CN115795282B

CN115795282B - Shock tube dynamic pressure reconstruction method and device, electronic equipment and storage medium

Info

Publication number: CN115795282B
Application number: CN202310045368.9A
Authority: CN
Inventors: 姚贞建; 李永生; 丁义凡
Original assignee: Wuhan Institute of Technology
Current assignee: Wuhan Institute of Technology
Priority date: 2023-01-30
Filing date: 2023-01-30
Publication date: 2023-05-09
Anticipated expiration: 2043-01-30
Also published as: CN115795282A

Abstract

The invention provides a shock tube dynamic pressure reconstruction method, a shock tube dynamic pressure reconstruction device, electronic equipment and a storage medium, wherein the shock tube dynamic pressure reconstruction method comprises the following steps: acquiring an initial dynamic pressure response signal, including a vibration signal and a response signal; preprocessing a vibration signal and a response signal based on a variation mode decomposition method and an empirical mode decomposition method to obtain a preprocessed signal and component signals of the signal in different frequency bands; then, according to the relation between each component signal obtained by the correlation coefficient and the preprocessing signal and the denoising vibration signal, a training set is constructed; constructing an initial inverse sensing network model based on the Bi-LSTM neural network model, and iteratively training the initial inverse sensing network model based on a training set to obtain a target inverse sensing network model; and reconstructing the dynamic pressure of the shock tube by adopting the target inverse sensing network model to obtain a target shock tube dynamic pressure reconstruction signal. The shock tube dynamic pressure reconstruction method provided by the invention can reconstruct the shock tube dynamic pressure signal more reasonably and accurately.

Description

Shock tube dynamic pressure reconstruction method and device, electronic equipment and storage medium

Technical Field

The invention relates to the technical field of metering test, in particular to a shock tube dynamic pressure reconstruction method, a shock tube dynamic pressure reconstruction device, electronic equipment and a storage medium.

Background

Shock tube dynamic pressure is widely used in the fields of explosion testing, medical instruments, material impact testing, aeroengines and the like. In actual measurement, the steady duration of dynamic pressure is typically several milliseconds to tens of milliseconds. The generation of the dynamic pressure of the shock tube is a transient dynamic process, and not only is the testing environment complex, but also the dynamic pressure is difficult to control, so that the dynamic pressure signal is difficult to accurately estimate, and the measurement accuracy of the dynamic pressure signal is seriously affected.

The existing methods all consider the dynamic pressure generated by the shock tube as ideal step pressure, namely the amplitude of the dynamic pressure is constant. However, the amplitude of the dynamic pressure generated by the actual shock tube fluctuates with time, and a certain idealized assumption exists in the representation of the dynamic pressure of the shock tube by using a constant amplitude, which inevitably leads to unreasonable representation results. In addition, in the working process of the shock tube, impact vibration is generated when incident shock waves reach the end face of the low-pressure chamber, the pressure sensor arranged on the end face is simultaneously excited by a dynamic pressure signal and a vibration signal, the acquired pressure sensor output signal is a mixed signal of the dynamic pressure response and the vibration response, the influence of the vibration response is not considered in the conventional method, and the obtained dynamic pressure amplitude estimation result is inaccurate.

In summary, in the prior art, the fluctuation characteristics of the dynamic pressure signal of the shock tube and the influence of the impact vibration on the dynamic pressure reconstruction result are not considered when the dynamic pressure of the shock tube is reconstructed, so that the dynamic pressure amplitude estimation result lacks rationality and accuracy.

Disclosure of Invention

In view of the foregoing, it is necessary to provide a shock tube dynamic pressure reconstruction method, apparatus, electronic device and storage medium, which solve the technical problem in the prior art that the dynamic pressure amplitude estimation result lacks rationality and accuracy due to the fact that the fluctuation characteristics of the shock tube dynamic pressure signal and the impact vibration influence on the dynamic pressure reconstruction result are not considered when the shock tube dynamic pressure is reconstructed.

In order to solve the technical problems, in one aspect, the present invention provides a shock tube dynamic pressure reconstruction method, including:

acquiring an initial dynamic pressure response signal, wherein the initial dynamic pressure response signal comprises a vibration signal and a response signal;

preprocessing the vibration signal and the response signal based on a variation modal decomposition method and an empirical modal decomposition method to obtain a denoising vibration signal, a preprocessing response signal and component signals of the preprocessing response signal in different frequency bands;

Constructing a training set according to the correlation between the component signals of different frequency bands and the denoising vibration signal and the preprocessing response signal respectively;

constructing an initial inverse sensing network model based on the Bi-LSTM neural network model, and iteratively training the initial inverse sensing network model based on the training set to obtain a target inverse sensing network model;

and acquiring a real-time dynamic pressure response signal, inputting the real-time dynamic pressure response signal into the target inverse sensing network model after the preprocessing operation, and obtaining a target shock tube dynamic pressure reconstruction signal.

