A method for long-distance measurement and evaluation of target radiated noise

Abstract

The invention discloses a long-distance measurement and evaluation method for target radiation noise. The method comprises the following steps: collecting target radiation noise signals, performing sample feature enhancement preprocessing, obtaining a target radiation noise signal data set, and dividing a short-range domain data set and a long-range domain data set according to distance; constructing a short-range domain and a long-range domain data propagation model, injecting environmental parameters into the short-range domain and the long-range domain data propagation model for correction and optimization, and establishing a correlation relationship between short-range domain and long-range domain sample data; compensating for on-site acoustic propagation loss of target radiation noise signals through the short-range domain data set, the long-range domain data set and the short-range domain and the long-range domain data propagation model, and obtaining target radiation noise feature information; constructing a long-distance measurement and evaluation model, judging whether the target radiation noise level and line spectrum noise characteristics meet the target values ​​required for actual use, and if they are lower than the target values, performing noise control for the target, and providing basic data support for the target acoustic stealth design.

Description

Technical Field

[0001] The present invention relates to the technical field of noise analysis, and in particular to a method for long-distance measurement and evaluation of target radiation noise. Background Art

[0002] The biggest advantage of a target is concealment, and radiated noise directly affects or even determines the concealment of a target. Usually, when a target is designed, built, and put into use, it will be fully designed for noise reduction and tested and evaluated. However, we are more concerned about the acoustic stealth performance in the actual use environment: on the one hand, the traditional radiated noise characteristic evaluation method is often carried out in a specific environment, and most of the measurement and evaluation results are obtained at a relatively close distance; on the other hand, the stealth performance in the actual environment is also related to the specific marine environmental conditions.

[0003] In view of this, it is necessary to establish a long-distance measurement and evaluation method of target radiation noise for actual use environment. Summary of the invention

[0004] In order to solve the deficiencies in the prior art, the purpose of the present invention is to provide a long-distance measurement and evaluation method for target radiated noise, which obtains the radiated noise information of the long-distance target by carrying out target characteristic measurement at a longer distance. The radiated noise information is closer to the acoustic signal characteristics detected in the far field of the actual environment, and can more accurately evaluate the acoustic characteristics of the target.

[0005] To achieve the purpose of the present invention, the technical solution adopted by the present invention is:

[0006] A method for long-distance measurement and evaluation of target radiation noise comprises the following steps:

[0007] (1) Collect the target radiation noise signal, perform sample extraction and sample feature enhancement preprocessing to obtain the target radiation noise signal dataset, and divide it into a short-range domain dataset and a long-range domain dataset according to the distance;

[0008] (2) Construct the data propagation model of the local domain and the remote domain, inject the environmental parameters measured on site into the data propagation model of the local domain and the remote domain, modify and optimize the parameters of the data propagation model, and establish the association relationship between the sample data of the local domain and the remote domain;

[0009] (3) Compensate for the on-site acoustic propagation loss of the target radiation noise signal through the near-domain dataset, the far-domain dataset, and the near-domain and far-domain data propagation models, obtain the target radiation noise feature information, and construct a target feature long-distance database;

[0010] (4) Construct a long-distance measurement evaluation model to determine whether the target radiation noise level and line spectrum noise characteristics meet the target values ​​required for actual use. If they are lower than the target values, noise control is required and basic data support is provided for the target acoustic stealth design.

[0011] Furthermore, in step (1), the test distance of 1 km is used as the dividing line, within 1 km is the short-range domain, and greater than 1 km is the long-range domain.

[0012] Furthermore, in step (2), the environmental parameters measured on-site include sound velocity profile and seabed acoustic parameters.

[0013] Furthermore, in step (2), the data propagation models of the near-domain and long-domain are constructed based on the beam spectrum integration method. The wavenumber spectrum integration includes integral transformation, global matrix method and integral inverse transformation. The global matrix method adopts the finite element solution of the depth equation.

