.

[0062]

[0063] The advantages of the present invention compared to the prior art are:

[0064] This invention proposes an underwater structure noise control method based on a parametric secondary sound source. Using sound pressure and normal vibration velocity information monitored by a vector hydrophone, a far-field radiated noise prediction method based on the Helmholtz integral theorem for sound pressure at an external field point is employed to predict noise at a virtual reference point in the far field. This method reduces approximation errors, avoids solving integral equations and their inverse matrices, and avoids problems related to singular integrals, high-dimensional matrix inversion, and the existence of non-unique solutions. Compared to using a traditional point sound source, using a parametric sound source as a secondary sound source reduces the impact of acoustic coupling and acoustic feedback between channels on the signals received by near-field monitoring sensors, improves the stability of the control system, reduces system complexity, and enhances control performance.

[0065] The secondary sound field, controlled by the emission weight coefficients of the parametric secondary sound source obtained through the least squares method, can better cancel the primary sound field, achieving effective noise reduction. When monitoring information is missing, using parametric sound sources as secondary sound sources allows for local control of areas with accurate far-field radiated noise prediction, effectively improving control performance. Furthermore, the operating state of each parametric secondary sound source can be determined based on whether the far-field radiated noise can be accurately predicted in the corresponding area, enhancing its flexibility and environmental adaptability to a certain extent. Attached Figure Description

[0066] Figure 1 This is a flowchart of the method of the present invention;

[0067] Figure 2 Diagram of the underwater target structure;

[0068] Figure 3 Sound pressure distribution on the surface of an underwater target structure;

[0069] Figure 4 Location of vector hydrophones for monitoring points;

[0070] Figure 5 The results are the predictions for far-field radiated noise.

[0071] Figure 6 Establishment of the parametric secondary sound source placement method and coordinate system;

[0072] Figure 7 To monitor the control effect when monitoring information is complete;

[0073] Figure 8This is the prediction result for far-field radiated noise when monitoring information is missing;

[0074] Figure 9 To monitor the effectiveness of control measures when monitoring information is missing;

[0075] Figure 10 To monitor the effectiveness of a local control strategy using secondary sound sources as parametric sources when monitoring information is missing;

[0076] Figure 11 To monitor the effectiveness of a local control strategy for secondary sound sources when information is missing. Detailed Implementation

[0077] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0078] The first step, taking an underwater target cylindrical shell structure as an example, is a hollow cylindrical shell with a radius of 1m, a length of 14m, a thickness of 0.05m, and made of Steel AISI 4340. Specifically, as follows... Figure 2 As shown. A radial harmonic force of 1 N with a frequency of 200 Hz is applied along the positive z-axis at the midpoint of its surface generatrix. The excitation position coordinates in the rectangular coordinate system are (0,0,1). The point excitation generates a sound pressure distribution on the surface of the cylindrical shell, as shown in the figure. Figure 3 As shown in the diagram, vector hydrophones are deployed on the surface of the cylindrical shell with an adjacent spacing of no more than one-fifth of a wavelength. In this case, an adjacent spacing of 0.03λ, approximately 0.22m, is used. Sound pressure and normal vibration velocity information on the shell surface are obtained through discrete points within the formed cylindrical envelope. The specific selection of the monitoring point locations for the vector hydrophones on the cylindrical shell surface is as follows: Figure 4 As shown.

[0079] The second step involves using the sound pressure and normal velocity information obtained from the aforementioned vector hydrophone deployment method. The Helmholtz integral theorem for the sound pressure at the external field point is then applied to estimate the amplitude and phase of the radiated noise at a far-field location of 1500m in the horizontal plane. This estimation is consistent with the amplitude and phase of the far-field radiated noise calculated using the finite element method in all directions except for the 90° and 270° directions. Within a small range at the 90° and 270° directions, the phase prediction shows a slight difference of approximately 0.2 rad. Specifically... Figure 5 As shown. The sound pressure matrix B of the radiated noise at each far-field virtual reference point is obtained using this prediction method:

[0080]

[0081] in, Let be the sound pressure radiated by the underwater target structure at the i-th virtual reference point in the far field. This represents the position of the i-th virtual reference point, where i = 1, 2, ..., M, and M is the number of far-field virtual reference points.

