CN115655631B

CN115655631B - Method and device for detecting voiceprint in wind tunnel environment based on hydraulic generator

Info

Publication number: CN115655631B
Application number: CN202211588517.8A
Authority: CN
Inventors: 曹祖杨; 陈启栋; 洪全付; 包君康; 张鑫; 方吉; 侯佩佩
Original assignee: Hangzhou Crysound Electronics Co Ltd
Current assignee: Hangzhou Crysound Electronics Co Ltd
Priority date: 2022-12-12
Filing date: 2022-12-12
Publication date: 2023-04-07
Anticipated expiration: 2042-12-12
Also published as: CN115655631A

Abstract

The application discloses a method and a device for detecting vocal prints in a wind tunnel environment based on a hydraulic generator, wherein the method comprises the steps of respectively acquiring first acoustic signals of the hydraulic generator based on eight measuring microphones arranged on the inner wall of the wind tunnel; processing the first sound signal to obtain a wave curve; dividing the wave curve into at least two sub-curve segments according to a preset frequency interval, and judging whether abnormal amplitude exists in all the sub-curve segments; when all the sub-curve segments are detected to have no abnormal amplitude, performing superposition calculation on all the sub-curve segments to obtain a superposed curve segment; and when the preset amplitude threshold value is exceeded in the superimposed curve segment, sending corresponding early warning information. The acoustic signal of the hydraulic generator is acquired by arranging the plurality of non-contact sensors, and corresponding early warning information is obtained by combining data analysis and various judgment conditions, so that abnormal sounds caused by various hidden dangers in the operation process of the hydraulic generator are effectively detected, and further the problem is prevented from being enlarged to cause economic loss.

Description

Method and device for detecting voiceprint in wind tunnel environment based on hydraulic generator

Technical Field

The application belongs to the technical field of voiceprint detection in complex environments, and particularly relates to a voiceprint detection method and device based on a hydraulic generator in a wind tunnel environment.

Background

The hydraulic generator is a generator which takes a hydraulic turbine as a prime mover to convert water energy into electric energy, when water flows through the hydraulic generator, the water energy can be converted into mechanical energy, and meanwhile, a rotating shaft of the hydraulic generator drives a rotor of the generator to convert the mechanical energy into the electric energy to be output.

The hydraulic generator needs to operate in a wind tunnel, and has the influences of strong wind, strong magnetism, strong vibration and the like during operation, the environment is complex and severe, at present, no effective detection means exists for the hydraulic generator, most of the hydraulic generator depends on a series of voltage and current values during the operation of the motor to assist in proving that the motor operates normally, but hidden dangers of abnormal sound, structure looseness, friction, matching of a rotor and a stator and the like of equipment cannot be detected, and only the outbreak of the problem can be passively waited; on the other hand, a small part of the hydraulic generators are equipped with contact sensors (vibration sensors), but since the environment of the hydraulic generator is strong vibration, the data detected by the vibration sensors are mostly vibration sound generated by the hydraulic generator itself, and only part of the problems can be solved.

Disclosure of Invention

The application provides a method and a device for detecting the voiceprint in the wind tunnel environment based on the hydraulic generator, for solving the technical problems that no effective detection means exists for the hydraulic generator at present, most of the prior methods are based on a series of voltage and current values during the operation of the motor to assist in proving that the operation of the motor is normal, but the hidden dangers of abnormal sound, structure looseness, friction, matching of a rotor and a stator and the like of equipment cannot be detected, and the problems can only be passively waited for, and the like, and the technical scheme is as follows:

in a first aspect, an embodiment of the present application provides a voiceprint detection method in a wind tunnel environment based on a hydraulic generator, including:

respectively acquiring first acoustic signals of the hydraulic generator based on eight measuring microphones arranged on the inner wall of the wind tunnel; the cross section of the wind tunnel is circular, and the included angles between any two adjacent measuring microphones and the circle center of the wind tunnel are kept consistent;

sampling each first acoustic signal based on the impact function, and performing Fourier transform processing on each first acoustic signal subjected to sampling processing to obtain a waveform curve corresponding to each first acoustic signal;

dividing a wave curve corresponding to each first sound signal into at least two sub-curve segments according to a preset frequency interval, and judging whether abnormal amplitude exists in all the sub-curve segments in the same frequency interval;

when detecting that all the sub-curve segments in the same frequency interval have no abnormal amplitude, performing superposition calculation on all the sub-curve segments in the same frequency interval to obtain a superposed curve segment;

and when the amplitude corresponding to any at least one frequency in the superimposed curve segment is determined to exceed the preset amplitude threshold, sending early warning information corresponding to the superimposed curve segment.

In an alternative of the first aspect, after acquiring the first acoustic signals of the hydraulic generators respectively based on eight measuring microphones disposed on the inner wall of the wind tunnel, the method further includes:

measuring the distance between each measuring microphone and the hydraulic generator;

when detecting that the distances from any at least two measuring microphones to the hydraulic generator are different, taking all first acoustic signals corresponding to the four measuring microphones closest to the hydraulic generator as a first set, and determining a first weight corresponding to the first set;

all first acoustic signals corresponding to the four measuring microphones with the farthest distances from the hydraulic generator serve as a second set, and a second weight corresponding to the second set is determined;

performing weighted summation calculation based on all the first acoustic signals in the first set, the first weight, all the first acoustic signals in the second set and the second weight to obtain a third acoustic signal;

determining a feature vector corresponding to the third sound signal, inputting the feature vector corresponding to the third sound signal and a preset standard feature vector into a trained deep learning model, and predicting the similarity between the third sound signal and the preset standard sound signal;

and when the similarity between the third sound signal and the preset standard sound signal is lower than a preset similarity threshold value, sending early warning information corresponding to the third sound signal.

In yet another alternative of the first aspect, after sampling each first acoustic signal based on the impulse function and performing fourier transform processing on each first acoustic signal subjected to the sampling processing to obtain a wave form curve corresponding to each first acoustic signal, the method further includes:

converting the voltage value corresponding to each frequency in each waveform curve to obtain a first pressure value corresponding to each frequency in each waveform curve;

acquiring atmospheric pressure at the position of the wind tunnel, and calculating a second pressure value corresponding to each frequency in each waveform curve according to the atmospheric pressure and a first pressure value corresponding to each frequency in each waveform curve;

bringing the second pressure value corresponding to each frequency in each waveform curve and the reference pressure parameter into a preset sound pressure calculation formula to obtain a third pressure value corresponding to each frequency in each waveform curve;

determining the pressure fluctuation range of each waveform curve according to the third pressure value corresponding to each frequency in each waveform curve;

and acquiring a pressure parameter interval of the hydraulic generator, and sending early warning information corresponding to the pressure fluctuation range of the waveform curve when detecting that the pressure fluctuation range of any waveform curve exceeds the pressure parameter interval of the hydraulic generator.

