Simulation test method for sonic boom signal of supersonic civil aircraft

Abstract

The invention provides a supersonic civil aircraft sonic boom signal simulation test method, which belongs to the field of sonic boom simulation test, and comprises the following steps: S1, obtaining an expected sonic boom signal, generating a simulated sonic boom signal of the expected sonic boom signal, and calculating to obtain an equilibrium parameter of the simulated sonic boom signal; s2, taking the expected acoustic explosion signal as an input signal and an expected signal, generating an actual acoustic explosion signal through a sound production device, and calculating to obtain a filter coefficient vector; s3, taking the expected sonic boom signal as a reference signal, and calculating to obtain an optimal filter coefficient vector according to the filter coefficient vector; and S4, according to the expected acoustic explosion signal, the equalization parameter and the optimal filter coefficient vector, calculating to obtain accurate input data, inputting the accurate input data to a sound production device, and outputting an actual acoustic explosion signal conforming to the expected acoustic explosion signal. According to the invention, the problem of inaccurate acoustic explosion simulation caused by amplitude-phase errors introduced by an electroacoustic system in the existing acoustic explosion signal simulation technology is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of sound blast simulation test, and particularly relates to a supersonic civil aircraft sound blast signal simulation test method. BACKGROUND

[0002] Supersonic civil aircrafts will produce sound blasts during cruising flight, and the sound blast is a kind of noise, which mainly appears near the lower part of the flight track. People will have negative feelings such as terror and restlessness when hearing the sound blast. Before the application of supersonic civil aircrafts, the subjective feelings and acceptance of people to different types of sound blasts need to be studied.

[0003] The sound blast subjective response evaluation test is a test method for studying the subjective feelings of people to sound blasts. The precondition for carrying out the above test is to accurately restore the sound blast signal of the supersonic civil aircraft for the test personnel to listen to. The sound blast is a kind of noise phenomenon with high nonlinearity. When the sound blast signal is restored by using the loudspeaker and other electro-acoustic systems, the amplitude and phase errors of the system itself will cause errors in the simulation of the sound blast signal, resulting in a large difference between the restored sound blast signal waveform and the expected value, and further unable to obtain the correct result of the subjective evaluation of people to the sound blast.

[0004] The existing sound blast signal simulation usually generates an electric signal. In the electro-acoustic system, the amplitude and phase errors will be introduced in the conversion process of the generation and transmission of the sound blast signal between the amplifier, transducing element, transmission line and other electro-acoustic systems. The amplitude and phase errors are caused by the frequency response imbalance and impedance mismatch of the components, and are dynamically fluctuated with the signal frequency and environmental temperature, which will directly cause the reproduced sound blast signal to deviate from the true sound blast, and the simulation is inaccurate. SUMMARY

[0005] In view of the above problems in the prior art, the supersonic civil aircraft sound blast signal simulation test method provided by the application solves the problem of inaccurate sound blast simulation caused by the amplitude and phase errors introduced by the electro-acoustic system in the existing sound blast signal simulation technology.

[0006] In order to achieve the above purpose, the technical scheme adopted by the application is as follows: a supersonic civil aircraft sound blast signal simulation test method, comprising the following steps: S1, obtaining an expected sound blast signal, generating a simulated sound blast signal of the expected sound blast signal, and calculating the equalization parameters of the simulated sound blast signal; S2, taking the expected sound blast signal as an input signal and an expected signal, generating an actual sound blast signal through a sound generating device, and calculating a filter coefficient vector; S3, taking the expected sound blast signal as a reference signal, and calculating an optimal filter coefficient vector according to the filter coefficient vector; S4, according to the desired sound blast signal, the equalization parameter and the optimal filter coefficient vector, the accurate input data is calculated and input into the sound emitting device, and the actual sound blast signal conforming to the desired sound blast signal is output.

[0007] Further, the expression of the equalization parameter is as follows:

[0008]

[0009]

[0010] wherein, is the equalization parameter, is the channel response of the sound emitting device, is the delay factor, is the sampling period of the delay, is the analog sound blast signal, is the first regularization factor, is the Fourier transform of the desired sound blast signal, is the Fourier transform, is the desired sound blast signal.