[0014] Furthermore, in step (2), the model parameters are corrected and optimized by inverting the sound velocity profile and seabed acoustic parameters measured in the field, combining the near-domain and far-domain data propagation models, the near-domain and far-domain data sets, and the broadband and single-frequency sound propagation characteristics analysis.

[0015] Furthermore, for broadband signals, the broadband acoustic signal received at the spatial point (r, z) is denoted as x(t), 0<t<T 1 , the broadband sound receiving signal energy level and sound propagation loss SL are:

[0016]

[0017] in, It is the result of filtering the broadband sound receiving signal x(t) through a narrowband filter with a center frequency of f and a bandwidth of Δf.

[0018] Furthermore, for a single-frequency signal, i.e., a long pulse or continuously transmitted signal, the power level is also adopted. The energy level of the single-frequency acoustic reception signal and the acoustic propagation loss SL are respectively:

[0019]

[0020] in, is the time-segment average signal, and TL is the environmental parameter of the measurement site, which is calculated from the data propagation model.

[0021] Furthermore, in step (3), the target feature long-distance database includes the target radiation noise level, spectral features, and line spectrum features; the spectral features and line spectrum features of the signals of the short-range domain dataset and the long-range domain dataset are compared, and the line spectrum information unique to the long distance is extracted and stored in the target feature long-distance database.

[0022] Furthermore, in step (4), the long-distance measurement evaluation model is constructed based on the passive sonar equation, and the passive sonar equation is in the form of:

[0023] DT=SL-TL+GS+GT-NL

[0024] Among them, TL is the sound propagation loss in the actual environment, SL is the target radiation noise level, NL is the background noise level, DT is the detection threshold, GT is the temporal processing gain and GS is the spatial processing gain.

[0025] Furthermore, step (4) also includes evaluating the distance range and detection probability detected by the horizontal hydrophone array measurement system based on the target characteristic parameters obtained by long-distance measurement, combined with known horizontal hydrophone array measurement system parameters and test sea area environmental parameters.

[0026] The beneficial effect of the present invention is that, compared with the prior art,

[0027] Aiming at the problem of difficulty in acquiring the correlation relationship between sample data in the short-range domain and the long-range domain, the present invention conducts in-depth customization and optimization design of the propagation model, solves the problem of low accuracy of the existing propagation model with the help of multiple groups of field test data, and obtains a high-performance feature space transformation mechanism by relying on the self-evaluation capability of the system, thereby realizing the target acoustic stealth effectiveness evaluation based on long-range measurement of noise.

[0028] The present invention utilizes environmental parameters measured on-site to optimize the propagation model of the target environment, improves the performance of the acoustic propagation model of the target domain scene, and through in-depth mining of radiation noise samples of the measured target under different working conditions, clarifies the intrinsic connection between samples and ultimately realizes the transmission network in the target environment.

[0029] The present invention extracts samples and uses a filtering feature selection method to perform feature enhancement preprocessing of sample data, and then uses a constructed acoustic transfer network to perform parameter association and comparison, so that the extracted features have target-independent feature boundaries.

[0030] In an actual environment, the long-distance measurement evaluation method of the present invention can evaluate the noise characteristics of a target based on its independent characteristic parameters, so that the evaluation result is closer to the actual situation.

[0031] The experimental results of the method of the present invention measured on the lake prove that the proposed long-distance measurement and evaluation method of target radiation noise can establish the correlation between the short-range and long-range domains of target radiation noise, effectively improving the evaluation effect of target acoustic stealth. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the target radiation noise long-distance measurement and evaluation method of the present invention;

[0033] Figure 2 This is a schematic diagram of underwater noise long-distance testing;

[0034] Figure 3 It is a sample feature enhancement preprocessing flow chart;

[0035] Figure 4 This is the result diagram of the measured target radiation noise analysis;

[0036] Figure 5 This is the target radiation noise long-distance measurement compensation analysis result diagram;

[0037] Figure 6 It is a schematic diagram of the signal-to-interference ratio probability density function and the predicted detection probability at different distances. DETAILED DESCRIPTION

[0038] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of this application.