In yet another alternative of the first aspect, before acquiring the first acoustic signals of the hydraulic generators respectively based on eight measuring microphones disposed on the inner wall of the wind tunnel, the method further includes:

sending a standard sound signal with a voltage value being a preset voltage value to each measuring microphone, and acquiring a second sound signal acquired by each measuring microphone based on the standard sound signal;

and when the voltage value corresponding to any at least one second acoustic signal is detected to be inconsistent with the preset voltage value, calibrating each measuring microphone.

performing octave calculation processing on the waveform curve corresponding to each first sound signal to obtain a histogram corresponding to each first sound signal;

calculating the difference value of the energy value corresponding to each columnar curve in the standard octave histogram and the energy value corresponding to each columnar curve in the histogram corresponding to each first acoustic signal, and accumulating the calculated difference value to obtain the energy difference value of each first acoustic signal;

and when detecting that the energy difference value of any at least one first sound signal exceeds a preset energy threshold value, sending early warning information corresponding to the energy difference value of the first sound signal.

In yet another alternative of the first aspect, the histogram corresponding to each first acoustic signal includes at least two histogram curves, and each histogram curve has a preset proportional relationship between the lowest frequency and the highest frequency.

In yet another alternative of the first aspect, the method further comprises:

when the early warning information corresponding to the superposed curve segments and the early warning information corresponding to the pressure fluctuation ranges of the waveform curves are detected to be sent respectively, storing and processing all the superposed curve segments and the pressure fluctuation ranges of the waveform curves, and returning to the step of respectively obtaining the first sound signals of the hydraulic generator based on the eight measuring microphones arranged on the inner wall of the wind tunnel.

In a second aspect, an embodiment of the present application provides a voiceprint detection device in a wind tunnel environment based on a hydraulic generator, including:

the signal acquisition module is used for respectively acquiring first acoustic signals of the hydraulic generator based on eight measuring microphones arranged on the inner wall of the wind tunnel; the cross section of the wind tunnel is circular, and the included angles between any two adjacent measuring microphones and the circle center of the wind tunnel are kept consistent;

the signal processing module is used for sampling each first acoustic signal based on the impact function and carrying out Fourier transform processing on each first acoustic signal after sampling processing to obtain a waveform curve corresponding to each first acoustic signal;

the data analysis module is used for dividing the waveform curve corresponding to each first sound signal into at least two sub-curve segments according to a preset frequency interval and judging whether abnormal amplitude values exist in all the sub-curve segments in the same frequency interval or not;

the first detection module is used for carrying out superposition calculation on all the sub-curve segments in the same frequency interval to obtain a superposed curve segment when detecting that all the sub-curve segments in the same frequency interval have no abnormal amplitude;

and the second detection module is used for sending the early warning information corresponding to the superimposed curve segment when the amplitude corresponding to any at least one frequency in the superimposed curve segment is determined to exceed the preset amplitude threshold.

In an alternative of the second aspect, the apparatus further comprises:

after the first acoustic signals of the hydraulic generators are respectively acquired on the basis of eight measuring microphones arranged on the inner wall of the wind tunnel,

all first acoustic signals corresponding to the four measuring microphones which have the farthest distances from the hydraulic generator are used as a second set, and a second weight corresponding to the second set is determined;

performing weighted summation calculation based on all the first acoustic signals in the first set, the first weight values, all the first acoustic signals in the second set and the second weight values to obtain a third acoustic signal;

determining a feature vector corresponding to the third acoustic signal, inputting the feature vector corresponding to the third acoustic signal and a preset standard feature vector into a trained deep learning model, and predicting the similarity between the third acoustic signal and the preset standard acoustic signal;

In yet another alternative of the second aspect, the apparatus further comprises:

after sampling each first acoustic signal based on the impact function and performing fourier transform processing on each first acoustic signal subjected to the sampling processing to obtain a waveform curve corresponding to each first acoustic signal,

performing conversion processing on the voltage value corresponding to each frequency in each waveform curve to obtain a first pressure value corresponding to each frequency in each waveform curve;

determining the pressure fluctuation range of each wave curve according to the third pressure value corresponding to each frequency in each wave curve;

the method comprises the steps of obtaining a pressure parameter interval of the hydraulic generator, and sending early warning information corresponding to the pressure fluctuation range of any one waveform curve when the pressure fluctuation range of the waveform curve exceeds the pressure parameter interval of the hydraulic generator.

before the first acoustic signals of the hydraulic generator are respectively acquired on the basis of eight measuring microphones arranged on the inner wall of the wind tunnel,

calculating the difference value of the energy value corresponding to each columnar curve in the standard octave histogram and the energy value corresponding to each columnar curve in the histogram corresponding to each first sound signal, and accumulating the calculated difference value to obtain the energy difference value of each first sound signal;

and when the energy difference value of any at least one first sound signal is detected to exceed a preset energy threshold value, sending early warning information corresponding to the energy difference value of the first sound signal.

In yet another alternative of the second aspect, the histogram corresponding to each first acoustic signal includes at least two histogram curves, and each histogram curve has a preset proportional relationship between the lowest frequency and the highest frequency.

and when detecting that the early warning information corresponding to the superposed curve segments and the early warning information corresponding to the pressure fluctuation ranges of the waveform curves are respectively sent, storing and processing the pressure fluctuation ranges of all the superposed curve segments and all the waveform curves, and returning to the step of respectively acquiring the first sound signals of the hydraulic generator based on eight measuring microphones arranged on the inner wall of the wind tunnel.

In a third aspect, an embodiment of the present application further provides a voiceprint detection device in a wind tunnel environment based on a hydraulic generator, including a processor and a memory;

the processor is connected with the memory;

a memory for storing executable program code;

the processor reads the executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method for detecting the voiceprint in the wind tunnel environment based on the hydraulic generator provided by the first aspect of the embodiment of the present application or any one implementation manner of the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer storage medium, where a computer program is stored in the computer storage medium, where the computer program includes program instructions, and when the program instructions are executed by a processor, the method for detecting a voiceprint in a wind tunnel environment based on a hydraulic generator, where the method is provided by the first aspect or any implementation manner of the first aspect of the embodiment of the present application, may be implemented.