In some possible implementations, the preprocessing operation is performed on the vibration signal and the response signal based on the variation modal decomposition method and the empirical modal decomposition method to obtain a denoising vibration signal, a preprocessing response signal and component signals of the preprocessing response signal in different frequency bands, including:

decomposing the vibration signal based on a variation mode decomposition method to obtain a plurality of vibration mode components, respectively calculating correlation coefficients of the vibration mode components and the vibration signal, and removing high-frequency noise components to obtain the denoising vibration signal;

decomposing the response signals based on a variation modal decomposition method to obtain a plurality of response modal components, and reconstructing the plurality of response modal components based on sensor ringing frequency to obtain a plurality of reconstructed signals;

Decomposing the plurality of reconstructed signals based on an empirical mode decomposition method to obtain a plurality of reconstructed signal eigenmode function components;

calculating correlation coefficients of the plurality of reconstructed signal eigenmode function components and the denoising vibration signal respectively, wherein a reconstructed signal corresponding to the reconstructed signal eigenmode function component with the largest correlation coefficient of the denoising vibration signal is the preprocessing response signal;

and decomposing the preprocessing response signal based on an empirical mode decomposition method to obtain component signals of the preprocessing response signal in different frequency bands.

In some possible implementations, the constructing a training set according to the correlation between the component signals of the different frequency bands and the denoising vibration signal and the preprocessing response signal respectively includes:

calculating correlation coefficients between the component signals of different frequency bands and the preprocessing response signals and the denoising vibration signals respectively, and taking the component signal with the largest correlation coefficient with the preprocessing response signals in the component signals of different frequency bands as a ringing component signal;

taking a component signal, except for a ringing frequency component, of the component signals in different frequency bands, of which the correlation coefficient value with a denoising vibration signal is smaller than a set threshold value as a noise component signal, and removing the noise component signal to obtain a denoising response signal, a reconstruction response signal, a vibration correlation component signal and a trend signal;

The training set is constructed based on the reconstructed response signal, the vibration related component signal, and the trend component signal.

In some possible implementations, the trend component signal corresponds to the amplitude of the target shock tube dynamic pressure reconstruction signal, which can embody the amplitude characteristic of the target shock tube dynamic pressure reconstruction signal.

In some possible implementations, the training the initial inverse sensor network model based on the training set iteration to obtain a target inverse sensor network model includes:

constructing a first training set input of an initial inverse sensing network model based on the reconstruction response signal, constructing a second training set input of the initial inverse sensing network model based on the vibration related component, and outputting the training set of the initial inverse sensing network model based on the trend component signal;

after setting the hidden layer number and super parameters of the initial inverse sensor network model, carrying out iterative training on the initial inverse sensor network model based on the constructed training set;

and when the output loss rate of the initial inverse sensor network model is lower than a set loss threshold value, obtaining the target inverse sensor network model.

In some possible implementations, after setting the hidden layer number and the super parameter of the initial inverse sensor network model, performing iterative training on the initial inverse sensor network model based on the constructed training set, including:

Setting an initial hidden layer number of the initial inverse sensing network model, wherein the hidden layer comprises a plurality of neuron units;

setting super parameters of the initial inverse sensor network model, wherein the super parameters comprise: optimizer parameters, learning rate, sequence length and training rounds;

and in the iterative training process, the initial hidden layer number, the optimizer parameters, the learning rate, the sequence length and the training rounds of the initial inverse sensor network model are adjusted, the weight and the bias of each neuron unit node in the hidden layer are determined, the output root mean square error of the initial inverse sensor network model is enabled to be minimum, the set loss threshold is reached, and the iterative training process of the initial inverse sensor network model is completed.

In some possible implementations, the inputting the real-time dynamic pressure response signal and the preprocessing operation into the target inverse sensor network model to obtain a target shock tube dynamic pressure reconstruction signal includes:

acquiring a real-time dynamic pressure response signal of the shock tube sensor, and performing the preprocessing operation on the real-time response signal to obtain a denoising vibration signal and a denoising response signal corresponding to the real-time dynamic pressure response signal;

And inputting the denoising vibration signal and the denoising response signal corresponding to the real-time dynamic pressure response signal into the target inverse sensing network model to obtain target inverse sensing network model output, and dividing the output by the amplification factor and the sensitivity of the pressure sensor to obtain the target shock tube dynamic pressure reconstruction signal.

On the other hand, the invention also provides a shock tube dynamic pressure reconstruction device, which comprises:

the signal acquisition module is used for acquiring an initial dynamic pressure response signal, wherein the initial dynamic pressure response signal comprises a vibration signal and a response signal;

the signal processing module is used for preprocessing the vibration signal and the response signal based on a variation modal decomposition method and an empirical modal decomposition method to obtain a denoising vibration signal, a preprocessing response signal and component signals of the preprocessing response signal in different frequency bands;

the signal construction module is used for constructing a training set according to the correlation between the component signals of the different frequency bands and the denoising vibration signal and the preprocessing response signal respectively;

the model training module is used for constructing an initial inverse sensing network model based on the Bi-LSTM neural network model, and iteratively training the initial inverse sensing network model based on the training set to obtain a target inverse sensing network model;

And the target reconstruction module is used for acquiring a real-time dynamic pressure response signal, inputting the real-time dynamic pressure response signal into the target inverse sensing network model after the preprocessing operation, and obtaining a target shock tube dynamic pressure reconstruction signal.