[0082] Thirdly, since the arrangement of secondary sound sources has a significant impact on noise reduction, this invention utilizes the directional noise control characteristics of parametric secondary sound sources to propose a method for uniformly distributing collimated beams across the solid angle of space. Specifically, the parametric secondary sound sources are arranged linearly at equal intervals along the generatrix of the underwater cylindrical shell structure's side surface, and circularly with equal arcs along the end face, so that the collimation direction of the parametric secondary sound sources is distributed as uniformly as possible across the entire solid angle of space. The distribution methods are as follows: Figure 6 As shown in -(a) and 6-(b), the two spherical coordinate systems are established as follows: Figure 6 As shown in -(c).

[0083] Establish a spherical coordinate system with the center of the underwater cylindrical shell structure as the origin. The position of the parametric secondary sound source in this spherical coordinate system is: The position of the far-field virtual reference point in this spherical coordinate system is: Then, establish another spherical coordinate system using the equivalent sound center of the secondary sound source with structural surface parametric parameters. The position of the far-field virtual reference point in this spherical coordinate system is: The deflection angle of the collimation direction of the parametric secondary sound source is ψ θ , By separately measuring the elevation and azimuth angles in the collimation direction of the secondary sound source, the position coordinates of the far-field virtual reference point are first deflected by ψ in the elevation direction. θ Then deflect it in the azimuth direction. Then along By performing a coordinate translation, with point T being the placement location of the parametric secondary sound source, the [source can be...] The coordinates of the far-field virtual reference point in the coordinate system are transformed to... In the coordinate system, the coordinates of each far-field virtual reference point are transformed to... In the coordinate system, the specific details are as follows:

[0084]

[0085]

[0086]

[0087] in, All of these are intermediate variables, as detailed below:

[0088]

[0089]

[0090]

[0091] Will Substitute R P '、θ P '、 In the middle, obtain and The coordinate system transformation relationship, through which the coordinate system is transformed... The position coordinates of each virtual reference point in the coordinate system are all transformed to In the coordinate system, using p d The expression (r,ξ) calculates the corresponding complex sound pressure value, since p d The sound field corresponding to the expression (r,ξ) is a two-dimensional axisymmetric sound field, and its input parameters are r and ξ. Therefore, The coordinates in the coordinate system are transformed into the form (r, ξ), and the specific transformation relationship is as follows:

[0092] r ij =R P '

[0093]

[0094] Where, r ij Let ξ be the distance from the j-th parameter secondary sound source point T to the i-th far-field virtual reference point P. ij Let P be the angle between the i-th far-field virtual reference point and the collimation direction of the j-th parametric secondary sound source. Using the complex sound pressure levels of each parametric secondary sound source at each virtual reference point as elements of the multi-channel sound transfer matrix, we can obtain the multi-channel sound transfer matrix C:

[0095]

[0096] By weighted summing the contributions of each parametric secondary sound source at the far-field virtual reference point, the total sound pressure matrix A of the weighted emission of the parametric secondary sound sources can be obtained:

[0097]

[0098] Where, p d (r sumi ,ξ sumi ξ represents the total sound pressure contribution of each parametric secondary sound source to the i-th virtual reference point. sumi Let r represent the azimuth angle of the i-th reference point. sumi p represents the distance from the i-th virtual reference point to the origin of the spherical coordinate system. d (r sumi ,ξ sumi ) is the weighted sum of the i-th row in the multi-channel acoustic transmission matrix, i = 1, 2, ..., M, where M is the number of far-field virtual reference points.