In the embodiment of the application, when the voiceprint detection of the hydraulic generator is carried out, the eight measuring microphones arranged on the inner wall of the wind tunnel are used for respectively acquiring the first acoustic signals of the hydraulic generator; sampling each first sound signal based on the impact function, and performing Fourier transform processing on each first sound signal subjected to sampling processing to obtain a waveform curve corresponding to each first sound signal; dividing a wave curve corresponding to each first sound signal into at least two sub-curve segments according to a preset frequency interval, and judging whether abnormal amplitude exists in all the sub-curve segments in the same frequency interval; when detecting that all the sub-curve segments in the same frequency interval have no abnormal amplitude, performing superposition calculation on all the sub-curve segments in the same frequency interval to obtain a superposed curve segment; and when the amplitude corresponding to any at least one frequency in the superimposed curve segment is determined to exceed a preset amplitude threshold, sending early warning information corresponding to the superimposed curve segment. The acoustic signal of the hydraulic generator is acquired by arranging the plurality of non-contact sensors, and corresponding early warning information is obtained by combining data analysis and various judgment conditions, so that abnormal sounds caused by various hidden dangers in the operation process of the hydraulic generator are effectively detected, and further the problem is prevented from being enlarged to cause economic loss.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

Fig. 1 is a schematic overall flow chart of a method for detecting a voiceprint in a wind tunnel environment based on a hydraulic generator according to an embodiment of the present disclosure;

fig. 2 is a schematic diagram of a waveform corresponding to an acoustic signal according to an embodiment of the present disclosure;

FIG. 3 is a histogram corresponding to an acoustic signal according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a voiceprint detection device in a wind tunnel environment based on a hydraulic generator according to an embodiment of the present application;

fig. 5 is a schematic structural diagram of another sound wave detection device in a wind tunnel environment based on a hydraulic generator according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

In the following description, the terms "first" and "second" are used for descriptive purposes only and are not intended to indicate or imply relative importance. The following description provides embodiments of the present application, where different embodiments may be substituted or combined, and thus the present application is intended to include all possible combinations of the same and/or different embodiments described. Thus, if one embodiment includes the feature A, B, C and another embodiment includes the feature B, D, then this application should also be considered to include embodiments that include one or more of all other possible combinations of A, B, C, D, although such embodiments may not be explicitly recited in text below.

The following description provides examples, and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made in the function and arrangement of elements described without departing from the scope of the disclosure. Various examples may omit, substitute, or add various procedures or components as appropriate. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

Referring to fig. 1, fig. 1 is a schematic overall flow chart of a method for detecting a voiceprint in a wind tunnel environment based on a hydraulic generator according to an embodiment of the present application.

As shown in fig. 1, the method for detecting the voiceprint in the wind tunnel environment based on the hydraulic generator at least includes the following steps:

and 102, respectively acquiring first acoustic signals of the hydraulic generator based on eight measuring microphones arranged on the inner wall of the wind tunnel.

In the embodiment of the application, the method for detecting the voiceprint in the wind tunnel environment based on the hydraulic generator can be applied to a control terminal, the control terminal can be connected with an acquisition card arranged in the wind tunnel through a server cabinet or a photoelectric converter, and the connection mode can be but is not limited to optical fiber wired connection. The inner wall of the wind tunnel can be provided with eight measuring microphones, the included angle between every two adjacent measuring microphones and the center of the wind tunnel is kept consistent, the wind tunnel can be understood as a channel with a circular section, the center of the wind tunnel can be correspondingly taken as the axis where the circle center is located, in other words, the included angle formed by every two adjacent measuring microphones and the center of the wind tunnel in the eight measuring microphones is 45 degrees, and each measuring microphone is connected with an acquisition card so as to facilitate the acquisition of the acoustic signal acquired by each measuring microphone by the acquisition card. The acoustic signals collected by each measuring microphone can be connected with the collecting card through an armored coaxial cable, and the whole wiring process of the armored coaxial cable needs to be combined with a metal waveguide tube arranged on the inner wall of the wind tunnel, so that signal interference caused by a magnetic field is effectively avoided.

It can be understood that the above-mentioned acquisition card can also be disposed in a specific waterproof and antimagnetic box to further avoid the interference of a severe environment, and an analog-to-digital conversion chip module, a gain module, a communication module, a data processing module, and the like are integrated in the acquisition card, so as to rapidly perform data processing on the acoustic signals acquired by the measurement microphone and ensure the accuracy of the data.

It should be noted that the purpose of setting eight measurement microphones in the embodiment of the present application is to determine the effectiveness and accuracy of each measurement microphone through the acoustic signals collected by a plurality of measurement microphones at the same time, and the setting mode in which the included angle formed between each two adjacent measurement microphones and the center of the wind tunnel is 45 degrees can further ensure the comparability and reliability of the acoustic signal collected by each measurement microphone, so that the acoustic line detection of the hydraulic generator can be realized together by combining the acoustic signals collected by each measurement microphone. Each measuring sensor can be further provided with a low-noise preamplifier and an outdoor waterproof protection device, and the low-noise preamplifier can be but is not limited to a preamplifier for supplying power to ICP (inductively coupled plasma) so as to be adapted to a collection card supporting IEPE/ICP (electronic equipment provider/inductively coupled plasma) power supply; the outdoor waterproof protection device can be provided with a windproof ball, so that the influence of wind noise on the measurement result can be effectively reduced.

Specifically, when the voiceprint detection of the hydraulic generator is carried out, the eight measuring microphones arranged on the inner wall of the wind tunnel can be controlled, and first acoustic signals sent by the hydraulic generator are respectively collected at the same moment, wherein the hydraulic generator can be arranged at any position in the wind tunnel, and before the voiceprint detection of the hydraulic generator is carried out, the hydraulic generator can be controlled by the control terminal to be in a working state in the wind tunnel.

It can be understood that the first acoustic signals collected by each measuring microphone can be sent to the acquisition card, and the acquisition card is controlled by the control terminal to simultaneously perform data processing and analysis on each first acoustic signal, so as to improve the overall testing efficiency and the accuracy of data.

As an optional option of the embodiment of the present application, after the eight measurement microphones disposed on the inner wall of the wind tunnel are used to respectively acquire the first acoustic signals of the hydraulic generator, the method further includes:

Because the position of the hydraulic generator cannot be accurately known in the wind tunnel, in order to further ensure the reliability and the accuracy of the first acoustic signal acquired by each measuring microphone, the first acoustic signal acquired by each measuring sensor can be analyzed by combining the distance between each measuring microphone and the hydraulic generator and the preset standard characteristic acoustic signal.