[0011] Further, the specific steps of S2 include: S201, setting an initial filter coefficient; S202, taking the desired sound blast signal as an input signal, passing through the filter of the sound emitting device to obtain an actual sound blast signal; S203, taking the desired sound blast signal as a reference signal, and calculating the reference signal and the actual sound blast signal to obtain an error signal; S204, based on the adaptive filtering algorithm, setting an adaptive convergence coefficient, and taking the minimum mean square value of the error signal as the target, calculating the filter coefficient vector.

[0012] Further, the expression of the filter coefficient vector is as follows:

[0013]

[0014]

[0015]

[0016]

[0017] wherein, is the filter coefficient vector of the nth iteration, is the filter coefficient vector of the (n-1)th iteration, is the filter coefficient vector of the (n+1)th iteration. filter coefficient vector of the next iteration, is an adaptive convergence coefficient, is an error signal, is an input signal, is a reference signal, is an actual sound blast signal, is an order of the filter, is a filter coefficient of the first order, is a filter coefficient of the first order, is a filter coefficient of the first order, is a filter coefficient of the first order, is an input signal of the first order, is an input signal of the first order, is a transpose symbol.

[0018] Further, the specific steps of S3 include: S301, taking the expected sound blast signal as a reference signal, and performing convolution operation with the filter coefficient vector to obtain a filtered signal; S302, constructing a filtered signal vector based on the filtered signal; S303, calculating the autocorrelation matrix and the cross-correlation variable through time average according to the filtered signal vector; S304, calculating the optimal filter coefficient through Wiener filtering according to the autocorrelation matrix and .

[0019] Further, the expression of the optimal filter coefficient is as follows:

[0020]

[0021]

[0022]

[0023] wherein, is the optimal filter coefficient, is the autocorrelation matrix, is the cross-correlation variable, is a second regularization factor, is a filter primary channel, is a time average mean square error of the expected sound blast signal, is a time average of the argument, is a filter coefficient in the iteration process, is a reference signal, is a filtered signal, is a transpose symbol.

[0024] Further: the expression of the accurate input data is as follows:

[0025] Wherein, is the accurate input data, is the equalization parameter, is the expected sound blast signal, is the optimal filter coefficient.

[0026] The beneficial effects of the present application: the present application solves the problem of signal distortion caused by the amplitude and phase error of the electro-acoustic system in the sound blast signal simulation, improves the fidelity and reliability of the sound blast simulation signal; corrects the amplitude and phase distortion in the signal transmission process, so that the simulated sound blast signal is close to the expected sound blast signal, and the accurate sound blast signal required by the technical personnel is reproduced; at the same time, the present application does not need to rely on specific hardware optimization, and through the adaptive adjustment ability of the algorithm, it can flexibly adapt to different sound blast signal simulation requirements, improves the stability of solving through regularization optimization, avoids the problems of signal extreme distortion and calculation divergence, improves the precision and consistency of sound blast signal generation, and provides high-precision and high-reliability sound blast signal generation guarantee for sound blast related research and test analysis. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 It is a supersonic civil aircraft sound blast signal simulation test method; Figure 2 It is a single-point sound blast signal simulation principle diagram; Figure 3 It is a channel identification adaptive filtering principle diagram; Figure 4 It is a transverse FIR filter structure diagram; Figure 5 It is a channel equalization filter solving algorithm structure; Figure 6 It is a sound blast signal simulation equipment principle diagram; Figure 7 It is a sound blast signal simulation software interface DETAILED DESCRIPTION The specific embodiments of the present application are described below to facilitate those skilled in the art to understand the present application, but it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, it is obvious that various changes are within the spirit and scope of the present application defined and determined by the appended claims, and all inventions utilizing the concept of the present application are within the scope of protection.