On the other hand, the invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the shock tube dynamic pressure reconstruction method in the implementation mode is realized when the processor executes the program.

Finally, the invention also provides a computer readable storage medium, on which a computer program is stored, which when being executed by a processor, implements the shock tube dynamic pressure reconstruction method described in the above implementation mode.

The beneficial effects of adopting the embodiment are as follows: according to the shock tube dynamic pressure reconstruction method provided by the invention, on one hand, the response signals and the vibration signals are decomposed and reconstructed by using the variation mode decomposition method and the empirical mode decomposition method, the correlation among the component signals, the response signals and the vibration signals is comprehensively considered, and meanwhile, the fluctuation of the actual dynamic pressure amplitude generated by the shock tube along with time and the influence of the vibration signals are considered, so that the preprocessed data has higher rationality and accuracy, and on the other hand, the inverse sensing network model constructed based on the Bi-LSTM neural network is used for reconstructing the dynamic pressure of the shock tube by adopting the bidirectional input, so that the extraction of the signal characteristics by the model is finer, and the rationality and accuracy of the reconstructed dynamic pressure of the shock tube are further improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of an embodiment of a shock tube dynamic pressure reconstruction method according to the present invention;

FIG. 2 is a flowchart illustrating an embodiment of step S102 in FIG. 1 according to the present invention;

FIG. 3 is a flowchart illustrating an embodiment of step S103 in FIG. 1 according to the present invention;

fig. 4 is a flowchart of an embodiment of step S104 in fig. 1 according to the present invention;

FIG. 5 is a diagram of an embodiment of raw measurement data of a sensor according to the present invention;

FIG. 6 is a diagram illustrating an embodiment of constructing training set data according to the present invention;

FIG. 7 is a schematic diagram of an embodiment of a model training result provided by the present invention;

FIG. 8 is a schematic diagram of an embodiment of a dynamic pressure reconstruction signal provided by the present invention;

FIG. 9 is a schematic flow chart diagram of an embodiment of a shock tube dynamic pressure reconstruction device according to the present invention;

Fig. 10 is a schematic structural diagram of an embodiment of an electronic device according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this disclosure, illustrates operations implemented according to some embodiments of the present invention. It should be appreciated that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Moreover, one or more other operations may be added to or removed from the flow diagrams by those skilled in the art under the direction of the present disclosure.

Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor systems and/or microcontroller systems.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

Before the description of the embodiments, the related words are interpreted:

shock tube: shock tubes are the core device for pressure sensor calibration to generate planar shock waves. Shock waves are those in which the pressure of a gas changes suddenly at a place and the pressure waves propagate at a high speed. The velocity of the wave is related to the intensity of the pressure change, and the greater the pressure change, the higher the wave velocity. During the propagation process, the wavefront reaches a point where abrupt changes occur in gas pressure, density and temperature; the wave front is not everywhere, and the gas is not disturbed by the wave; after the wave front, the temperature and pressure of the gas behind the wave front are higher than those in front of the wave front, and the gas particles flow in the advancing direction of the wave front at a speed lower than the advancing speed of the wave front.

Bi-LSTM neural network: the Bi-LSTM neural network is a double-fed neural network, and is composed of two independent long-short-term memory (LSTM) neural networks, and input sequences are respectively input into the two long-short-term memory networks in positive sequence and reverse sequence, and then feature extraction is carried out.

The variation modal decomposition method comprises the following steps: (Variational mode decomposition, VMD) is a method of adaptive, completely non-recursive modal variation and signal processing. The technology has the advantages that the number of modal decomposition can be determined, the adaptivity is represented by determining the number of modal decomposition of a given sequence according to actual conditions, the optimal center frequency and the limited bandwidth of each modal can be adaptively matched in the subsequent searching and solving process, the effective separation of inherent modal components (IMFs), the frequency domain division of signals and further the effective decomposition components of given signals can be realized, and finally the optimal solution of the variation problem is obtained. The method solves the problems of end effect and modal component aliasing of an EMD method, has a firmer mathematical theory basis, can reduce the time sequence non-stationarity with high complexity and strong nonlinearity, decomposes to obtain a relatively stable subsequence containing a plurality of different frequency scales, is suitable for the sequence of the non-stationarity, and the core idea of the VMD is to construct and solve the variational problem.

Empirical mode decomposition method: (Empirical Mode Decomposition, EMD) is a time-frequency domain signal processing method that performs signal decomposition according to the time-scale characteristics of the data itself, without presetting any basis functions. The EMD has obvious advantages in processing non-stationary and non-linear data, is suitable for analyzing a non-linear non-stationary signal sequence, and has higher signal-to-noise ratio. The key point of the method is empirical mode decomposition, so that a complex signal is decomposed into a limited number of eigenmode functions (Intrinsic Mode Function, IMF), and each decomposed IMF component contains local characteristic information of different time scales of the original signal.

