Bolt axial force detection method and device and electronic equipment

Abstract

The invention relates to the field of ultrasonic nondestructive testing, and provides a bolt axial force detection method and device and electronic equipment, and the method comprises the steps: receiving a trigger signal, and transmitting a pulse excitation signal to a piezoelectric probe and / or an electromagnetic probe through a processor based on the trigger signal; a digital signal fed back by the processor is received, the digital signal is input into a pre-constructed detection model, the digital signal is obtained by the processor based on an echo signal, the echo signal is generated by the piezoelectric probe and / or the electromagnetic probe under the triggering of the pulse excitation signal, and the echo signal comprises bolt axial force information collected by the piezoelectric probe and / or the electromagnetic probe; the detection model is a neural network model; and converting the digital signal into bolt axial force information based on the detection model. The bolt axial force detection method and device are used for solving the technical problem that in the prior art, various detection requirements and high-frequency field switching scenes cannot be met, a piezoelectric probe and / or an electromagnetic probe can be applied to bolt axial force detection from multiple dimensions, and various detection requirements can be met.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic non-destructive testing, and in particular to a method and device for detecting bolt axial force, and electronic equipment. Background Art

[0002] Bolts are an important connecting element in steel structure engineering, responsible for transmitting and distributing loads. Bolt axial force is a key factor in evaluating bolt connection performance.

[0003] Bolt axial force is a force acting along the bolt axis. The magnitude and direction of the bolt axial force directly affect the stability and safety of the bolt connection. Therefore, it is of great significance to detect the bolt axial force.

[0004] In related technologies, there are multiple detection methods for bolt axial force detection, but the current technical solutions still cannot meet various detection needs and high-frequency on-site switching scenarios. Summary of the Invention

[0005] The present invention provides a bolt axial force detection method and device, and electronic equipment to solve the defects of the solutions in the related art that cannot meet various detection needs and high-frequency on-site switching scenarios. The solution of this application can use piezoelectric probes and / or electromagnetic probes to detect bolt axial force from multiple dimensions, which can meet various detection needs.

[0006] The present invention provides a method for detecting the axial force of a bolt, comprising:

[0007] receiving a trigger signal, and sending a pulse excitation signal to a piezoelectric probe and / or an electromagnetic probe through a processor based on the trigger signal, wherein the piezoelectric probe and the electromagnetic probe are arranged at the end of the bolt to be tested;

[0008] receiving a digital signal fed back by the processor and inputting the digital signal into a pre-built detection model, wherein the digital signal is obtained by the processor based on an echo signal, the echo signal is generated by the piezoelectric probe and / or the electromagnetic probe when triggered by the pulse excitation signal, the echo signal includes bolt axial force information collected by the piezoelectric probe and / or the electromagnetic probe, and the detection model is a neural network model;

[0009] The digital signal is converted into bolt axial force information based on the detection model.

[0010] According to the bolt axial force detection method provided by the present invention, the detection model includes a one-dimensional convolution layer, a memory network layer and a feature fusion layer;

[0011] The one-dimensional convolutional layer conforms to the following formula (1):

[0012]

[0013] Among them, * is a one-dimensional convolution operation, l is the layer index, c is the output channel, is the output feature map of the lth convolutional layer, is the convolution kernel weight of the lth convolution layer, is the bias term of the lth convolutional layer; BatchNorm is the batch normalization operation;

[0014] The memory network layer complies with the following formula (2):

[0015]

[0016] in, is the hidden state of the l-th layer LSTM at time step t, is the cell state of the l-th layer LSTM at time step t, is the final output feature of the previous layer;

[0017] The feature fusion layer complies with the following formula (3):

[0018]

[0019] Among them, h p is the fused feature vector, is the final temporal feature extracted from the BiLSTM network, Δt ML The sound time difference predicted by machine learning, T comp is the temperature compensation value;

[0020] According to the bolt axial force detection method provided by the present invention, when the detection model is constructed, the model evaluation is performed using the following loss function:

[0021]

[0022] in, is the total loss function, F pred is the bolt axial force value predicted by the model, F true is the actual measured axial force value, λ is the trade-off coefficient used to control the importance of the physical constraint term, E is the elastic modulus of the material, A is the cross-sectional area of the bolt, L is the bolt length, Δt ML is the acoustic time difference predicted by the machine learning model.