[0099] Let X be the optimal emission weighting coefficient of the parametric secondary sound source, which is an N×1 constant matrix, as shown below:

[0100]

[0101] After adjusting the emission weighting coefficient X, the total sound pressure matrix A of the parametric secondary sound source weighted emission is obtained, which is the complex sound pressure value formed at each far-field virtual reference point, as shown in the following formula:

[0102]

[0103] The fourth step involves minimizing the cost function J(X) and then using the least squares method to solve for the emission weighting coefficients X of the parametric secondary sound sources deployed on the surface of the underwater cylindrical shell. The cost function J(X) is shown in the following equation:

[0104] J(X) = ||A+B|| 2 =||CX+B|| 2 =(CX+B) H (CX+B)

[0105] =X H C H CX+X H C H B+B H CX+B H B

[0106] Where X is the weighting coefficient matrix of the N×1 dimensional parametric secondary sound source, A is the total sound pressure matrix of the weighted emission of the parametric secondary sound source, B is the radiated sound pressure matrix composed of the sound pressure of the radiated noise at the far-field virtual reference point, C is the multi-channel sound transmission matrix, (·) H The conjugate transpose operation is represented by ||·||, which represents the 2-norm of a matrix-vector pair, i.e., the square root of the sum of the squares of the elements in the vector. To minimize J(X), the derivative of J(X) with respect to X must be zero, thus yielding the emission weighting coefficient matrix X of the parametric secondary sound source.

[0107] X = -C -1 B

[0108] in,(·) -1 To invert the matrix, the emission intensity of each parametric secondary sound source is adjusted by the weighting coefficient X of the emitted signal of the parametric secondary sound source, so that the sound pressure amplitude contribution of all parametric secondary sound sources at each virtual reference point in the far field is the same as that of the radiated noise sound pressure amplitude, and their phase contribution is opposite to that of the radiated noise sound pressure. Thus, the desired sound field generated by the parametric secondary sound sources after weighting adjustment can cancel the radiated noise sound field.

[0109] If the number of far-field virtual reference points exceeds the number of parametric secondary sound sources placed on the cylindrical shell surface, since C is a non-square matrix and has no inverse matrix, an ill-posed problem arises in the solution process for the emission weight coefficient matrix of the parametric secondary sound sources. When using the least squares method, Tikhonov regularization is required to transform the cost function J(X) into:

[0110] J(X) = ||A+B|| 2 +ε 2 ||X|| 2

[0111] =||CX+B|| 2 +ε 2 ||X|| 2

[0112] Where ε is the regularization parameter, the optimal parameter value ε is obtained by iterative weighting using the Bayesian criterion, and the derivative of J(X) with respect to X is set to 0, thus obtaining the weight coefficient matrix X:

[0113] X = -(C H C+ε 2 ·I N×N ) -1 ·C H ·B

[0114] Among them, I N×N It is an N-row, N-column identity matrix.

[0115] like Figure 7 As shown, by adjusting the parametric secondary sound sources on the surface of the cylindrical shell through the weighting coefficient matrix X, the obtained parametric secondary sound field is superimposed with the radiated sound field of the underwater target structure to obtain the controlled sound field. The noise control effect evaluation index J is used to evaluate this sound field. m =0.094%, Risk coefficient L m =1.6, noise decreases significantly in all directions on the horizontal plane, but there is a slight increase in energy in some directions.

[0116] Fifth, when monitoring information is missing, shutting down the parametric secondary sound sources in areas with inaccurate predictions on the surface of the underwater cylindrical shell can prevent noise spillover and increased exposure caused by controlling based on erroneous predictions. Utilizing the high directivity of the parametric secondary sound sources, a local control strategy can be employed to precisely and effectively control areas with accurate far-field radiated noise predictions. This not only reduces the average noise level across the entire space but also enhances the underwater vehicle's flexibility and maneuverability.