Based on the description of the technical terms, the common reconstruction methods of dynamic pressure in the prior art are two kinds of theoretical calculation methods and inverse reconstruction methods. The theoretical calculation method establishes a theoretical amplitude calculation model of the dynamic pressure of the shock tube by utilizing the propagation characteristics of the incident shock wave in the shock tube and a motion shock wave formula, and obtains the amplitude of the dynamic pressure of the shock tube by measuring the initial temperature and the initial pressure of a low-pressure chamber of the shock tube and the propagation speed of the incident shock wave; the reverse reconstruction method comprises the steps of firstly carrying out trend estimation on a response signal of the pressure sensor, obtaining a stable section of the trend signal, further obtaining a stable value of the response signal, and finally dividing the stable value of the response signal by the sensitivity of the sensor and an acquisition amplification factor to obtain a dynamic pressure signal stable value of the shock tube. However, in the prior art, the fluctuation characteristics of the dynamic pressure signals of the shock tube and the influence of impact vibration on the dynamic pressure reconstruction result are not considered when the dynamic pressure of the shock tube is reconstructed, so that the dynamic pressure amplitude estimation result lacks rationality and accuracy.

The following detailed description of specific embodiments is provided, and it should be noted that the description order of the following embodiments is not to be taken as a limitation on the preferred order of the embodiments.

The embodiment of the invention provides a shock tube dynamic pressure reconstruction method, a shock tube dynamic pressure reconstruction device, electronic equipment and a storage medium.

Referring to fig. 1, fig. 1 is a schematic flow chart of an embodiment of a shock tube dynamic pressure reconstruction method according to the present invention, where the shock tube dynamic pressure reconstruction method includes:

s101, acquiring an initial dynamic pressure response signal, wherein the initial dynamic pressure response signal comprises a vibration signal and a response signal;

s102, preprocessing the vibration signal and the response signal based on a variation modal decomposition method and an empirical modal decomposition method to obtain a denoising vibration signal, a preprocessing response signal and component signals of the preprocessing response signal in different frequency bands;

s103, constructing a training set according to the correlation between the component signals of the different frequency bands and the denoising vibration signal and the preprocessing response signal respectively;

s104, constructing an initial inverse sensor network model based on the Bi-LSTM neural network model, and iteratively training the initial inverse sensor network model based on the training set to obtain a target inverse sensor network model;

S105, acquiring a real-time dynamic pressure response signal, inputting the real-time dynamic pressure response signal into the target inverse sensing network model after the preprocessing operation, and obtaining a target shock tube dynamic pressure reconstruction signal.

Compared with the prior art, the shock tube dynamic pressure reconstruction method provided by the embodiment of the invention has the advantages that on one hand, the response signals and the vibration signals are decomposed and reconstructed by utilizing the variation modal decomposition method and the empirical modal decomposition method, the correlation among the component signals, the response signals and the vibration signals is comprehensively considered, and meanwhile, the fluctuation of the actual dynamic pressure amplitude generated by the shock tube along with time and the influence of the vibration signals are considered, so that the preprocessed data has higher rationality and accuracy, and on the other hand, the reverse sensing network model constructed based on the Bi-LSTM neural network is used for reconstructing the dynamic pressure of the shock tube by adopting the bidirectional input, so that the model is more careful in extracting the signal characteristics, and the rationality and accuracy of the shock tube reconstruction dynamic pressure are further improved.

Further, in some embodiments of the present invention, in step S101, when an initial response signal of the shock tube sensor is obtained, since impact vibration is generated when an incident shock wave reaches the end surface of the low pressure chamber during the operation of the shock tube, the pressure sensor mounted on the end surface is simultaneously excited by a dynamic pressure signal and a vibration signal, and the collected output signal of the pressure sensor is a mixed signal of the dynamic pressure response and the vibration response, and the vibration signal can be collected separately by using the acceleration sensor, so that the initial response signal includes the vibration signal and the response signal.

In step S105, the real-time response signal of the shock tube sensor is obtained and preprocessed, and the real-time response signal is preprocessed by the empirical mode decomposition method and the variation mode decomposition method in step S102.

Further, in some embodiments of the present invention, as shown in fig. 2, fig. 2 is a flowchart of an embodiment of step S102 in fig. 1 provided in the present invention, where step S102 includes:

s201, decomposing the vibration signal based on a variation modal decomposition method to obtain a plurality of vibration modal components, respectively calculating correlation coefficients of the plurality of vibration modal components and the vibration signal, and removing high-frequency noise components to obtain the denoising vibration signal;

s202, decomposing the response signals based on a variation modal decomposition method to obtain a plurality of response modal components, and reconstructing the plurality of response modal components based on sensor ringing frequency to obtain a plurality of reconstructed signals;

s203, decomposing the plurality of reconstruction signals based on an empirical mode decomposition method to obtain a plurality of reconstruction signal eigenmode function components;

s204, respectively calculating correlation coefficients of the plurality of reconstructed signal eigenmode function components and the denoising vibration signal, wherein a reconstructed signal corresponding to the reconstructed signal eigenmode function component with the largest correlation coefficient of the denoising vibration signal is the preprocessing response signal;

S205, decomposing the preprocessing response signal based on an empirical mode decomposition method to obtain component signals of the preprocessing response signal in different frequency bands.