Specifically, the distance between each measurement microphone and the hydraulic generator may be acquired by a laser sensor provided on the measurement microphone, and the distances between each measurement microphone and the hydraulic generator may be sorted in order from near to far. It is possible that, when it is detected that the distances from any at least two measuring microphones to the hydraulic generator are different, it is determined that the hydraulic generator is not located at the axis of the center of the wind tunnel, then the first acoustic signals collected by the four measuring microphones closest to the center of the wind tunnel may be used as a first set, and the weights corresponding to the first set are determined, and the first acoustic signals collected by the four measuring microphones farthest from the center of the wind tunnel may be used as a second set, and the weights corresponding to the second set are determined. It can be understood that, in order to balance the accuracy and the error of the first acoustic signal collected by each measuring microphone, the accuracy of the first acoustic signal collected by the measuring microphone that is closer to the measuring microphone may be significantly higher than the accuracy of the first acoustic signal collected by the measuring microphone that is farther from the measuring microphone, and the second weight may be, but is not limited to being greater than the first weight, and the sum of the second weight and the first weight is 1. Then, a weighted summation calculation may be performed based on all the first acoustic signals in the first set, the first weight, all the first acoustic signals in the second set, and the second weight, so as to obtain a third acoustic signal, which may be, but is not limited to, represented as follows:

B=（A1+A2+A3+A4）× 0.3 +（A5+A6+A7+A8）× 0.7

in the above formula, B may correspond to an expression of the third acoustic signal, A1, A2, A3, and A4 may correspond to the first acoustic signal collected by the four closest measurement microphones, 0.3 may correspond to the first weight, A5, A6, A7, and A8 may correspond to the first acoustic signal collected by the four farthest measurement microphones, and 0.7 may correspond to the second weight.

Further, after the third acoustic signal is obtained, the third acoustic signal may be, but is not limited to, subjected to transform processing to obtain a wave-shaped curve corresponding to the third acoustic signal, and a plurality of feature points are extracted from the wave-shaped curve corresponding to the third acoustic signal as a feature vector corresponding to the third acoustic signal, where the manner of extracting the plurality of feature points from the wave-shaped curve corresponding to the third acoustic signal may be, but is not limited to, extraction by an image recognition technique, and a ratio between slopes corresponding to line segments before and after each feature point is a negative number.

Further, a feature vector corresponding to the third acoustic signal and a preset standard feature vector may be input to the trained deep learning model to predict the similarity between the third acoustic signal and the preset standard acoustic signal, wherein the preset standard feature vector may be extracted from the standard waveform, and the extraction manner may be similar to the feature vector corresponding to the third acoustic signal. It is understood that, in the embodiment of the present application, the structure of the deep learning model may be consistent with that of a common deep learning neural network, so as to be trained from sample feature vectors with known similarity to the preset standard acoustic signal, the output form of the deep learning model may be, but is not limited to, any value between 0 and 1, and the arbitrary data may be represented as the similarity between the third acoustic signal and the preset standard acoustic signal, and a value closer to 1 may indicate that the similarity between the third acoustic signal and the preset standard acoustic signal is higher. A closer to 0 may indicate that the third acoustic signal has a lower similarity to the predetermined standard acoustic signal.

It can be understood that the method can be used for preliminarily determining whether the first acoustic signals collected by each measuring microphone are abnormal, when the similarity between the third acoustic signal and the preset standard acoustic signal is lower than the preset similarity threshold, it is indicated that the first acoustic signal collected by at least one measuring microphone is obviously abnormal in the eight measuring microphones, and further, in order to ensure the accuracy of the overall test, the early warning information corresponding to the third acoustic signal can be generated and sent to the worker, and the early warning information can be, but is not limited to, representing that the third acoustic signal is abnormal, and each measuring microphone needs to be checked.

Possibly, when it is detected that the distances between any two measuring microphones and the hydraulic generator are kept the same, it can be determined that the hydraulic generator is located at the axis where the center of the wind tunnel is located, and the similarity between the first acoustic signals acquired by each measuring microphone can be directly judged, but not limited to, so that the measuring microphone with the abnormal acquired signal can be determined more quickly.

And 104, sampling each first acoustic signal based on the impact function, and performing Fourier transform processing on each first acoustic signal subjected to sampling processing to obtain a waveform curve corresponding to each first acoustic signal.

Specifically, after obtaining the first acoustic signals collected by each measurement microphone, the control terminal may perform sampling processing on each first acoustic signal based on an impulse function to ensure the continuity of each first acoustic signal after the sampling processing, where the impulse function may be, but is not limited to, represented as follows:

in the above-mentioned formula, the compound has the following structure,

the time may correspond to a time corresponding to the acquisition of each first acoustic signal.

Next, after each first acoustic signal is subjected to sampling processing, fourier transform processing may be further performed on each first acoustic signal subjected to sampling processing to obtain a time-domain discretized first acoustic signal, and then, in order to further reduce the amount of calculation, classification processing may be performed on each time-domain discretized first acoustic signal, and the first acoustic signal subjected to classification processing may be represented in the form of a waveform curve.

Here, the manner of performing fourier transform processing on each first acoustic signal after the sampling processing may be represented by, but is not limited to, the following:

in the above formula

May correspond to each first acoustic signal (also understood as aperiodic continuous-time signal) after the sampling process, since what is obtained in the practical process is

Based on this need by a discrete sampling signal in

The signal is sampled discretely to obtain a wave curve.

Wherein, will

The discrete sampled signal in (1) is denoted as x (n), which may be, but is not limited to being, expressed as follows:

x (n) can be decomposed into the sum of the even and odd sequences, i.e.:

here, the

And

the lengths of the two groups of the optical fibers are N/2,

the correspondence is to the even-numbered sequence,

corresponding to an odd number sequence, the expression corresponding to the above formula x (n) can be converted into:

further convertible to:

due to the fact that

Having periodicity, when n is an even number (denoted as n here by 2 n), the above equation can also be converted to:

it can further be deduced that:

this time game

It can be seen that when n is an even number,

when n is an odd number, it can be extracted

As a common factor to let the rest

The condition that n is an even number is satisfied.

Then, the even part can be made to be

The result of extracting the common factor from the odd part is

Then the following can be obtained:

due to the fact that

And

are all known and fixed, and

satisfy the requirement of

Then, then

And an

Further derivation can be derived from the following formula

：

The following conclusions can be drawn:

that is, as long as it is obtained

Can obtain

。

Based on the above, can also

To convert to:

through the derivation process and the conversion process, the acquired first acoustic signals can be rapidly processed in a large scale in the embodiment of the application, and can be connected in a point-to-line mode in a curve window of corresponding software, so that a corresponding wave curve is obtained.