[0117] When it is impossible to deploy vector hydrophones on both ends of an underwater cylindrical shell structure for end-face area monitoring, vector hydrophones are deployed only on the side surface of the cylindrical shell. The prediction effect of far-field radiated noise is as follows: Figure 8As shown, the prediction accuracy drops significantly near the 90° and 270° directions, with a sound pressure level error of approximately 10 dB and a phase error of 1.3 rad. A global control strategy is employed based on the prediction results. However, in the two directions where the predictions are inaccurate (90° and 270°), the global control strategy causes noise overflow, as shown in the specific control effect. Figure 9 As shown. Its noise control effect evaluation index J m =1.6%, risk coefficient L m =3.25. When using a local control strategy to control the accurately predicted regions of -60° to 60° and 120° to 240°, the control effect is as follows: Figure 10 As shown. Noise control effectiveness evaluation index J m =1.51%, Risk coefficient L m =1.25. (Comparison) Figure 9 and Figure 10 It can be seen that when monitoring information is missing, the local control strategy is not only better than the global control strategy in terms of noise reduction, but also does not produce energy overflow in the direction of inaccurate prediction, and only produces a small amount of energy overflow in the area of ​​accurate prediction.

[0118] Under conditions of missing monitoring information, comparison Figure 10 and Figure 11 Therefore, by replacing the parametric secondary sound source with a traditional point sound source and adopting a local control strategy, the noise control effect evaluation index J is [value missing]. m =1.37%, risk coefficient L m =45.84. It is not difficult to see that although the residual noise effect is slightly higher than that of the parametric secondary sound source by 0.14%, the risk coefficient is about 37 times that of the parametric secondary sound source for local noise control, with a risk coefficient of about 45.84. This makes the underwater vehicle's exposure increase sharply, and noise control becomes meaningless.

[0119] Therefore, using a parametric sound source with strong directional characteristics as a secondary sound source for active noise control can not only reduce the overlap area of ​​secondary beams between channels and improve the stability of the system, but also achieve good noise reduction when monitoring information is missing, and ensure that the exposure of the underwater vehicle is not too large after noise control.

Claims

1. A method for controlling underwater structure noise based on parametric secondary sound sources, characterized in that, Includes the following steps: (1.1) Deploy multiple vector hydrophones on the surface of the underwater structure to obtain the sound pressure and normal vibration velocity information of the underwater structure radiated noise. The distance between adjacent vector hydrophones is no more than one-fifth of the wavelength of the sound wave in water. (1.2) Using the Helmholtz integral prediction method, based on the sound pressure and normal velocity information obtained by each vector hydrophone on the surface of the underwater structure, the sound pressure amplitude and phase at each virtual reference point in the far field are calculated, and the radiation sound pressure matrix composed of the complex sound pressure values ​​of the radiated noise at the virtual reference point in the far field is obtained. ; (1.3) The parametric secondary sound sources are evenly and uniformly distributed on the surface of the underwater structure at equal intervals, ensuring that their collimation direction is uniformly distributed across the entire solid angle range; the radiated sound field of the parametric array sound sources is derived according to the wave equation to obtain the radiated sound field of a single parametric secondary sound source; the radiated sound field generated by each parametric secondary sound source is transformed to the same coordinate system through coordinate transformation to obtain the complex sound pressure value of each parametric secondary sound source at the spatial position of each virtual reference point in the far field; all the obtained complex sound pressure values ​​are used as one element of the multi-channel sound transfer matrix to obtain the multi-channel sound transfer matrix. ; (1.4) The number of far-field virtual reference points selected should not be less than the number of parametric secondary sound sources selected. When the number of far-field virtual reference points is the same as the number of parametric secondary sound sources selected, the emission weight coefficients of the parametric secondary sound sources can be solved by the least squares method with the defined cost function minimized. When the number of far-field virtual reference points exceeds the number of parametric secondary sound sources, it may cause poor stability of the noise control system and ill-posed problems in the matrix solution process. In this case, the weighting coefficients of the transmitted signals of the parametric secondary sound sources need to be obtained by using the Tikhonov regularized least squares method. ; (1.5) When monitoring information is missing, the parametric secondary sound source in the area where the monitoring information is inaccurate should be shut down. Taking advantage of the high directivity of the parametric secondary sound source, a local control strategy should be adopted to carry out precise and effective noise control in the local area where the monitoring information is accurate. Without generating noise leakage, the noise control effect in the entire space range can be obtained.