Further, as shown in fig. 3, fig. 3 is a flow chart of an embodiment of step S103 in fig. 1 provided in the present invention, where step S103 includes:

s301, respectively calculating correlation coefficients between the component signals of different frequency bands and the preprocessing response signals and the denoising vibration signals, and taking the component signal with the largest correlation coefficient with the preprocessing response signals in the component signals of different frequency bands as a ringing component signal;

s301, removing a noise component signal which is used as a noise component signal and has a correlation coefficient value smaller than a set threshold value with a noise-removing vibration signal except for a ringing frequency component in the component signals of different frequency bands, so as to obtain a noise-removing response signal, a reconstruction response signal, a vibration-related component signal and a trend signal;

s301, constructing the training set based on the reconstruction response signal, the vibration related component signal and the trend component signal.

The trend component signal corresponds to the amplitude of the dynamic pressure reconstruction signal of the target shock tube, and the amplitude characteristic of the dynamic pressure reconstruction signal of the target shock tube can be embodied.

In a specific embodiment of the invention, the vibration signal is first decomposed by using a variation mode decomposition method

Decomposing, VMD can make vibration signal +.>

Decomposition into k narrowband bimfs, denoted +.>

The center frequency is +.>

，/>

And->

The method can be obtained by solving the variational problem:

(1)

(2)

wherein, the liquid crystal display device comprises a liquid crystal display device,

representing deviation measure->

Representing dirac distribution function, +.>

The decomposed kth modal component and the center frequency are respectively corresponding.

Solving the constrained optimal by adopting a quadratic penalty function and a Lagrangian operatorThe resolution problem is solved, and the resolution mode is calculated by adopting an alternating direction method of Lagrange's operator

And its corresponding center frequency->

The following are provided:

(3)

(4)

in the method, in the process of the invention,

respectively correspond to->

Is a fourier transform of (a).

Each vibration mode component obtained

Can be expressed as:

(5)

in the method, in the process of the invention,

representing k modal components.

Obtaining a calculated vibration signal

Is +.>

The correlation coefficient CC between them is as follows:

(6)

in the method, in the process of the invention,

and->

A discrete signal representing the vibration signal and corresponding to the ith modal component; />

And->

Respectively indicate->

And->

Average value of (2); n is the signal length.

According to the meaning of the correlation coefficient, taking the component with the phase relation number smaller than 0.2 as a noise component signal, and removing the noise component signal to obtain a denoising vibration signal

This can be expressed as:

(7)

in the method, in the process of the invention,

is a high frequency noise component. />

Further, response signals are decomposed by using a variation mode decomposition method

Decomposition into K response modality components

Firstly, reconstructing several components with central frequency less than or equal to sensor ringing frequency to obtain base signal +.>

The method comprises the steps of carrying out a first treatment on the surface of the The residual components are added into the base signal in turn according to the central frequency to reconstruct the signal, the base signal is +.>

Reconstructing a signal

The expression is:

(8)

(9)

in the method, in the process of the invention,

；/>

；/>

representing the J-th reconstructed signal.

And (3) sequentially processing the reconstructed signals by adopting an empirical mode decomposition method to obtain a series of narrow-band components called eigenmode functions.

Wherein each eigenmode function has to fulfil the following two conditions: (1) The extreme point and the zero crossing point number are equal or at most one different on the whole data set; (2) At any time, the mean value of the upper envelope estimated by the local maximum point and the lower envelope estimated by the local minimum point is zero.

The basic steps of decomposition are as follows:

step (1): identifying reconstructed signals

Local minimum and local maximum points of (a);

step (2): by using cubic splineThe lines are respectively connected with all local minimum value points and local maximum value points to obtain

Lower envelope of->

And upper envelope->

The mean value of the upper envelope curve and the lower envelope curve is calculated as follows:

(10)

step (3): from the signal

Less->

The difference signal is obtained as:

(11)

if it is

Satisfying two conditions of the eigenmode function, +.>

Is->

Is a first eigenmode function component of (a); otherwise, let->

Repeating the calculation process k times from the step (1) to the step (3) until the +.>

Two conditions of the eigenmode function are satisfied, in this case +.>

Is +.>

The method comprises the following steps:

(12)

step (4): from reconstructing a signal

Less->

Obtaining residual signal->

Is that

(13)

Order the

Repeating the calculation process i times from the step (1) to the step (4), and obtaining the ith eigenmode function component as follows:

(14)

continuing the decomposition process until a final residual component

Becomes a monotonic function or contains only one extreme point, in this case from +.>

The eigenmode function components can no longer be resolved. The combination of formula (13) and formula (14) reconstructs the signal +.>

Expressed as:

(15)/>

thus, the signal is reconstructed

Is decomposed into an eigenmode function component IMF of the h reconstructed signals and a residual component, and the frequency bands of these components vary from high to low.

All component signals obtained by calculation empirical mode decomposition method

And denoising vibration signal- >

The correlation coefficient between the two signals is found out, and the corresponding reconstruction signal is the preprocessing signal +.>

. The following are provided:

(16)

in the method, in the process of the invention,

，/>

is->

Decomposed eigenmode component->

Is a trend component.