Fig. 2 is a schematic diagram of a waveform curve corresponding to an acoustic signal according to an embodiment of the present disclosure. As shown in fig. 2, the horizontal axis of the waveform curve corresponding to each first acoustic signal may correspond to a frequency interval, and the vertical axis may correspond to an energy (decibel value) interval, and the horizontal axis may be, but is not limited to, labeled with 100Hz, 1000Hz, and 10000Hz, and the vertical axis may be, but is not limited to, labeled with 50dB, 100dB, and 150dB, which may not be limited thereto.

And 106, dividing the waveform curve corresponding to each first acoustic signal into at least two sub-curve segments according to a preset frequency interval, and judging whether abnormal amplitude exists in all the sub-curve segments in the same frequency interval.

After the waveform curve corresponding to each first acoustic signal is obtained, the waveform curve corresponding to each first acoustic signal may be divided according to a preset frequency interval to obtain at least two sub-curve segments, wherein the same frequency interval may correspond to eight divided sub-curve segments, and whether all the sub-curve segments in the same frequency interval have abnormal amplitudes may be determined, in which the determination method may be, but is not limited to, placing each sub-curve segment in an energy interval corresponding to the frequency interval, and possibly, when a partial line segment of the sub-curve segment exceeds the energy interval corresponding to the frequency interval, it may indicate that the sub-curve segment has abnormal amplitudes, and may notify an operator to perform inspection and analysis on the sub-curve segment. It can be understood that, here, the staff may, but is not limited to, compare the sub-curve segments with abnormal amplitudes according to the combined historical fault data record, for example, when it is determined that there is an error in the working parameters of the hydraulic generator according to the comparison result, the staff may remotely perform parameter correction processing on the hydraulic generator, and return to the step of respectively acquiring the first acoustic signals of the hydraulic generator based on the eight measurement microphones disposed on the inner wall of the wind tunnel, so as to acquire all the sub-curve segments within the same frequency interval again until all the sub-curve segments are within the energy interval corresponding to the frequency interval. It is possible that when the sub-curve segments are all in the energy interval corresponding to the frequency interval, it can be shown that there is no abnormal amplitude for the sub-curve segments.

And 108, when detecting that all the sub-curve segments in the same frequency interval have no abnormal amplitude, performing superposition calculation on all the sub-curve segments in the same frequency interval to obtain a superposed curve segment.

Specifically, when it is determined that there is no abnormal amplitude in the eight sub-curve segments located in the same frequency interval, the superposition calculation may be performed on all the sub-curve segments in a manner corresponding to the frequency to obtain a superposed curve segment corresponding to each frequency interval. It can be understood that, in the embodiment of the present application, it may be preliminarily determined whether each sub-curve segment has an abnormal amplitude value through an energy interval corresponding to a frequency interval, and then it may be determined whether the superimposed amplitude values of all sub-curve segments are abnormal through an amplitude threshold, so as to ensure the accuracy of data in a multiple detection manner.

Of course, if one or more extra peaks (i.e., a large energy difference) appear at the same frequency in the process of performing superposition calculation on all the sub-curve segments in a frequency corresponding manner, the warning information corresponding to the sub-curve segments may be sent to prompt the staff that there may be an abnormality in the acoustic signal in the frequency interval.

And 110, when the amplitude corresponding to any at least one frequency in the superimposed curve segment is determined to exceed a preset amplitude threshold, sending early warning information corresponding to the superimposed curve segment.

Specifically, after all the sub-curve segments are subjected to superposition calculation in a frequency corresponding mode, when the fact that the amplitude corresponding to any at least one frequency exceeds a preset amplitude threshold value is determined, it can be shown that the acoustic signal in the frequency interval is abnormal, and then early warning information corresponding to the superposed curve segments is sent to inform workers of timely closing the hydraulic generator and carrying out inspection processing on the hydraulic generator.

As a further optional option of the embodiment of the present application, after sampling each first acoustic signal based on the impulse function and performing fourier transform processing on each first acoustic signal subjected to the sampling processing to obtain a waveform curve corresponding to each first acoustic signal, the method further includes:

In this application embodiment, except that detecting hydraulic generator through the energy amplitude who corresponds with the frequency, still can further detect hydraulic generator through the pressure fluctuation scope that calculates, when reducing the manual inspection cost to improve the accuracy of whole detection through multiple detection mode, and then avoid because detect the economy and the safety problem that hidden danger caused.

Specifically, in the process that the control terminal processes each first acoustic signal to obtain the waveform curve corresponding to each first acoustic signal, the control terminal may further detect a voltage value corresponding to each frequency in the waveform curve corresponding to each first acoustic signal based on the aforementioned acquisition card, where the detection mode may be, but is not limited to, detection by a multimeter disposed inside the acquisition card, and perform conversion processing on the voltage value to obtain a first pressure value corresponding to each frequency in each waveform curve. Wherein the sensitivity of the measuring microphone applied in the embodiment of the present application is 15mV/Pa, in other words, every pressure increase of 1Pa, the voltage increases by 15mV for stool and urine, and here, the first pressure value can be obtained according to the following formula:

P1=（V1-V0）/ 15

in the above formula, P1 may correspond to a first pressure value corresponding to each frequency in the waveform curve, V1 may correspond to a voltage value corresponding to each frequency in the waveform curve, and V0 may correspond to an initial voltage value detected by the acquisition card before the measurement microphone acquires the acoustic signal.

Further, the termination terminal may obtain an atmospheric pressure at a position where the wind tunnel is located, the atmospheric pressure may be a fixed value, a specific numerical value of the atmospheric pressure may be directly obtained through internet query, and the atmospheric pressure and the first pressure value are subjected to addition calculation to obtain a current pressure value of the hydraulic generator in a working state, that is, a second pressure value corresponding to each frequency in each waveform curve.

Further, after calculating the second pressure value corresponding to each frequency in each waveform curve, the second pressure value corresponding to each frequency in each waveform curve and the reference pressure parameter may be substituted into a preset sound pressure calculation formula to obtain a third pressure value corresponding to each frequency in each waveform curve, wherein the preset sound pressure calculation formula may be, but is not limited to, expressed as follows:

in the above equation, SPL corresponds to a third pressure value corresponding to each frequency in each waveform curve,

corresponding to the second pressure value corresponding to each frequency in each waveform curve,

the correspondence is a reference pressure parameter, and the reference pressure parameter may be, but is not limited to, fixed at 20uPa.