[0039] Aiming at the problem of difficulty in acquiring the correlation relationship between sample data in the short-range domain and the long-range domain, the present invention conducts in-depth customization and optimization design of the propagation model, solves the problem of low accuracy of the existing propagation model with the help of multiple groups of field test data, and obtains a high-performance feature space transformation mechanism by relying on the self-evaluation capability of the system, thereby realizing the target acoustic stealth effectiveness evaluation based on long-range measurement of noise.

[0040] like Figure 1 As shown, the target radiation noise long-distance measurement and evaluation method described in the present invention includes the following steps:.

[0041] The first step is to collect the target radiation noise signal, perform sample extraction and sample feature enhancement preprocessing to obtain the target radiation noise signal dataset, and divide it into a short-range domain dataset and a long-range domain dataset according to the distance;

[0042] like Figure 2 As shown, the near and far domains of the embodiment of the present invention are divided by the distance of the horizontal hydrophone array measurement system from the tested target at 1 km, within 1 km is the near domain, and greater than 1 km is the far domain. Based on this distance, multiple different distances are selected in the near domain and the far domain to collect target radiation noise signals, and multiple groups of samples are collected at each distance.

[0043] like Figure 3As shown, samples are extracted from the collected target radiation noise signal. The principle of sample extraction is to group the channel test data with the same element interval, and perform feature enhancement preprocessing on the grouped samples; the sample feature enhancement preprocessing includes filtering and denoising by combining bandpass filtering and compensation filters; the samples after filtering and denoising are subjected to frequency domain beamforming and frequency band equalization to obtain the target radiation noise signal after removing the interference data.

[0044] The second step is to build a data propagation model for the local and remote domains, inject the environmental parameters measured on site into the data propagation model for the local and remote domains, modify and optimize the parameters of the data propagation model, and establish the association relationship between the sample data of the local and remote domains;

[0045] The environmental parameters measured on site, including the sound velocity profile (spatial non-uniformity and time variation), seabed acoustic parameters (seabed sound velocity and attenuation coefficient), etc., can be collected and acquired through special equipment.

[0046] For the sound velocity profile data, a temperature-salinity-depth instrument can be used to directly measure and obtain it at the test site, or historical measurement data can be used.

[0047] For the seabed acoustic parameters, acoustic field measurement equipment can be used to measure and obtain them at the test site or use historical measurement data. The required seabed acoustic parameters, including seabed sound speed and attenuation coefficient, are inverted using the hydrological environment and acoustic environment characteristic measurement data obtained on site. The seabed acoustic characteristics are mainly obtained by inverting the active sound source signal received by the vertical array of the buoy, and the sound propagation is verified based on the established seabed acoustic model. In shallow waters, medium and low frequency active sound sources are used, and acoustic methods such as the spatial distribution characteristics of the sound field and propagation loss are used to locally invert the seabed geoacoustic parameters.

[0048] The cost function for obtaining the seafloor attenuation coefficient using propagation loss is:

[0049]

[0050] in, To numerically calculate the propagation loss at different distances, is the propagation loss obtained from the experiment. The propagation calculated numerically is the average of the single-frequency calculation results over the distance, and the propagation loss obtained from the experimental data is the average of the 1 / 3 octave frequency.

[0051] For the data propagation models in the short-range domain and the long-range domain, the data propagation models in the short-range domain and the long-range domain are constructed based on the beam spectrum integration method. The wavenumber spectrum integration includes integral transformation, global matrix method and integral inverse transformation.