[0023] According to the bolt axial force detection method provided by the present invention, the digital signal is converted into bolt axial force information based on the detection model, and then the method further includes:

[0024] Displaying and storing the bolt axial force information;

[0025] The bolt axial force information is input into the detection model to train the detection model.

[0026] According to the bolt axial force detection method provided by the present invention, the pulse excitation signal includes a first pulse excitation signal and / or a second pulse excitation signal;

[0027] The first pulse excitation signal is used to control the piezoelectric probe to collect piezoelectric information of the bolt to be tested;

[0028] The second pulse excitation signal is used to control the electromagnetic probe to collect electromagnetic information of the bolt to be tested.

[0029] The present invention also provides a bolt axial force detection device, comprising a PC, a controller, a piezoelectric probe and an electromagnetic probe;

[0030] Wherein, the PC is used to execute any of the above-mentioned bolt axial force detection methods;

[0031] a controller, configured to send a pulse excitation signal to the piezoelectric probe and / or the electromagnetic probe under the triggering of the PC, receive an echo signal sent by the piezoelectric probe and / or the electromagnetic probe, and convert the echo signal into a digital signal;

[0032] A piezoelectric probe is used to receive a pulse excitation signal, collect piezoelectric information of the bolt to be tested under the triggering of the pulse excitation signal, and send the piezoelectric information to the controller;

[0033] The electromagnetic probe is used to receive the pulse excitation signal, collect electromagnetic information of the bolt to be tested under the triggering of the pulse excitation signal, and send the electromagnetic information to the controller.

[0034] According to the bolt axial force detection device provided by the present invention, the piezoelectric probe includes a first piezoelectric probe and a second piezoelectric probe;

[0035] The first piezoelectric probe is used to receive a pulse excitation signal, and under the triggering of the pulse excitation signal, collect shear wave information of the bolt to be tested, and send the shear wave information to the controller;

[0036] The second piezoelectric probe is used to receive a pulse excitation signal, and under the triggering of the pulse excitation signal, collect longitudinal wave information of the bolt to be tested, and send the longitudinal wave information to the controller.

[0037] The bolt axial force detection device provided by the present invention further includes a data transceiver interface;

[0038] The data transceiver interface is used to receive the pulse excitation signal sent by the controller, and send the pulse excitation signal to the piezoelectric probe and / or the electromagnetic probe;

[0039] and receiving the echo signal sent by the piezoelectric probe and / or the electromagnetic probe, and sending the echo signal to the controller.

[0040] The bolt axial force detection device provided according to the present invention further includes a channel switching device, which is used to merge the shear wave signal collected by the first piezoelectric probe and the longitudinal wave signal collected by the second piezoelectric probe.

[0041] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, any of the above-mentioned bolt axial force detection methods is implemented.

[0042] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, any of the above-mentioned bolt axial force detection methods is implemented.

[0043] The present invention also provides a computer program product, comprising a computer program, which implements any of the above-mentioned bolt axial force detection methods when executed by a processor.

[0044] In the bolt axial force detection method provided by the present invention, a pulse excitation signal can be sent to the piezoelectric probe and / or the electromagnetic probe. That is to say, in the scheme of the present application, the two probes can be controlled simultaneously to detect the bolt to be tested, or the two probes can be controlled separately to detect the bolt to be tested, which can meet various detection needs and high-frequency on-site switching scenarios. After the piezoelectric probe and / or the electromagnetic probe completes the detection of the bolt to be tested, the collected information can be sent back to the processor in the form of an echo signal. Furthermore, the processor sends the processed echo signal to the PC, and the PC can convert the digital signal feedback from the processor to obtain the bolt axial force information corresponding to the bolt to be tested. In this process, the PC can apply a neural network model to analyze and convert the digital signal, and the accuracy is also higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 1 is a flow chart of a method for detecting bolt axial force provided by an embodiment of the present invention;

[0047] Figure 21 is a schematic structural diagram of a bolt axial force detection device provided by an embodiment of the present invention;

[0048] Figure 3 It is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0050] Figure 1 It is a flow chart of a method for detecting bolt axial force provided by an embodiment of the present invention.