Further, after obtaining the third pressure value corresponding to each frequency in each waveform curve, a pressure fluctuation range of each waveform curve may be determined according to the third pressure value corresponding to each frequency in each waveform curve, a minimum value of the pressure fluctuation range is a minimum pressure value of the third pressure values corresponding to each frequency in each waveform curve, and a maximum value of the pressure fluctuation range is a maximum pressure value of the third pressure values corresponding to each frequency in each waveform curve.

Furthermore, the control terminal can also inquire a pressure parameter interval in the operating parameters of the hydraulic generator, and when the pressure fluctuation range of any one waveform curve exceeds the pressure parameter interval of the hydraulic generator, the acoustic signal corresponding to the waveform curve is indicated to be abnormal, and then the early warning information corresponding to the pressure fluctuation range of the waveform curve can be generated to remind a worker of processing the pressure parameter interval immediately.

As a further alternative of the embodiment of the present application, before the eight measurement microphones disposed on the inner wall of the wind tunnel are used to respectively obtain the first acoustic signals of the hydro-generator, the method further includes:

In the embodiment of the application, before the test is performed, whether the detected voltage value has a deviation or not can be judged through the standard sound signal, so that the accuracy of the voltage value is improved.

Specifically, before first acoustic signals acquired by eight measuring microphones are acquired, the control terminal can also send a standard acoustic signal with a voltage value of a preset voltage value to each measuring microphone, the acquisition card acquires an acoustic signal acquired by each measuring microphone under the standard acoustic signal, and the multimeter arranged in the acquisition card acquires a corresponding voltage value, when the voltage value corresponding to any one measuring microphone is detected to be not the preset voltage value, it is indicated that a device for detecting the measuring microphone may have a deviation, and meanwhile, it is also indicated that the measuring microphone may have a deviation, so that the measuring microphones and voltage detection devices corresponding to the measuring microphones can be corrected, and the accuracy of the acoustic signals is guaranteed.

calculating the difference value of the energy value corresponding to each frequency in the standard octave histogram and the energy value corresponding to each frequency in the histogram corresponding to each first acoustic signal, and accumulating the calculation result of the difference value to obtain the energy difference value of each first acoustic signal;

In this application embodiment, except that detecting hydraulic generator through the energy amplitude who corresponds with the frequency, still can further detect hydraulic generator through the histogram that calculates, when reducing artifical inspection cost to improve the accuracy of whole detection through multiple detection mode, and then avoid because the economy and the safety problem that find out hidden danger and cause.

Since the frequency of the sound heard by human ears is about 20 to 20000hz, but the sensitivity is different for different frequencies: it is easy to distinguish low frequencies, e.g. 50Hz and 100Hz, but it is increasingly difficult to distinguish high frequencies, e.g. 10000H at and 10050Hz. Therefore, in the process of carrying out voiceprint detection on the hydraulic generator, the low frequency can be divided into a little as possible, and the characteristics of the low frequency can be more fully reflected so as to be convenient for analysis.

Specifically, in the process of processing each first acoustic signal to obtain a waveform curve corresponding to each first acoustic signal, the control terminal may further perform octave calculation processing on the waveform curve corresponding to each first acoustic signal to obtain a histogram corresponding to each first acoustic signal, where the histogram corresponding to each first acoustic signal may include at least two histogram curves, a lowest frequency and a highest frequency corresponding to each histogram curve are in a preset proportional relationship, and a highest frequency of a previous histogram curve in two adjacent histogram curves is equal to a lowest frequency of a next histogram curve, in other words, a frequency interval of an nth histogram curve may be sequentially calculated according to a preset proportional relationship between the lowest frequency and the highest frequency of the first histogram curve and the highest frequency, and the frequency interval may specifically include the lowest frequency, the highest frequency, and a center frequency.

It is understood that the control terminal may determine the lowest frequency of the first histogram in the waveform corresponding to each first acoustic signal, or set a preset lowest frequency as the lowest frequency of the first histogram, and may obtain, but is not limited to, the highest frequency and the center frequency of the first histogram by the following formula:

in the above formula, the first and second carbon atoms are,

may correspond to the highest frequency of the first histogram,

may correspond to the lowest frequency of the first histogram, i.e. a known value,

can be aligned withThe center frequency of the first histogram.

By the above formula and the lowest frequency of the first histogram, the frequency interval of the nth histogram can be sequentially calculated, and the energy value corresponding to the ith histogram can be calculated by, but not limited to, the following formula:

in the above formula, k is an integer from 3i-2 to 3i,

for representing the energy values corresponding to the lowest frequency, the highest frequency and the center frequency constituting the ith histogram, the energy values can be determined in the waveform corresponding to each first acoustic signal.

Referring to fig. 3, a histogram corresponding to an acoustic signal according to an embodiment of the present disclosure is shown, and as shown in fig. 3, the histogram includes a waveform curve corresponding to the acoustic signal and a histogram corresponding to the acoustic signal, an abscissa may correspond to a frequency interval, and an ordinate may correspond to an energy value. And a central frequency value corresponding to each columnar curve is plotted in the abscissa, and an energy value corresponding to each columnar curve is plotted in the ordinate.

Further, after obtaining the histogram corresponding to each first acoustic signal, the control terminal may further perform difference calculation on an energy value corresponding to each histogram curve in the standard octave histogram and an energy value corresponding to each histogram curve in the histogram corresponding to each first acoustic signal, and perform accumulation processing on the difference calculation result to obtain an energy difference value of each first acoustic signal, where the energy difference value of each first acoustic signal may be, but is not limited to, calculated by the following formula:

in the above-mentioned formula, the compound has the following structure,

may correspond to the energy difference of the ith histogram,

may correspond to the energy value of the ith histogram,

which may correspond to the energy value of the ith standard histogram.

Further, when the energy difference value of the columnar curve in any at least one first sound signal is detected to exceed a preset energy threshold value, the first sound signal is indicated to be abnormal, and then early warning information corresponding to the energy difference value of the first sound signal is sent to prompt a worker that the energy value of the first sound signal is abnormal, and noise exists in a corresponding frequency interval and needs to be processed immediately. It can be understood that, when a worker performs denoising processing on a frequency interval with an abnormal energy value, the worker may, but is not limited to, perform wavelet continuous transformation processing on an acoustic signal in the frequency interval, then divide the processed acoustic signal into at least two sub-signals, select a value with a larger amplitude as a preset threshold through a cmp function, perform screening processing on a signal parameter of each sub-signal according to the preset threshold, and may, but is not limited to, retain the signal parameter of any sub-signal when the absolute value of the signal parameter of the sub-signal is detected to be larger than the preset threshold; when the absolute value of the signal parameter of any sub-signal is detected to be smaller than the preset threshold, the signal parameter of the sub-signal can be set to 0 (that is, the sub-signal is subjected to filtering processing).