Based on the idea of keeping the vibration signal components to the greatest extent, the correlation coefficients between the imf (t) component signals and the preprocessing signals and the denoising vibration signals are calculated respectively and recorded as

And->

The method comprises the steps of carrying out a first treatment on the surface of the Will->

The component signal with the largest correlation coefficient is used as the ringing component signal, and the vibration signal component is reserved for the maximum>

In addition to the ringing frequency component, the component signal with the value less than 0.1 is taken as the noise component signal to be removed, and the reconstruction response signal of the pressure sensor is obtained>

And denoising response signal->

. The following are provided:

(17)

(18)

in the method, in the process of the invention,

for vibration-related component signals +.>

Wherein->

。

The embodiment of the invention decomposes and reconstructs the response signals and the vibration signals by utilizing the variation modal decomposition method and the empirical modal decomposition method, comprehensively considers the correlation between each component signal and the response signals and the vibration signals, and simultaneously considers the fluctuation of the actual dynamic pressure amplitude generated by the shock tube along with time and the influence of the vibration signals, so that the data obtained by preprocessing has higher rationality and accuracy

Further, in some embodiments of the present invention, as shown in fig. 4, fig. 4 is a flowchart of an embodiment of step S104 in fig. 1 provided in the present invention, and step S103 includes:

s401, constructing a first training set input of an initial inverse sensing network model based on the reconstruction response signal, constructing a second training set input of the initial inverse sensing network model based on the vibration related component, and outputting the training set of the initial inverse sensing network model based on the trend component signal;

s402, after setting the hidden layer number and super parameters of the initial inverse sensor network model, carrying out iterative training on the initial inverse sensor network model based on the constructed training set;

s403, when the output loss rate of the initial inverse sensor network model is lower than a set loss threshold value, obtaining the target inverse sensor network model.

The step S402 specifically includes:

In a specific embodiment of the invention, a variation modal decomposition method and an empirical modal decomposition method are carried out on a shock tube initial response signal to obtain related data required for constructing an inverse sensing network model training set, a first inverse sensing network training set input is constructed by a reconstruction response signal of a pressure sensor, a second inverse sensing network training set input is constructed by a vibration related component, and an inverse sensing network training set output is constructed by a trend component signal.

The method for establishing the Bi-LSTM neural network based initial inverse sensor network model mainly comprises three parts of hidden layer number setting, super parameter setting and learning training.

The hidden layers execute the internal information transfer function of the neural network, namely, the information transmitted by the input layer is processed by a plurality of hidden layers to obtain an output layer; the hidden layer is composed of a plurality of nerve units, and one working nerve unit is composed of a forgetting gate, an input gate, a temporary cell state, a cell state, an output gate and a hidden layer state. The forgetting gate determines how much of the cell unit state remains at the current moment, the input gate determines how much of the input of the network remains at the current moment, the temporary cell state and the input gate jointly act on the updating of the cell unit state, the cell state provides the updating of the next cell state, and the output gate jointly acts on the hidden layer state and the cell state to provide the updating of the hidden layer state of the next cell unit. The working principle is as follows:

(19)

(20)

(21)

(22)

(23)

(24)

Wherein:

、/>

respectively amnestic door, input door, temporary cell state, output door and hidden layer state, ++>

Is weight, & gt>

Is biased.

Setting super parameters, namely setting an optimizer, a learning rate, a sequence length, training rounds and the like;

the neural network learning is to determine weights and biases of all neural unit nodes in the hidden layer according to training samples, and the neural network training is to search proper weights and biases so as to minimize the output root mean square error; and when the model output loss rate is lower than a given loss threshold value, completing identification of the inverse sensing network model to obtain the target inverse sensing network model.

When the inverse sensing network model is established, super parameters in the inverse sensing network, including the number of layers of hidden units, learning rate, learning decline rate, optimizer, training times and the like, are adjusted to enable the model output loss rate to be lower than a set loss threshold value, so that identification of the inverse sensing network model is completed.

According to the embodiment of the invention, the reverse sensing network model is constructed based on the two-way long-short-term memory neural network, iterative training is carried out through the training set, the hidden layer number and super parameters of the neural network are determined, the loss rate of model output is reduced, and the accuracy and rationality of the dynamic pressure reconstruction of the shock tube are further improved.

Further, in some embodiments of the present invention, step S105 includes:

acquiring a real-time response signal of the shock tube sensor, and performing the preprocessing operation on the real-time response signal to obtain the denoising vibration signal and the denoising response signal corresponding to the real-time response signal;

and inputting the denoising vibration signal and the denoising response signal corresponding to the real-time response signal into the target inverse sensing network model to obtain target inverse sensing network model output, and dividing the output by the amplification factor and the sensitivity of the pressure sensor to obtain the target shock tube dynamic pressure reconstruction signal.