[0051] like Figure 1 As shown, this embodiment provides a method for detecting bolt axial force, including:

[0052] Step 101: receiving a trigger signal and sending a pulse excitation signal to a piezoelectric probe and / or an electromagnetic probe via a processor based on the trigger signal, wherein the piezoelectric probe and the electromagnetic probe are arranged at the end of a bolt to be tested;

[0053] During implementation, the bolt axial force detection solution provided in this application can be executed by a PC, a tablet computer, etc.

[0054] In actual applications, the trigger signal in this step may refer to a command sent by the user to an execution entity such as a PC or tablet computer through a key or touch screen. After receiving the user's command, the execution entity may start executing the bolt axial force detection method.

[0055] The piezoelectric probe in the above steps can be equipped with a piezoelectric chip inside. After receiving the pulse excitation signal, the piezoelectric probe can generate an ultrasonic signal to pass through the bolt. After passing through the entire bolt, the ultrasonic signal reaches the bottom surface and is reflected back, thereby obtaining piezoelectric information that can reflect the axial force information of the bolt to be tested.

[0056] The electromagnetic probe may include an induction coil or a magnetic core. When the electromagnetic probe receives a pulse excitation signal, it may generate a rapidly changing strong magnetic field, thereby inducing eddy currents in the bolt to be tested. The electromagnetic probe receives the secondary magnetic field generated by the eddy currents and extracts electromagnetic information that can reflect the axial force information of the bolt to be tested.

[0057] The pulse excitation signal can be generated by a processor, and the processor can be FPGA+ARM.

[0058] The pulse excitation signal sent by the processor in this step can control the piezoelectric probe or the electromagnetic probe to collect information independently, or can control the piezoelectric probe and the electromagnetic probe to collect information simultaneously.

[0059] Step 102: Receive a digital signal fed back by the processor and input the digital signal into a pre-built detection model. The digital signal is obtained by the processor based on an echo signal. The echo signal is generated by the piezoelectric probe and / or the electromagnetic probe when triggered by the pulse excitation signal. The echo signal includes bolt axial force information collected by the piezoelectric probe and / or the electromagnetic probe. The detection model is a neural network model.

[0060] In actual applications, after the processor sends a pulse excitation signal to the piezoelectric probe and / or electromagnetic probe, it can also receive the echo signal sent by the piezoelectric probe and / or electromagnetic probe. The echo signal contains the piezoelectric information and / or electromagnetic information collected by the piezoelectric probe and / or electromagnetic probe. The processor can use the preamplifier circuit to process it and then convert the echo signal into a digital signal. The preamplifier circuit is also called a preamplifier. Its main function is to amplify and preliminarily filter the weak echo signal. After that, the echo signal can be converted into a digital signal through analog-to-digital conversion (ADC) and digital signal processing (DSP) for further analysis and processing.

[0061] Step 103: Convert the digital signal into bolt axial force information based on the detection model.

[0062] During implementation, the processor converts the echo signal into a digital signal and feeds it back to the PC. The PC stores a pre-built neural network model, which can further process the digital signal. Specifically, the pre-built neural network model is used to convert the digital signal into bolt axial force information, thereby achieving the effect of bolt axial force detection by controlling the piezoelectric probe and / or electromagnetic probe through the PC.

[0063] In the bolt axial force detection method provided in this embodiment, a pulse excitation signal can be sent to the piezoelectric probe and / or the electromagnetic probe. That is to say, in the solution of the present application, the two probes can be controlled simultaneously to detect the bolt to be tested, or the two probes can be controlled separately to detect the bolt to be tested, which can meet various detection needs and high-frequency on-site switching scenarios. After the piezoelectric probe and / or the electromagnetic probe completes the detection of the bolt to be tested, the collected information can be sent back to the processor in the form of an echo signal. Furthermore, the processor sends the processed echo signal to the PC, and the PC can convert the digital signal feedback from the processor to obtain the bolt axial force information corresponding to the bolt to be tested. In this process, the PC can apply a neural network model to analyze and convert the digital signal, and the accuracy is also higher.