2. The underwater structure noise control method based on a parametric secondary sound source according to claim 1, characterized in that, When the number of far-field virtual reference points is less than the number of parametric secondary sound sources, that is... At that time, for the multi-channel acoustic transfer matrix Elementary matrix transformations are used to simplify the expression containing indivual The system of equations of the elementary equations has infinitely many solutions, making it impossible to obtain the emission weight coefficients of the parametric secondary sound sources. Therefore, the number of far-field virtual reference points should not be less than the number of parametric secondary sound sources. When the number of far-field virtual reference points is chosen to be equal to the number of parametric secondary sound sources, that is... At that time, the constructed multi-channel sound transmission matrix for OK Given a square matrix, solving the simplified system of equations derived from this matrix yields a unique solution. Therefore, in... At that time, the emission weighting coefficients of the parametric secondary sound source can be determined; At this point, through the defined cost function The noise control effect is described, and the least squares method is used to obtain the cost function. Minimal parametric secondary sound source emission weighting coefficient matrix Cost function As shown in the following formula: Wherein, cost function yes The function, The radiated sound field matrix is ​​composed of the complex sound pressure levels of the radiated noise at the far-field virtual reference point. This represents the multi-channel sound transfer matrix of each parametric secondary sound source at each virtual reference point. This represents the conjugate transpose operation. The 2-norm of a matrix-vector structure is represented by the square root of the sum of the squares of the elements in the vector; it makes the sum of the squares of the elements in the vector equal to the 2-norm of the vector structure. Reaching the minimum value yields the optimal parametric secondary sound source emission weighting coefficient matrix. : in, The matrix inversion operation is performed, and the emission intensity of each parametric secondary sound source is adjusted by the emission signal weight coefficients so that the parametric secondary sound source array radiates the desired secondary sound field. When the number of far-field virtual reference points is greater than the number of parametric secondary sound sources, that is... At that time, the constructed multi-channel acoustic transmission matrix The system of equations simplified using elementary matrix transformations is as follows: indivual The system of equations is overdetermined, so it has no solution. That is, there is no parametric secondary source emission weighting coefficient that makes the sum of the complex sound pressure levels at each virtual reference point in the far field zero. Therefore, by defining the residual function of the initial superimposed sound field at the virtual reference point in the far field, and using the square of the residual function as the cost function... Calculate the cost function The smallest set of parametric secondary sound source emission weighting coefficient matrices ; At this point, when using the least squares method, Tikhonov regularization is required to adjust the cost function. The change is as follows: in, For the regularization parameters, the Bayesian criterion is used for iterative weighting to obtain the optimal parameter values, so that... By reaching a minimum value, the optimal parametric secondary sound source emission weighting coefficient matrix can be obtained. : in, for OK The identity matrix of the column is used to adjust the emission intensity of each parametric secondary sound source by obtaining the emission signal weight coefficients of the parametric secondary sound source array, so that the parametric secondary sound source array radiates the desired secondary sound field.

3. The underwater structure noise control method based on a parametric secondary sound source according to claim 1, characterized in that, When underwater structure monitoring information is missing, it is necessary to turn on the parametric secondary sound sources deployed in the areas that can be accurately predicted, and turn off the parametric secondary sound sources in the directions that are not accurately predicted. Only the areas with accurate prediction should be precisely controlled locally to produce a good local spatial noise control effect and reduce the average noise level in the entire space.

Images