[0028] Example one As Figure 1As shown, the present application provides a supersonic civil aircraft sound blast signal simulation test method, comprising the following steps: S1, obtaining the expected sound blast signal, generating the simulated sound blast signal of the expected sound blast signal, and calculating the equalization parameter of the simulated sound blast signal; S2, taking the expected sound blast signal as the input signal and the expected signal, generating the actual sound blast signal through the sound generating device, and calculating the filter coefficient vector; S3, taking the expected sound blast signal as the reference signal, and calculating the optimal filter coefficient vector according to the filter coefficient vector; S4, calculating the accurate input data according to the expected sound blast signal, the equalization parameter and the optimal filter coefficient vector, inputting the accurate input data into the sound generating device, and outputting the actual sound blast signal conforming to the expected sound blast signal.

[0029] In an embodiment of the present application, in the sound blast signal simulation technology, as shown in the single-point sound blast signal simulation principle block diagram, the sound blast signal simulation mainly includes two parts, one part is 1 / 3 octave equalization, and the other part is channel equalization. Figure 2 The 1 / 3 octave equalization is used for correcting the frequency response performance of the whole system, that is, flattening the frequency response of the whole sound generating system; and the channel equalization mainly makes the sound waves at the sound blast signal receiving point and the microphone receiving signal close to the set expected sound blast signal, and the basic principle is to use the sound pressure matching method. S1 of the present application is based on the channel equalization, which firstly obtains the expected sound blast signal , that is, sets the expected sound blast signal, and collects the sound waves at the microphone receiving signal at this time, that is, the simulated sound blast signal , calculates the channel response of the channel equalization at this time based on Fourier transform as:

[0030]

[0031] Among them, is the channel response of the sound generating device, is the simulated sound blast signal, is the Fourier transform of the expected sound blast signal, is the Fourier transform, is the expected sound blast signal.

[0032] When the response of the channel equalization satisfies , the expression of the microphone receiving signal at this time is as follows:

[0033] Among them, is the microphone receiving signal, is the equalization parameter, It is a delay factor; At this point, the signal received at the microphone is exactly the same as the desired sonic boom signal, and the transfer function that the channel equalizer should satisfy can be calculated:

[0034] Considering when When the value is small, inverting it results in instability, potentially exceeding the speaker's output capacity. Therefore, a regularization factor is added to eliminate this problem, resulting in the following expression for the transfer function:

[0035] in, It is the first regularization factor; In one embodiment of the present invention, such as Figure 3 The diagram shown is a block diagram of the channel identification adaptive filtering principle. The specific steps of S2 include: S201, Set the initial filter coefficients; S202. The desired sonic boom signal is used as the input signal and passed through the filter of the sound-generating device to obtain the actual sonic boom signal. S203. Using the desired sonic boom signal as a reference signal, the reference signal and the actual sonic boom signal are used to calculate the error signal. S204. Based on the adaptive filtering algorithm, set the adaptive convergence coefficients and calculate the filter coefficient vector with the goal of minimizing the mean square value of the error signal.

[0036] In a specific embodiment of the present invention, an FIR filter is used to characterize the channel response. ,like Figure 4 The diagram shown is a schematic of a transverse FIR filter. The expression for the initial filter coefficients is defined as follows:

[0037] The expression for the input signal is as follows:

[0038] In the In the next iteration, the output of the FIR filter, i.e., the actual sonic boom signal, is expressed as follows:

[0039] in, This is the actual sonic boom signal. Let the order be the filter order. For the filter number Step, For the filter number The filter coefficients of order, For the filter number input signal of the stage, is a transposition symbol.