[0064] In an exemplary embodiment, the detection model includes a one-dimensional convolution layer, a memory network layer, and a feature fusion layer;

[0065] The one-dimensional convolutional layer conforms to the following formula (1):

[0066]

[0067] Among them, * is a one-dimensional convolution operation, l is the layer index, c is the output channel, is the output feature map of the lth convolutional layer, is the convolution kernel weight of the lth convolution layer, is the bias term of the lth convolutional layer; BatchNorm is the batch normalization operation;

[0068] The memory network layer complies with the following formula (2):

[0069]

[0070] in, is the hidden state of the l-th layer LSTM at time step t, is the cell state of the l-th layer LSTM at time step t, is the final output feature of the previous layer;

[0071] The feature fusion layer complies with the following formula (3):

[0072]

[0073] Among them, h p is the fused feature vector, is the final temporal feature extracted from the BiLSTM network, Δt ML The sound time difference predicted by machine learning, T comp is the temperature compensation value;

[0074] The monitoring model provided in this embodiment can adopt a hybrid neural network architecture. The one-dimensional convolution layer can extract local wave characteristics, including the rising edge and attenuation characteristics of the echo. The memory network layer uses a bidirectional long short-term memory network (BiLSTM) to capture timing dependencies. Physical features can also be fused through the feature fusion layer, and physical measurement values such as acoustic time difference and temperature can be explicitly integrated to improve interpretability.

[0075] In an exemplary embodiment, when the detection model is constructed, the model is evaluated using the following loss function:

[0076]

[0077] in, is the total loss function, F pred is the bolt axial force value predicted by the model, F true is the actual measured axial force value, λ is the trade-off coefficient used to control the importance of the physical constraint term, E is the elastic modulus of the material, A is the cross-sectional area of the bolt, L is the bolt length, Δt ML is the acoustic time difference predicted by the machine learning model.

[0078] In an exemplary embodiment, the converting the digital signal into bolt axial force information based on the detection model further includes:

[0079] Displaying and storing the bolt axial force information;

[0080] The bolt axial force information is input into the detection model to train the detection model.

[0081] In practical applications, bolt axial force information can be uploaded to a remote database for storage.

[0082] While storing the bolt axial force information, it can also be used to train the neural network model in the PC. By continuously training the neural network model, its accuracy can be improved.

[0083] In an exemplary embodiment, the pulse excitation signal includes a first pulse excitation signal and / or a second pulse excitation signal;

[0084] The first pulse excitation signal is used to control the piezoelectric probe to collect piezoelectric information of the bolt to be tested;

[0085] The second pulse excitation signal is used to control the electromagnetic probe to collect electromagnetic information of the bolt to be tested.

[0086] As described in the above embodiments, the pulse signal can control the piezoelectric probe or the electromagnetic probe alone to collect information on the bolt to be tested, or can control the piezoelectric probe and the electromagnetic probe at the same time to collect information on the bolt to be tested. Specifically, the pulse excitation signal can include a first pulse excitation signal and / or a second pulse excitation signal. When the piezoelectric probe is required to collect information alone, the processor can send the first pulse excitation signal to the piezoelectric probe alone. When the electromagnetic probe is required to collect information alone, the processor can send the second pulse excitation signal to the electromagnetic probe alone. When the piezoelectric probe and the electromagnetic probe are required to collect information on the bolt to be tested at the same time, the processor can send the first pulse excitation signal to the piezoelectric probe and the second pulse excitation signal to the electromagnetic probe at the same time.

[0087] In an exemplary embodiment, after the PC receives the digital signal fed back by the processor, it can also first display the waveform corresponding to the digital signal in real time, and mark the start gate and end gate in the waveform, based on which the first wave and the echo are identified.

[0088] The bolt axial force detection device provided by the present invention is described below. The bolt axial force detection device described below and the bolt axial force detection method described above can be referenced to each other.

[0089] Figure 2 It is a structural schematic diagram of a bolt axial force detection device provided in an embodiment of the present invention.

[0090] like Figure 2 As shown, the bolt axial force detection device provided by the present application includes a PC, a controller, a piezoelectric probe and an electromagnetic probe;

[0091] Wherein, PC is used to perform the bolt axial force detection method according to any one of claims 1 to 5;

[0092] a controller, configured to send a pulse excitation signal to the piezoelectric probe and / or the electromagnetic probe under the triggering of the PC, receive an echo signal sent by the piezoelectric probe and / or the electromagnetic probe, and convert the echo signal into a digital signal;

[0093] A piezoelectric probe is used to receive a pulse excitation signal, collect piezoelectric information of the bolt to be tested under the triggering of the pulse excitation signal, and send the piezoelectric information to the controller;

[0094] The electromagnetic probe is used to receive the pulse excitation signal, collect electromagnetic information of the bolt to be tested under the triggering of the pulse excitation signal, and send the electromagnetic information to the controller.