As another option of the embodiment of the application, when it is detected that the control terminal sends the early warning information corresponding to the superimposed curve segments and the early warning information corresponding to the pressure fluctuation ranges of the waveform curves respectively, it indicates that the hydraulic generator has a very high possibility of abnormality, based on this, all the superimposed curve segments and the pressure fluctuation ranges of all the waveform curves need to be stored and processed, so as to perform data analysis, simultaneously stop the current test stage and control the hydraulic generator to be in a stop working state, and after a preset time interval, respectively obtain the first acoustic signals of the hydraulic generator based on the eight measurement microphones arranged on the inner wall of the wind tunnel again, so as to perform sound streak detection on the hydraulic generator again.

Referring to fig. 4, fig. 4 is a schematic structural diagram illustrating a voiceprint detection device in a wind tunnel environment based on a hydraulic generator according to an embodiment of the present application.

As shown in fig. 4, the voiceprint detection device based on the hydraulic generator in the wind tunnel environment at least may include a signal acquisition module 401, a signal processing module 402, a data analysis module 403, a first detection module 404, and a second detection module 405, where:

the signal acquisition module 401 is configured to respectively acquire first acoustic signals of the hydraulic generator based on eight measurement microphones arranged on an inner wall of the wind tunnel; the cross section of the wind tunnel is circular, and the included angles between any two adjacent measuring microphones and the circle center of the wind tunnel are kept consistent;

the signal processing module 402 is configured to sample each first acoustic signal based on the impact function, and perform fourier transform processing on each sampled first acoustic signal to obtain a waveform curve corresponding to each first acoustic signal;

the data analysis module 403 is configured to divide the waveform curve corresponding to each first acoustic signal into at least two sub-curve segments according to a preset frequency interval, and determine whether abnormal amplitudes exist in all the sub-curve segments in the same frequency interval;

the first detection module 404 is configured to, when it is detected that all sub-curve segments in the same frequency interval do not have abnormal amplitudes, perform superposition calculation on all sub-curve segments in the same frequency interval to obtain a superposed curve segment;

the second detection module 405 is configured to send the warning information corresponding to the superimposed curve segment when it is determined that the amplitude corresponding to any at least one frequency in the superimposed curve segment exceeds a preset amplitude threshold.

In some possible embodiments, the apparatus further comprises:

after the first acoustic signals of the hydro-generator are respectively acquired based on the eight measuring microphones arranged on the inner wall of the wind tunnel,

In some possible embodiments, the apparatus further comprises:

before the eight measuring microphones arranged on the inner wall of the wind tunnel are used for respectively acquiring the first acoustic signals of the hydraulic generator,

and when the voltage value corresponding to any at least one second acoustic signal is not consistent with the preset voltage value, calibrating each measuring microphone.

In some possible embodiments, the apparatus further comprises:

In some possible embodiments, the histogram corresponding to each first acoustic signal includes at least two histogram curves, and each histogram curve has a preset proportional relationship between the lowest frequency and the highest frequency.

In some possible embodiments, the apparatus further comprises:

It is clear to a person skilled in the art that the solution according to the embodiments of the present application can be implemented by means of software and/or hardware. The "unit" and "module" in this specification refer to software and/or hardware that can perform a specific function independently or in cooperation with other components, where the hardware may be, for example, a Field-Programmable Gate Array (FPGA), an Integrated Circuit (IC), or the like.

Referring to fig. 5, fig. 5 is a schematic structural diagram illustrating another voiceprint detection device based on a hydraulic generator in a wind tunnel environment according to an embodiment of the present application.

As shown in fig. 5, the voiceprint detection apparatus 500 based on the hydraulic generator in the wind tunnel environment may include at least one processor 501, at least one network interface 504, a user interface 503, a memory 505, and at least one communication bus 502.

The communication bus 502 can be used for realizing the connection communication of the above components.

The user interface 503 may include keys, and the optional user interface may also include a standard wired interface or a wireless interface.

The network interface 504 may include, but is not limited to, a bluetooth module, an NFC module, a Wi-Fi module, and the like.

Processor 501 may include one or more processing cores, among other things. Processor 501 connects various portions of the entire hydraulic generator-based voiceprint detection apparatus 500 in a wind tunnel environment using various interfaces and lines, and performs various functions of routing hydraulic generator-based voiceprint detection apparatus 500 in a wind tunnel environment and processing data by executing or executing instructions, programs, code sets, or instruction sets stored in memory 505, and calling data stored in memory 505. Optionally, the processor 501 may be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 501 may integrate one or a combination of several of a CPU, GPU, modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 501, but may be implemented by a single chip.

The memory 505 may include a RAM or a ROM. Optionally, the memory 505 includes a non-transitory computer-readable medium. The memory 505 may be used to store instructions, programs, code, sets of codes, or sets of instructions. The memory 505 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the various method embodiments described above, and the like; the storage data area may store data and the like referred to in the above respective method embodiments. The memory 505 may alternatively be at least one memory device located remotely from the processor 501. As shown in fig. 5, the memory 505, which is a type of computer storage medium, may include an operating system, a network communication module, a user interface module, and a voiceprint detection application in a wind tunnel environment based on a hydraulic generator.

Specifically, the processor 501 may be configured to invoke a voiceprint detection application program based on the hydraulic generator in the wind tunnel environment, which is stored in the memory 505, and specifically perform the following operations:

and when the amplitude corresponding to any at least one frequency in the superimposed curve segment is determined to exceed a preset amplitude threshold, sending early warning information corresponding to the superimposed curve segment.

In some possible embodiments, after acquiring the first acoustic signals of the hydraulic generators based on the eight measuring microphones disposed on the inner wall of the wind tunnel, the method further includes:

In some possible embodiments, after sampling each first acoustic signal based on the impulse function and performing fourier transform processing on each first acoustic signal subjected to the sampling processing to obtain a waveform curve corresponding to each first acoustic signal, the method further includes:

In some possible embodiments, before acquiring the first acoustic signals of the hydraulic generators respectively based on the eight measuring microphones disposed on the inner wall of the wind tunnel, the method further includes:

In some possible embodiments, the method further comprises:

The present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the above-described method. The computer-readable storage medium may include, but is not limited to, any type of disk including floppy disks, optical disks, DVD, CD-ROMs, microdrive, and magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic or optical cards, nanosystems (including molecular memory ICs), or any type of media or device suitable for storing instructions and/or data.