[0040] output signal and the desired signal the difference is the error signal , the expression of the error signal is as follows:

[0041] After the error signal is obtained, the adaptive filter coefficient vector is constantly adjusted to make the error signal reach the minimum mean square value, in this process, the filter coefficient is a time-varying quantity, which is updated in real time by the adaptive algorithm, and the LMS least mean square algorithm can be used, that is, the square of the difference between the reference signal and the filter output signal is minimized, and the coefficient vector of the filter is updated according to this criterion, and the adaptive filtering algorithm generated thereby is called the least mean square algorithm. According to the objective function defined by the mean square error, it can be expressed as:

[0042]

[0043] When the filter coefficient is fixed, the objective function can also be written as:

[0044] wherein, is the cross-correlation vector of the desired signal and the input signal ;

[0045] is the autocorrelation matrix of the input signal :

[0046]

[0047] wherein, is the autocorrelation function of . From the cost function, it can be seen that the mean square error function is a quadratic equation of the filter coefficient vector , thereby forming a multi-dimensional hyper-parabolic surface, which is called the error performance surface of the adaptive filter, and the surface has a unique minimum point. Derive the cost function with respect to , and let it equal to zero, that is:

[0048] Assumption Since it is non-singular, the optimal filter coefficients that minimize the objective function can be obtained as follows:

[0049] This solution is optimal in terms of minimizing the mean square error. The steepest descent algorithm is used to solve it. This method makes each step change in the filter coefficients proportional to the negative value of the gradient vector of the mean square error. The formula is called the update formula, which is:

[0050] in, For the first The filter coefficient vector of the next iteration. For the first The filter coefficient vector of the next iteration. The adaptive convergence coefficient is a constant greater than 0, which controls the search step size and determines the stability and convergence of the iteration. Let be the gradient vector of the mean square error at the nth iteration. For any given initial coefficients of the adaptive filter... By selecting an appropriate step size and adaptively adjusting the filter coefficients according to the update formula, the corresponding filter coefficients approximate the Wiener solution infinitely. Having the update formula for the filter coefficients, the gradient of the mean square error is calculated below:

[0051] Substituting the gradient of the mean squared error into the update formula, we get:

[0052] Obtained using single sample data To replace mean square error Therefore, gradient calculation is performed, and the expression for the new gradient is as follows:

[0053] Substituting the expression for the new gradient into the update formula, we obtain the update formula for the filter coefficient vector as follows:

[0054] in, For the first The filter coefficient vector of the next iteration. For the first The filter coefficient vector of the next iteration. For adaptive convergence coefficients, For error signals, This is the input signal.

[0055] In one embodiment of the present invention, such asFigure 5 The specific steps of S3 include: S301, taking the desired sound blast signal as a reference signal, and performing convolution operation with the filter coefficient vector to obtain a filtered signal; S302, constructing a filtered signal vector based on the filtered signal; S303, calculating the autocorrelation matrix and cross-correlation variable through time average according to the filtered signal vector; S304, calculating the optimal filter coefficient through Wiener filtering according to the autocorrelation matrix and .

[0056] In specific embodiments of the present application, Figure 5 In the channel equalization filter, is a reference signal, representing the sound blast signal to be simulated, which reaches the error microphone through the primary channel and is recorded as the desired signal The primary channel is usually a pure delay, or a low-pass filter to remove the low-frequency energy components that the loudspeaker cannot produce. After the reference signal is input into the controller , the secondary signal is output, which passes through the secondary channel , interacts with at the error microphone, and obtains the error signal . The objective function of the channel equalization filter is the mean square error:

[0057] wherein is the time average of the independent variable; the desired signal and the error signal are respectively:

[0058]

[0059] The most commonly used implementation of the equalization controller is the transversal structure FIR filter, and the filter order is , so the weight coefficient of the filter of the th iteration and the reference input are represented in vector form as:

[0060]

[0061] secondary sound source input signal The filter output is expressed as follows:

[0062] The reference signal is convolved with the filter coefficient vector to obtain the filtered signal. The expression is as follows:

[0063] From the filtered signal The resulting column vector is the filtered signal vector, expressed as follows:

[0064] make , The expression for the cost function is obtained as follows:

[0065] in, To represent the mean square error of the desired sonic boom signal, for stationary primary noise, It is the filter weight vector The quadratic function, due to the autocorrelation matrix It is positive definite and symmetric, therefore There exists a unique minimum value, and the Wiener solution corresponding to the optimal filter coefficients is:

[0066] In the formula Represents a matrix To find the inverse, considering system stability, a second regularization factor can be added. The expression for the optimal filter coefficients is as follows:

[0067] After obtaining the optimal filter coefficients, in step S4 of this invention, accurate input data is calculated based on the desired sonic boom signal, equalization parameters, and the optimal filter coefficient vector. This accurate input data is then input to the sound-generating device, and the actual sonic boom signal that matches the desired sonic boom signal is output. The expression for the accurate input data is as follows:

[0068] in, To ensure accurate data input, For balancing parameters, For the desired sonic boom signal, These are the optimal filter coefficients.

[0069] In one embodiment of the present application, the conventional sonic boom signal simulation device generates the desired supersonic civil aircraft sonic boom signal in a laboratory environment for subjective listening experiments. Due to the influence of the loudspeaker and the surrounding environment reflection, the directly emitted N-type wave has a large distortion at the microphone receiving position. Therefore, the sonic boom signal simulation test method provided by the present application is used to pre-process the loudspeaker input signal, offset the distortion, and make the signal emitted in space close to the ideal supersonic civil aircraft sonic boom signal.

[0070] As shown in Figure 6 , it is a schematic diagram of the sonic boom signal simulation device. The method of the present application can be implemented by software, which can be called a sonic boom signal simulation software. The sonic boom signal simulation software is installed on an upper computer. The upper computer controls and interacts with the signal simulator through a network port. There are mainly two working modes. The first interactive mode is to set the initial parameters, including the sampling frequency, the signal simulator acquisition working mode, the sensor sensitivity, the buffer size setting, etc. The signal simulator working mode includes limited data acquisition and continuous data acquisition. The second interactive mode is the control command transmission, including four control commands: start data output, start data acquisition, stop data output, and stop data acquisition. After receiving the control command, the signal simulator transmits the collected data to the acquisition and control system according to different working modes, and at the same time, transmits the data sent by the acquisition and control system to the buffer in the signal simulator.

[0071] As shown in Figure 7 , it is a sonic boom signal simulation software interface, which can implement a supersonic civil aircraft sonic boom signal simulation test method of the present application. The specific implementation steps are as follows: (1) Step 1: Sound emitting system equalization The desired sonic boom signal is made as an audio file as the software input. Click "Load calibration file". The software automatically reads the sonic boom signal audio file data, directly simulates the input sonic boom signal, and automatically drives the microphone to collect the actual output sonic boom signal of the sound emitting system . The transfer function is calculated according to the formula , which is the equalization parameter.

[0072] (2) Step 2: Channel response solving Use the sonic boom signal data input in step 1 as the input signal and the desired signal . Click "Output calibration file". The software will drive the sound emitting system to simulate the sonic boom signal, use the microphone to collect the actual output sonic boom signal of the sound emitting system , and get the error signal . Set the initial value of the filter , and set a reasonable iteration step according to experience , the update formula of the filter coefficient vector is brought in to obtain the final filter coefficient .

[0073] (3) Step 3: Channel filter solving The expected sound blast signal is used as a reference signal And the filter coefficient obtained in step 2 Clicking the "sound blast signal output", the software can calculate the filter signal vector , and the formula The optimal filter coefficient can be obtained .

[0074] (4) Step 4: Preprocessing input signal data calculation The expected sound blast signal data, the transfer function obtained in step 1 And the optimal filter coefficient obtained in step 3 , the formula The sound emitting system input signal data can be calculated, and the data is directly used as the input, and the expected sound blast signal can be obtained at the microphone through the sound emitting system simulation.