[0095] In an exemplary embodiment, the piezoelectric probe includes a first piezoelectric probe and a second piezoelectric probe;

[0096] The first piezoelectric probe is used to receive a pulse excitation signal, and under the triggering of the pulse excitation signal, collect shear wave information of the bolt to be tested, and send the shear wave information to the controller;

[0097] The second piezoelectric probe is used to receive a pulse excitation signal, and under the triggering of the pulse excitation signal, collect longitudinal wave information of the bolt to be tested, and send the longitudinal wave information to the controller.

[0098] In an exemplary embodiment, the bolt axial force detection device further includes a data transceiver interface;

[0099] The data transceiver interface is used to receive the pulse excitation signal sent by the controller, and send the pulse excitation signal to the piezoelectric probe and / or the electromagnetic probe;

[0100] And receive the echo signal sent by the piezoelectric probe and / or the electromagnetic probe, and send the echo signal to the controller.

[0101] In an exemplary embodiment, the bolt axial force detection device further includes a channel switching device, which is used to merge the shear wave signal collected by the first piezoelectric probe and the longitudinal wave signal collected by the second piezoelectric probe.

[0102] Figure 3 An example of a physical structure diagram of an electronic device is shown below. Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the bolt axial force detection method, which includes:

[0103] receiving a trigger signal, and sending a pulse excitation signal to a piezoelectric probe and / or an electromagnetic probe through a processor based on the trigger signal, wherein the piezoelectric probe and the electromagnetic probe are arranged at the end of the bolt to be tested;

[0104] receiving a digital signal fed back by the processor and inputting the digital signal into a pre-built detection model, wherein the digital signal is obtained by the processor based on an echo signal, the echo signal is generated by the piezoelectric probe and / or the electromagnetic probe when triggered by the pulse excitation signal, the echo signal includes bolt axial force information collected by the piezoelectric probe and / or the electromagnetic probe, and the detection model is a neural network model;

[0105] The digital signal is converted into bolt axial force information based on the detection model.

[0106] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0107] On the other hand, the present invention further provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the bolt axial force detection method provided by the above methods, which includes:

[0108] receiving a trigger signal, and sending a pulse excitation signal to a piezoelectric probe and / or an electromagnetic probe through a processor based on the trigger signal, wherein the piezoelectric probe and the electromagnetic probe are arranged at the end of the bolt to be tested;

[0109] receiving a digital signal fed back by the processor and inputting the digital signal into a pre-built detection model, wherein the digital signal is obtained by the processor based on an echo signal, the echo signal is generated by the piezoelectric probe and / or the electromagnetic probe when triggered by the pulse excitation signal, the echo signal includes bolt axial force information collected by the piezoelectric probe and / or the electromagnetic probe, and the detection model is a neural network model;

[0110] The digital signal is converted into bolt axial force information based on the detection model.

[0111] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for detecting the bolt axial force provided by the above methods is implemented, and the method includes:

[0112] receiving a trigger signal, and sending a pulse excitation signal to a piezoelectric probe and / or an electromagnetic probe through a processor based on the trigger signal, wherein the piezoelectric probe and the electromagnetic probe are arranged at the end of the bolt to be tested;

[0113] receiving a digital signal fed back by the processor and inputting the digital signal into a pre-built detection model, wherein the digital signal is obtained by the processor based on an echo signal, the echo signal is generated by the piezoelectric probe and / or the electromagnetic probe when triggered by the pulse excitation signal, the echo signal includes bolt axial force information collected by the piezoelectric probe and / or the electromagnetic probe, and the detection model is a neural network model;

[0114] The digital signal is converted into bolt axial force information based on the detection model.

[0115] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units. That is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0116] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus the necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or certain parts of the embodiment.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for detecting the axial force of a bolt, characterized in that: include: receiving a trigger signal, and sending a pulse excitation signal to a piezoelectric probe and / or an electromagnetic probe through a processor based on the trigger signal, wherein the piezoelectric probe and the electromagnetic probe are arranged at the end of the bolt to be tested; receiving a digital signal fed back by the processor and inputting the digital signal into a pre-built detection model, wherein the digital signal is obtained by the processor based on an echo signal, the echo signal is generated by the piezoelectric probe and / or the electromagnetic probe when triggered by the pulse excitation signal, the echo signal includes bolt axial force information collected by the piezoelectric probe and / or the electromagnetic probe, and the detection model is a neural network model; The digital signal is converted into bolt axial force information based on the detection model.