[0052] Among them, the direct global matrix method uses the finite element solution of the depth equation. The specific steps are:

[0053] First, consider each layer of the hierarchical structure as a finite element, and the magnitude of the homogeneous solution to be determined in the mth layer is and as well as and Combined into a local degree of freedom vector have:

[0054]

[0055] The kernel function vector v that defines the sound field parameters m (k r ,z):

[0056]

[0057] Where ω and u are displacements, and σ zz and σ rz are normal stress and tangential stress respectively; the matrix relationship of the homogeneous part in the mth layer is obtained:

[0058] v m (k r ,z)=c m (k r ,z)a m (k r )

[0059] in,

[0060] c m (k r ,z)=d m (k r ) m (k r ,z)

[0061] Where, d m (k r ) is a depth-independent matrix; and e m (k r ,z) is a diagonal matrix containing exponential terms.

[0062] Then, the kernel function of the active sound field is Kernel function v superimposed on the homogeneous solution m (k r ,z). According to the continuity condition of the sound field parameters on the interface between the m layer and the m+1 layer, we have:

[0063]

[0064] It can be seen that the discontinuity of the active sound field and the discontinuity in the homogeneous solution can cancel each other out. The left side of the above equation is defined as the interface discontinuity vector v m(k r ),have Similarly, the active sound field discontinuity vector can be introduced

[0065] Finally, the mapping relationship between the local vector and the global system is established. The global degree of freedom vector of the uplink and downlink wave field amplitudes is defined, and there is a mapping relationship:

[0066] a m (k r )=S m A(k r ),m=1,2,...,N

[0067] Substituting in:

[0068]

[0069] Next, v m (k r ) into a global discontinuity vector V(k r ), there are:

[0070]

[0071] Formation:

[0072]

[0073] Similarly, the discontinuity vector of the global active sound field can be obtained:

[0074]

[0075] Therefore, the global cancellation should satisfy the linear equation system, And C(k r ) is the global coefficient matrix:

[0076]

[0077] The topological matrix S in the above formula m and T m is a very sparse matrix containing only 0s and 1s.

[0078] The model parameters are corrected and optimized by monitoring the sound velocity profile at the test site, inverting the seabed acoustic parameters, combining the near-domain and far-domain data propagation models, near-domain and far-domain data sets, and broadband and single-frequency sound propagation characteristics analysis.

[0079] According to the target noise spectrum characteristics, it can be divided into broadband (spectrum) and single frequency (line spectrum).

[0080] For broadband signals, the broadband acoustic signal received at the spatial point (r, z) is denoted as x(t), 0<t<T 1 , which is considered to have been converted to hydrophone sensitivity, in μPa. The broadband acoustic reception signal energy level and acoustic propagation loss SL are:

[0081]

[0082] in, It is the result of filtering the broadband sound receiving signal x(t) through a narrowband filter with a center frequency of f and a bandwidth of Δf.

[0083] For single-frequency signals, i.e. long pulses or continuous transmission signals, the power level is also used. The energy level of the single-frequency sound receiving signal and the sound propagation loss SL are:

[0084]

[0085] in, is the time-segment average signal, and TL is the environmental parameter of the measurement site, which is calculated from the data propagation model.

[0086] The third step is to compensate for the on-site acoustic propagation loss of the target radiation noise signal through the short-range domain dataset, long-range domain dataset and the short-range and long-range domain data propagation model, obtain the target radiation noise characteristic information, and construct the target characteristic long-range database;

[0087] The target characteristic long-distance database includes the target radiation noise level, spectrum characteristics, line spectrum characteristics (line spectrum frequency and its time stability, line spectrum height relative to continuous spectrum, etc.).

[0088] The spectral characteristics and line spectrum characteristics of the signals in the short-range domain dataset and the long-range domain dataset are compared, and the line spectrum information unique to the long-range is extracted and stored in the target feature long-range database.

[0089] The fourth step is to construct a long-distance measurement and evaluation model to determine whether the target radiation noise level and line spectrum noise characteristics meet the target values ​​required for actual use. If they are lower than the target values, noise control is required and basic data support is provided for the target acoustic stealth design.

[0090] The measured results are as follows Figure 4 and Figure 5 As shown, the results of long-distance measurement can reflect the actual acoustic characteristics of the target.