It should be noted that for simplicity of description, the above-mentioned embodiments of the method are described as a series of acts, but those skilled in the art should understand that the present application is not limited by the described order of acts, as some steps may be performed in other orders or simultaneously according to the present application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.

In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is only a logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed coupling or direct coupling or communication connection between each other may be through some service interfaces, indirect coupling or communication connection of devices or units, and may be electrical or in other forms.

Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable memory. Based on such understanding, the technical solution of the present application may be substantially implemented or a part of or all or part of the technical solution contributing to the prior art may be embodied in the form of a software product stored in a memory, and including several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method of the embodiments of the present application. And the aforementioned memory comprises: various media capable of storing program codes, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program which instructs associated hardware to perform the steps, and the program may be stored in a computer readable memory, and the memory may include: flash disks, read-Only memories (ROMs), random Access Memories (RAMs), magnetic or optical disks, and the like.

The above are merely exemplary embodiments of the present disclosure, and the scope of the present disclosure should not be limited thereby. That is, all equivalent changes and modifications made in accordance with the teachings of the present disclosure are intended to be included within the scope of the present disclosure. Embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

Claims

1. A method for detecting voiceprint in a wind tunnel environment based on a hydraulic generator is characterized by comprising the following steps:

sampling each first acoustic signal based on an impact function, and performing Fourier transform processing on each first acoustic signal subjected to sampling processing to obtain a waveform curve corresponding to each first acoustic signal;

dividing a waveform curve corresponding to each first sound signal into at least two sub-curve segments according to a preset frequency interval, and judging whether abnormal amplitude exists in all the sub-curve segments in the same frequency interval;

when the amplitude corresponding to any at least one frequency in the superimposed curve segment is determined to exceed a preset amplitude threshold, sending early warning information corresponding to the superimposed curve segment;

wherein, after respectively obtaining hydraulic generator's first acoustic signal based on eight measuring microphones of setting at the wind tunnel inner wall, still include:

when detecting that the distances from any at least two measuring microphones to the hydraulic generator are different, taking all the first acoustic signals corresponding to the four measuring microphones with the shortest distances to the hydraulic generator as a first set, and determining a first weight corresponding to the first set;

taking all the first acoustic signals corresponding to the four measuring microphones with the farthest distances between the measuring microphones and the hydraulic generator as a second set, and determining a second weight corresponding to the second set;

determining a feature vector corresponding to the third acoustic signal, inputting the feature vector corresponding to the third acoustic signal and a preset standard feature vector into a trained deep learning model, and predicting the similarity between the third acoustic signal and a preset standard acoustic signal;

and when the similarity between the third sound signal and the preset standard sound signal is lower than a preset similarity threshold, sending early warning information corresponding to the third sound signal.

2. The method according to claim 1, wherein after the sampling each of the first acoustic signals based on the impact function and performing fourier transform processing on each of the sampled first acoustic signals to obtain a waveform curve corresponding to each of the first acoustic signals, the method further comprises:

bringing a second pressure value corresponding to each frequency in each waveform curve and a reference pressure parameter into a preset sound pressure calculation formula to obtain a third pressure value corresponding to each frequency in each waveform curve;

and acquiring a pressure parameter interval of the hydraulic generator, and sending early warning information corresponding to the pressure fluctuation range of the waveform curve when detecting that the pressure fluctuation range of any one waveform curve exceeds the pressure parameter interval of the hydraulic generator.

3. The method according to claim 2, wherein before the step of obtaining the first acoustic signals of the hydraulic generators respectively based on the eight measuring microphones arranged on the inner wall of the wind tunnel, the method further comprises the following steps:

4. The method according to claim 1, wherein after the sampling each of the first acoustic signals based on the impact function and performing fourier transform processing on each of the sampled first acoustic signals to obtain a waveform curve corresponding to each of the first acoustic signals, the method further comprises:

performing octave calculation processing on the waveform curve corresponding to each first acoustic signal to obtain a histogram corresponding to each first acoustic signal;

calculating difference values of energy values corresponding to each columnar curve in a standard octave histogram and energy values corresponding to each columnar curve in the histogram corresponding to each first sound signal, and accumulating the calculated difference values to obtain the energy difference value of each first sound signal;

5. The method of claim 4, wherein the histogram corresponding to each of the first acoustic signals comprises at least two histogram curves, and each histogram curve has a predetermined ratio between the lowest frequency and the highest frequency.

6. The method of claim 2, further comprising:

and when detecting that the early warning information corresponding to the superposed curve segments and the early warning information corresponding to the pressure fluctuation ranges of the waveform curves are respectively sent, storing and processing all the superposed curve segments and the pressure fluctuation ranges of the waveform curves, and returning to the step of respectively acquiring the first sound signals of the hydraulic generator based on eight measuring microphones arranged on the inner wall of the wind tunnel.

7. The utility model provides a vocal print detection device in wind tunnel environment based on hydraulic generator which characterized in that includes:

the signal acquisition module is used for respectively acquiring first acoustic signals of the hydraulic generator based on the eight measuring microphones arranged on the inner wall of the wind tunnel; the cross section of the wind tunnel is circular, and the included angles between any two adjacent measuring microphones and the circle center of the wind tunnel are kept consistent;

the signal processing module is used for sampling each first acoustic signal based on an impact function and carrying out Fourier transform processing on each first acoustic signal subjected to sampling processing to obtain a waveform curve corresponding to each first acoustic signal;

the data analysis module is used for dividing the waveform curve corresponding to each first acoustic signal into at least two sub-curve segments according to a preset frequency interval and judging whether abnormal amplitude exists in all the sub-curve segments in the same frequency interval or not;

the second detection module is used for sending early warning information corresponding to the superimposed curve segment when determining that the amplitude corresponding to any at least one frequency in the superimposed curve segment exceeds a preset amplitude threshold;

wherein, after eight measurement microphones based on setting up at the wind-tunnel inner wall obtain hydraulic generator's first acoustic signal respectively, still include:

all the first acoustic signals corresponding to the four measuring microphones which have the farthest distances from the hydraulic generator are used as a second set, and a second weight corresponding to the second set is determined;

8. A sound pattern detection device based on a hydraulic generator in a wind tunnel environment is characterized by comprising a processor and a memory;

the processor is connected with the memory;

the memory for storing executable program code;

the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory for performing the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium on which a computer program is stored, having instructions stored therein, which when run on a computer or processor, cause the computer or processor to perform the steps of the method according to any one of claims 1-6.