In a specific embodiment of the present invention, for a shock tube sensor response signal obtained in real time, the same empirical mode decomposition method and variation mode decomposition method as in the above embodiment are adopted to perform a preprocessing operation on the real-time response signal, and the obtained denoising response signal and denoising vibration signal are used as inputs of a target inverse sensing network model, and the obtained output of the model is divided by the amplification factor and sensitivity of the pressure sensor to obtain a target shock tube dynamic pressure reconstruction signal, where the relationship between the dynamic pressure reconstruction signal and the output of the inverse sensing network model is:

(24)

In the method, in the process of the invention,

for the amplification factor, S is sensitivity.

According to the embodiment of the invention, the reverse sensing network model constructed based on the Bi-LSTM neural network is used for reconstructing the dynamic pressure of the shock tube by adopting bidirectional input, so that the extraction of the signal characteristics by the model is finer, and the rationality and accuracy of the reconstructed dynamic pressure of the shock tube are further improved.

In order to more intuitively embody the rationality and accuracy of the shock tube dynamic pressure reconstruction method provided by the invention, the dynamic pressure response data and vibration signal data generated by a shock tube system are measured and analyzed by an ENDEVCO 8510B PR pressure sensor and a YA1102 ICP type acceleration sensor, wherein the sensitivity of the pressure sensor is 0.16V/MPa, the amplification factor is 50, the sampling frequency of the data is 5MHz, and the dynamic pressure reconstruction is carried out:

raw measurement data of the ENDEVCO 8510B PR pressure sensor and the YA1102 ICP type acceleration sensor are shown in FIG. 5, and FIG. 5 is a schematic diagram of an embodiment of the raw measurement data of the sensor according to the present invention.

The response signal and the vibration signal in fig. 5 are processed by using the same empirical mode decomposition method and the variation mode decomposition method in the above embodiment, so as to obtain an inverse sensing network training set as shown in fig. 6, and fig. 6 is a schematic diagram of an embodiment of constructing training set data provided by the present invention.

The training result of the target inverse sensor network model obtained through the training set iterative training in fig. 6 is shown in fig. 7, and fig. 7 is a schematic diagram of an embodiment of the model training result provided by the invention.

The dynamic pressure reconstruction signal obtained by using the target inverse sensor network model obtained in fig. 7 is shown in fig. 8, and fig. 8 is a schematic diagram of an embodiment of the dynamic pressure reconstruction signal provided by the present invention.

From fig. 5 to fig. 8, it can be seen that, according to the shock tube dynamic pressure reconstruction method provided by the embodiment of the present invention, the shock tube dynamic pressure reconstruction signal considers the influence of the vibration signal, and simultaneously considers the characteristic that the dynamic pressure amplitude of the shock tube changes with time, so that the shock tube dynamic pressure reconstruction signal can more reasonably and accurately restore the dynamic pressure signal generated by the shock tube.

In order to better implement the shock tube dynamic pressure reconstruction method in the embodiment of the present invention, correspondingly, on the basis of the shock tube dynamic pressure reconstruction method, the embodiment of the present invention further provides a shock tube dynamic pressure reconstruction device, as shown in fig. 9, where the shock tube dynamic pressure reconstruction device 900 includes:

a signal acquisition module 901, configured to acquire an initial dynamic pressure response signal, where the initial dynamic pressure response signal includes a vibration signal and a response signal;

The signal processing module 902 is configured to perform a preprocessing operation on the vibration signal and the response signal based on a variation modal decomposition method and an empirical modal decomposition method, so as to obtain a denoising vibration signal, a preprocessing response signal, and component signals of the preprocessing response signal in different frequency bands;

the signal construction module 903 is configured to construct a training set according to correlations between the component signals in the different frequency bands and the denoising vibration signal and the preprocessing response signal, respectively;

the model training module 904 is configured to construct an initial inverse sensor network model based on the Bi-LSTM neural network model, and iteratively train the initial inverse sensor network model based on the training set to obtain a target inverse sensor network model;

the target reconstruction module 905 is configured to obtain a real-time dynamic pressure response signal, perform the preprocessing operation, and input the real-time dynamic pressure response signal into the target inverse sensor network model to obtain a target shock tube dynamic pressure reconstruction signal.

The shock tube dynamic pressure reconstruction device 900 provided in the foregoing embodiment may implement the technical solution described in the foregoing shock tube dynamic pressure reconstruction method embodiment, and the specific implementation principle of each module or unit may refer to the corresponding content in the foregoing shock tube dynamic pressure reconstruction method embodiment, which is not described herein again.

As shown in fig. 10, the present invention further provides an electronic device 1000 accordingly. The electronic device 1000 comprises a processor 1001, a memory 1002 and a display 1003. Fig. 10 shows only some of the components of the electronic device 1000, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead.

The processor 1001 may in some embodiments be a central processing unit (CentralProcessing Unit, CPU), microprocessor or other data processing chip for executing program code or processing data stored in the memory 1002, such as a shock tube dynamic pressure reconstruction program in the present invention.

In some embodiments, the processor 1001 may be a single server or a group of servers. The server farm may be centralized or distributed. In some embodiments, the processor 1001 may be local or remote. In some embodiments, the processor 1001 may be implemented in a cloud platform. In an embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-internal, multiple clouds, or the like, or any combination thereof.