[0091] The long-range measurement evaluation model is constructed based on the passive equation. The traditional passive sonar equation is in the form of:

[0092] DT=SL-TL+GS+GT-NL

[0093] Among them, TL is the sound propagation loss in the actual environment (uncertainty ΔTL), SL is the target radiation noise level, and NL is the background noise level. The detection threshold DT, the time processing gain GT, and the spatial processing gain GS, etc., for a given sonar system, are considered to be some known parameters. These parameters are all in decibel values. The sonar quality factor is:

[0094] FOM=SL-DT+GS+GT-NL

[0095] It reflects the basic detection capability of a given sonar system.

[0096] Based on the target characteristic parameters obtained from long-distance measurements, combined with the knowledge of known horizontal hydrophone array measurement system parameters and test sea area environmental parameters, the distance range and detection probability (detection probability) detected by the horizontal hydrophone array measurement system are evaluated. From the perspective of target detection, the target characteristics that can be detected at a long distance are captured and measured, which is closer to the actual situation.

[0097] The distance where the "propagation loss TL and the quality factor FOM" or the "signal-to-interference ratio SIR and the detection threshold DT" are equal is the sonar operating range. The signal-to-interference ratio is defined as:

[0098] SIR=SL-TL+GS+GT-NL

[0099] At a given distance, the more the SIR of the signal received by the sonar system exceeds DT, the greater the probability of the target being detected. Considering the influence of the uncertainty in the acquisition and evaluation of each parameter in the sonar equation, when evaluating the effective distance, the probability density function description of the uncertainty change of the sonar equation parameters is introduced to predict the detection probability at different distances, such as Figure 6 As shown. The probability density function distribution at each distance is predicted, and then the probability integral is performed on the part greater than the detection threshold to obtain the predicted detection probability at that distance. The predicted probability of detection (PPD) that varies with distance has taken into account the uncertainty effects of target characteristics and propagation loss fluctuations.

[0100] According to the ocean acoustic environment parameters of the on-site sea area, the distribution of the sound propagation loss TL and its uncertainty in the sea area can be obtained, and the probability density function of the signal-to-interference ratio SIR at different distances can be obtained (in terms of P SIR (x) indicates), we can also get its expected value SIR 0 (r). The detection probability at the distance is calculated based on the probability distribution at different distances (the probability that the signal-to-interference ratio SIR is greater than or equal to the detection threshold DT), that is:

[0101]

[0102] Assume PSIR (x) is the mean value μ 0 (r,z r ,z s )≡SIR 0 (r), variance σ 0 (r,z r ,z s ), we have:

[0103]

[0104] Note that the signal margin is defined as:

[0105] SE=FOM-TL=SIR-DT;FOM-TL 0 =SIR 0 -DT

[0106] The above formula can also be written as:

[0107]

[0108] In SIR 0 (r)=DT or FOM=TL 0 At a distance of (r), the calculated detection probability is 0.5, which is traditionally considered to correspond to the sonar operating range R.

[0109] Using these formulas, calculate (0~R max ) or (R min ~R max ) The corresponding detection probability P at each distance within the distance range D (r), and the expected sonar range estimate is obtained through probability weighted integration:

[0110]

[0111]

[0112] The experimental results of the method of the present invention measured on the lake prove that the proposed long-distance measurement and evaluation method of target radiation noise can establish the correlation between the short-range and long-range domains of target radiation noise, effectively improving the evaluation effect of target acoustic stealth.