2. The bolt axial force detection method according to claim 1, characterized in that: The detection model includes a one-dimensional convolution layer, a memory network layer and a feature fusion layer; The one-dimensional convolutional layer conforms to the following formula (1): Among them, * is a one-dimensional convolution operation, l is the layer index, c is the output channel, is the output feature map of the lth convolutional layer, is the convolution kernel weight of the lth convolution layer, is the bias term of the lth convolutional layer; BatchNorm is the batch normalization operation; The memory network layer complies with the following formula (2): in, is the hidden state of the l-th layer LSTM at time step t, is the cell state of the l-th layer LSTM at time step t, is the final output feature of the previous layer; The feature fusion layer complies with the following formula (3): Among them, h p is the fused feature vector, is the final temporal feature extracted from the BiLSTM network, Δt ML The sound time difference predicted by machine learning, T comp is the temperature compensation value.

3. The bolt axial force detection method according to claim 1, characterized in that: When the detection model is constructed, the model is evaluated using the following loss function: in, is the total loss function, F pred is the bolt axial force value predicted by the model, F true is the actual measured axial force value, λ is the trade-off coefficient used to control the importance of the physical constraint term, E is the elastic modulus of the material, A is the cross-sectional area of the bolt, L is the bolt length, Δt ML is the acoustic time difference predicted by the machine learning model.

4. The bolt axial force detection method according to claim 1, characterized in that: The step of converting the digital signal into bolt axial force information based on the detection model further includes: Displaying and storing the bolt axial force information; The bolt axial force information is input into the detection model to train the detection model.

5. The bolt axial force detection method according to claim 1, characterized in that: The pulse excitation signal includes a first pulse excitation signal and / or a second pulse excitation signal; The first pulse excitation signal is used to control the piezoelectric probe to collect piezoelectric information of the bolt to be tested; The second pulse excitation signal is used to control the electromagnetic probe to collect electromagnetic information of the bolt to be tested.

6. Bolt axial force detection device, characterized in that: Includes PC, controller, piezoelectric probe and electromagnetic probe; Wherein, the PC is used to execute the bolt axial force detection method according to any one of claims 1 to 5; The controller is configured to send a pulse excitation signal to the piezoelectric probe and / or the electromagnetic probe under the triggering of the PC, receive an echo signal sent by the piezoelectric probe and / or the electromagnetic probe, and convert the echo signal into a digital signal; The piezoelectric probe is used to receive a pulse excitation signal, collect piezoelectric information of the bolt to be tested under the triggering of the pulse excitation signal, and send the piezoelectric information to the controller; The electromagnetic probe is used to receive a pulse excitation signal, collect electromagnetic information of the bolt to be tested under the triggering of the pulse excitation signal, and send the electromagnetic information to the controller.

7. The bolt axial force detection device according to claim 6, characterized in that: The piezoelectric probe includes a first piezoelectric probe and a second piezoelectric probe; The first piezoelectric probe is used to receive a pulse excitation signal, and under the triggering of the pulse excitation signal, collect shear wave information of the bolt to be tested, and send the shear wave information to the controller; The second piezoelectric probe is used to receive a pulse excitation signal, and under the triggering of the pulse excitation signal, collect longitudinal wave information of the bolt to be tested, and send the longitudinal wave information to the controller.

8. The bolt axial force detection device according to claim 6, characterized in that: Also includes a data transceiver interface; The data transceiver interface is used to receive the pulse excitation signal sent by the controller, and send the pulse excitation signal to the piezoelectric probe and / or the electromagnetic probe; and receiving the echo signal sent by the piezoelectric probe and / or the electromagnetic probe, and sending the echo signal to the controller.

9. The bolt axial force detection device according to claim 6, characterized in that: It also includes a channel switching device, which is used to merge the shear wave signal collected by the first piezoelectric probe and the longitudinal wave signal collected by the second piezoelectric probe.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the bolt axial force detection method according to any one of claims 1 to 5 is implemented.

Images