[0075] The present application has the advantages that: the present application solves the signal distortion problem caused by the amplitude and phase error of the electro-acoustic system in the sound blast signal simulation, improves the fidelity and reliability of the sound blast simulation signal, corrects the phase distortion in the signal transmission process, makes the simulated sound blast signal close to the expected sound blast signal, and reproduces the accurate sound blast signal required by the technical personnel; meanwhile, the present application does not need to rely on specific hardware optimization, and through the adaptive adjustment ability of the algorithm, it can flexibly adapt to different sound blast signal simulation requirements, improves the stability of solving through regularization optimization, avoids the problems of signal extreme distortion and calculation divergence, improves the precision and consistency of the sound blast signal generation, and provides high-precision and high-reliability sound blast signal generation guarantee for sound blast related research and test analysis.

Claims

1. A method for simulating sonic boom signals in supersonic civil aircraft, characterized in that, Includes the following steps: S1. Obtain the desired sonic boom signal, generate a simulated sonic boom signal of the desired sonic boom signal, and calculate the equalization parameters of the simulated sonic boom signal. S2. Using the desired sonic boom signal as the input signal and the desired signal, the actual sonic boom signal is generated through the sound-generating device, and the filter coefficient vector is calculated. S3. Using the desired sonic boom signal as a reference signal, and based on the filter coefficient vector, calculate the optimal filter coefficient vector. S4. Based on the desired sonic boom signal, equalization parameters, and optimal filter coefficient vector, calculate the accurate input data, input the accurate input data to the sound-generating device, and output the actual sonic boom signal that conforms to the desired sonic boom signal.

2. The method for simulating sonic boom signals of supersonic civil aircraft according to claim 1, characterized in that, The expression for the equilibrium parameter is as follows: in, For balancing parameters, For the channel response of the sound-generating device, As a delay factor, For the delayed sampling period, To simulate a sonic boom signal, As the first regularization factor, For the desired Fourier transform of the sonic boom signal, For Fourier transform, This is the desired sonic boom signal.

3. The method for simulating sonic boom signals of supersonic civil aircraft according to claim 1, characterized in that, The specific steps of S2 include: S201, Set the initial filter coefficients; S202. The desired sonic boom signal is used as the input signal and passed through the filter of the sound-generating device to obtain the actual sonic boom signal. S203. Using the desired sonic boom signal as a reference signal, the reference signal and the actual sonic boom signal are used to calculate the error signal. S204. Based on the adaptive filtering algorithm, set the adaptive convergence coefficients and calculate the filter coefficient vector with the goal of minimizing the mean square value of the error signal.

4. The method for simulating sonic boom signals of supersonic civil aircraft according to claim 3, characterized in that, The expression for the filter coefficient vector is as follows: in, For the first The filter coefficient vector of the next iteration. For the first The filter coefficient vector of the next iteration. For adaptive convergence coefficients, For error signals, For input signal, For reference signal, This is the actual sonic boom signal. Let the order be the filter order. For the filter number Step, For the filter number The filter coefficients of order, For the filter number The input signal of order 1 This is the transpose symbol.

5. The method for simulating sonic boom signals of supersonic civil aircraft according to claim 1, characterized in that, The specific steps of S3 include: S301. Take the desired sonic boom signal as a reference signal and perform a convolution operation with the filter coefficient vector to obtain the filtered signal. S302. Construct a filtered signal vector based on the filtered signal; S303. Based on the filtered signal vector, calculate the autocorrelation matrix and cross-correlation variables by time averaging. S304. Based on the autocorrelation matrix and cross-correlation variables, the optimal filter coefficients are calculated using Wiener filtering.

6. The method for simulating sonic boom signals of supersonic civil aircraft according to claim 5, characterized in that, The expression for the optimal filter coefficients is as follows: in, For optimal filter coefficients, The autocorrelation matrix is... For cross-correlated variables, As the second regularization factor, This is the primary channel of the filter. For the desired sonic boom signal, Let the time-averaged mean square error of the desired sonic boom signal be... To take the time average of the independent variable, These are the filter coefficients during the iteration process. For filtered signals, This is the transpose symbol.

7. The method for simulating sonic boom signals of supersonic civil aircraft according to claim 1, characterized in that, The expression for the accurate input data is as follows: in, To ensure accurate data input, For balancing parameters, For the desired sonic boom signal, These are the optimal filter coefficients.

Images