[0113] The applicant of the present invention has made a detailed explanation and description of the implementation examples of the present invention in conjunction with the drawings in the specification. However, those skilled in the art should understand that the above implementation examples are only preferred implementation schemes of the present invention, and the detailed description is only to help readers better understand the spirit of the present invention, but not to limit the scope of protection of the present invention. On the contrary, any improvements or modifications based on the inventive spirit of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for long-distance measurement and evaluation of target radiation noise, characterized in that: Includes steps: (1) Collect the target radiation noise signal, perform sample extraction and sample feature enhancement preprocessing to obtain the target radiation noise signal data set, and divide it into a short-range domain data set and a long-range domain data set according to the distance: the test distance of 1 km is used as the dividing line, within 1 km is the short-range domain, and greater than 1 km is the long-range domain; (2) Construct the data propagation model of the local and remote domains, inject the environmental parameters measured on site into the data propagation model of the local and remote domains, modify and optimize the parameters of the data propagation model, and establish the association relationship between the sample data of the local and remote domains; (3) By using the near-domain data set, the far-domain data set, and the near-domain and far-domain data propagation models, the on-site acoustic propagation loss of the target radiation noise signal is compensated to obtain the target radiation noise characteristic information and construct a target characteristic long-distance database; (4) Construct a long-distance measurement evaluation model to determine whether the target radiation noise level and line spectrum noise characteristics meet the target values ​​required for actual use. If they are lower than the target values, noise control is required and basic data support is provided for the target acoustic stealth design.

2. The target radiation noise long-distance measurement and evaluation method according to claim 1 is characterized in that: In step (2), the environmental parameters measured on site include sound velocity profile and seabed acoustic parameters.

3. The target radiation noise long-distance measurement and evaluation method according to claim 1, characterized in that: In step (2), the data propagation models of the near-domain and long-domain are constructed based on the beam spectrum integration method. The wavenumber spectrum integration includes integral transformation, global matrix method and integral inverse transformation. The global matrix method adopts the finite element solution of the depth equation.

4. The target radiation noise long-distance measurement and evaluation method according to claim 1, characterized in that: In step (2), the model parameters are corrected and optimized by inverting the sound velocity profile and seabed acoustic parameters measured in the field, combining the data propagation model of the near-range and long-range domains, the near-range and long-range domain data sets, and the analysis of broadband and single-frequency sound propagation characteristics.

5. The target radiation noise long-distance measurement and evaluation method according to claim 4, characterized in that: For broadband signals, note the spatial points (r,z) The broadband acoustic signal received at x(t), 0<t<T 1 , broadband acoustic receiving signal energy level E x ( f ) and the acoustic propagation loss SL are: SL=TL+10lg E x ( f )dB in, is the broadband acoustic receiving signal x(t) passing through the center frequency f , bandwidth is Δ f The result after filtering with a narrowband filter.

6. The target radiation noise long-distance measurement and evaluation method according to claim 4, characterized in that: For single-frequency signals, that is, long pulse or continuous transmission signals, the power level is also used. The energy level of the single-frequency sound receiving signal P x The sound propagation losses SL are: SL=TL+10lg P x dB Among them, record the space point (r,z) The broadband acoustic signal received at x(t), 0<t<T 1 , Time segment ΔT The average signal, TL is the environmental parameter of the measurement site, calculated from the data propagation model.

7. The target radiation noise long-distance measurement and evaluation method according to claim 1, characterized in that: In step (3), the target feature long-distance database includes the target radiation noise level, spectrum features, and line spectrum features; the spectrum features and line spectrum features of the signals of the short-range domain dataset and the long-range domain dataset are compared, and the line spectrum information unique to the long-range is extracted and stored in the target feature long-distance database.

8. The target radiation noise long-distance measurement and evaluation method according to claim 1, characterized in that: In step (4), the long-range measurement evaluation model is constructed based on the passive sonar equation, which is in the form of: DT=SL-TL+GS+GT-NL Among them, TL is the sound propagation loss in the actual environment, SL is the target radiation noise level, NL is the background noise level, DT is the detection threshold, GT is the temporal processing gain and GS is the spatial processing gain.

9. The target radiation noise long-distance measurement and evaluation method according to claim 1, characterized in that: Step (4) also includes evaluating the distance range and detection probability detected by the horizontal hydrophone array measurement system based on the target characteristic parameters obtained by long-distance measurement, combined with known horizontal hydrophone array measurement system parameters and test sea area environmental parameters.

Images