The memory 1002 may be an internal storage unit of the electronic device 1000 in some embodiments, such as a hard disk or memory of the electronic device 1000. The memory 1002 may also be an external storage device of the electronic device 1000 in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device 1000.

Further, the memory 1002 may also include both internal storage units and external storage devices of the electronic device 1000. The memory 1002 is used for storing application software and various types of data for installing the electronic device 1000.

The display 1003 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like in some embodiments. The display 1003 is used for displaying information at the electronic device 1000 and for displaying a visualized user interface. The components 1001-1003 of the electronic device 1000 communicate with each other over a system bus.

In one embodiment, when the processor 1001 executes the shock tube dynamic pressure reconstruction program in the memory 1002, the following steps may be implemented:

It should be understood that: the processor 1001 may perform other functions in addition to the above functions when executing the shock tube dynamic pressure reconstruction program in the memory 1002, see in particular the description of the corresponding method embodiments above.

Further, the type of the electronic device 1000 is not particularly limited, and the electronic device 1000 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (personal digital assistant, PDA), a wearable device, a laptop (laptop), etc. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices that carry IOS, android, microsoft or other operating systems. The portable electronic device described above may also be other portable electronic devices, such as a laptop computer (laptop) or the like having a touch-sensitive surface, e.g. a touch panel. It should also be appreciated that in other embodiments of the invention, the electronic device 1000 may not be a portable electronic device, but rather a desktop computer having a touch-sensitive surface (e.g., a touch panel).

Correspondingly, the embodiment of the application also provides a computer readable storage medium, which is used for storing a computer readable program or instruction, and when the program or instruction is executed by a processor, the steps or functions in the shock tube dynamic pressure reconstruction method provided by the embodiments of the method can be realized.

Those skilled in the art will appreciate that all or part of the flow of the methods of the embodiments described above may be accomplished by way of a computer program stored in a computer readable storage medium to instruct related hardware (e.g., a processor, a controller, etc.). The computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory.

The shock tube dynamic pressure reconstruction method, the shock tube dynamic pressure reconstruction device, the electronic equipment and the storage medium provided by the invention are described in detail, specific examples are applied to the explanation of the principle and the implementation mode of the invention, and the explanation of the above examples is only used for helping to understand the method and the core idea of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present invention, the present description should not be construed as limiting the present invention.

Claims

1. The shock tube dynamic pressure reconstruction method is characterized by comprising the following steps of:

acquiring a real-time dynamic pressure response signal, inputting the real-time dynamic pressure response signal into the target inverse sensing network model after the preprocessing operation, and obtaining a target shock tube dynamic pressure reconstruction signal;

the preprocessing operation is performed on the vibration signal and the response signal based on the variation modal decomposition method and the empirical modal decomposition method to obtain a denoising vibration signal, a preprocessing response signal and component signals of the preprocessing response signal in different frequency bands, and the preprocessing method comprises the following steps:

decomposing the preprocessing response signal based on an empirical mode decomposition method to obtain component signals of the preprocessing response signal in different frequency bands;

the constructing a training set according to the correlation between the component signals of the different frequency bands and the denoising vibration signal and the preprocessing response signal respectively comprises the following steps:

taking a component signal, except for a ringing frequency component, of the component signals in different frequency bands, of which the correlation coefficient value with a denoising vibration signal is smaller than a set threshold value as a noise component signal, and removing the noise component signal to obtain a denoising response signal, a reconstruction response signal, a vibration correlation component signal and a trend component signal;

constructing the training set based on the reconstructed response signal, the vibration-related component signal, and the trend component signal;

the training set based on the training iteration training of the initial inverse sensor network model, to obtain a target inverse sensor network model, includes:

constructing a first training set input of an initial inverse sensing network model based on the reconstruction response signal, constructing a second training set input of the initial inverse sensing network model based on the vibration related component, and taking the trend component signal as an output of the initial inverse sensing network model;

2. The shock tube dynamic pressure reconstruction method of claim 1 wherein the trend component signal corresponds to the target shock tube dynamic pressure reconstruction signal amplitude.

3. The shock tube dynamic pressure reconstruction method according to claim 1, wherein after setting the hidden layer number and the super parameter of the initial inverse sensor network model, performing iterative training on the initial inverse sensor network model based on the constructed training set comprises:

4. The shock tube dynamic pressure reconstruction method as set forth in claim 3 wherein said obtaining a dynamic pressure real-time response signal and inputting said pre-processing operation into said target inverse sensor network model to obtain a target shock tube dynamic pressure reconstruction signal comprises:

5. A shock tube dynamic pressure reconstruction device, comprising:

the target reconstruction module is used for acquiring a real-time dynamic pressure response signal, inputting the real-time dynamic pressure response signal into the target inverse sensing network model after the preprocessing operation, and obtaining a target shock tube dynamic pressure reconstruction signal;

6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the shock tube dynamic pressure reconstruction method according to any one of claims 1 to 4 when the program is executed by the processor.

7. A computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements a shock tube dynamic pressure reconstruction method according to any one of claims 1 to 4.