Short bolt axial stress detection method and device based on electromagnetic ultrasonic resonance

Abstract

The invention provides a short bolt axial stress detection method and device based on electromagnetic ultrasonic resonance, and relates to the technical field of industrial fastener stress detection. The method comprises the following steps: determining a preset period and frequency of an excitation signal according to material characteristics of a to-be-measured short bolt, and calibrating resonant frequency and stress to obtain a calibration curve; classifying the lengths of the short bolts to be measured to obtain bolts with known original lengths and in-service bolts with unknown original lengths; based on the determined preset period and frequency, first-order resonant frequency corresponding to the bolt with the known original length is obtained through an electromagnetic ultrasonic transducer and a single-wave resonance method; based on the determined preset period and frequency, first-order resonant frequency corresponding to the in-service bolt with the unknown original length is obtained through an electromagnetic ultrasonic transducer and a double-wave resonance method; and calculating the axial stress of the corresponding short bolt to be measured according to the calibration curve and the first-order resonant frequency. The problems that in the prior art, the energy conversion efficiency is low, and the short bolt detection effect is poor are solved.

Description

Technical Field

[0001] This invention relates to the field of industrial fastener stress testing technology, and in particular to a method and apparatus for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance. Background Technology

[0002] Bolts, as one of the most common fasteners, are widely used in various fields such as industry, defense, construction, transportation, and medical equipment. They are key components in facilities and equipment in aerospace, bridge construction, automobiles, ships, and railways. The vast majority of bolts in important structures are made of metal and mate with nuts and washers. Their main function is to connect and fix structural components, ensuring their stability in predetermined positions. During service, bolts must maintain good working performance, requiring sufficient strength and stiffness to guarantee the stability and reliability of the structure. The stress conditions on bolts can affect their lifespan and connection performance, thus impacting the stability and reliability of the connected structure. Bolts are subjected to significant stress during structural assembly and operation, primarily axial stress. If the axial stress is too low, it indicates bolt loosening or failure, rendering the bolt ineffective and significantly impacting the stability of the entire equipment. In this case, the bolt should be replaced or tightened with appropriate tools, and anti-loosening measures should be implemented. Conversely, if the axial stress is too high, exceeding the material's yield strength or allowable limit under specific conditions, the bolt may undergo plastic deformation or fatigue fracture, affecting connection stability and posing a risk of breakage. In this situation, the bolt material should be replaced or other measures taken to reduce the axial stress. Therefore, detecting bolt axial stress is of significant engineering importance.

[0003] When performing axial stress testing on bolts in industrial applications, especially when the bolt length is short, the accuracy of the measurement and how to achieve rapid testing without damaging the original bolt structure are current challenges. Currently, there are various methods for testing bolt axial stress, including resistance strain gauge method, torque wrench method, ultrasonic method, and strain gauge method.

[0004] Ultrasonic testing is a safe, non-contact measurement method, giving it unique advantages in defect detection and leading to its widespread application in defect monitoring. Particularly for axial stress detection of in-service bolts, it only requires placement on the surface of the test piece to perform measurements without damaging the bolt's original structure, making it suitable for harsh industrial environments and a crucial method for determining bolt failure. However, traditional ultrasonic measurements require the use of a coupling agent to prevent excessive sound energy loss due to the acoustic impedance difference between the test piece material and the piezoelectric crystal material. Furthermore, the use of coupling agents in piezoelectric ultrasound makes long-term stable measurements difficult under high and low temperature conditions and hinders accurate measurements on workpieces with poor surface quality. Additionally, a single probe cannot excite a wideband signal. On the other hand, while traditional electromagnetic ultrasonic testing methods overcome the limitations of coupling agents and probe center frequency, they still have shortcomings in axial stress detection of short bolts and suffer from low transduction efficiency. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a method and device for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance. This invention solves the problems of low transduction efficiency and poor detection effect on short bolts in the existing detection technology.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] A method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance, comprising:

[0008] Based on the material properties of the short bolt to be tested, the preset period and frequency of the excitation signal are determined, and the resonant frequency and stress are calibrated to obtain the calibration curve.

[0009] The lengths of the short bolts to be tested are classified to obtain bolts with known original lengths and in-service bolts with unknown original lengths;

[0010] Based on the predetermined preset period and frequency, the first-order resonant frequency of the bolt with a known original length is obtained by using an electromagnetic ultrasonic transducer and the single-wave resonance method.

[0011] Based on a predetermined preset period and frequency, the first-order resonant frequency of an in-service bolt of unknown original length is obtained by using an electromagnetic ultrasonic transducer and the dual-wave resonance method.

[0012] The axial stress of the short bolt to be tested is calculated based on the calibration curve and the first-order resonant frequency.

[0013] Preferably, the step of obtaining the first-order resonant frequency corresponding to a bolt of known original length using an electromagnetic ultrasonic transducer and a single-wave resonance method based on a predetermined preset period and frequency includes:

[0014] The electromagnetic ultrasonic transducer is installed on the surface of a bolt of known original length and activated based on a preset period and frequency to emit a target excitation signal for detection and obtain an echo signal.

[0015] The echo signal was subjected to Fourier transform analysis to extract the resonant frequency;

[0016] The corresponding first-order resonant frequency is determined based on the resonant frequency.

[0017] Preferably, the step of performing Fourier transform analysis on the echo signal to extract the resonant frequency includes:

[0018] Empirical mode decomposition (EMD) is performed on the echo signal to obtain multiple intrinsic mode functions;

[0019] The instantaneous frequency and instantaneous amplitude of each intrinsic mode function are extracted using the Hilbert transform;

[0020] The intrinsic mode function is denoised based on the instantaneous frequency and instantaneous amplitude to obtain a denoising function;

[0021] The resonant frequency is obtained by performing inverse wavelet transform and Fourier transform on the denoising function.

[0022] Preferably, obtaining the first-order resonant frequency corresponding to an in-service bolt of unknown original length using the dual-wave resonance method includes:

[0023] The initial weights of the three methods in the two-wave resonance method are determined based on the resonant energy dominance criterion. The three methods include instantaneous frequency blind source separation (TF-BSS), resonant sparse dictionary learning (RSDL), and resonant subspace filtering (RSF).

[0024] The initial weights of the three methods are updated and adjusted accordingly using the separation effect feedback criterion, the reconstruction accuracy driving criterion, and the signal-to-noise ratio gain to obtain the final weights of the three methods.

[0025] The filtering results of each method are fused according to the final weights to obtain the fused result.

[0026] The first-order resonant frequency corresponding to the in-service bolt with unknown original length is determined based on the SNC method and the fused result.

[0027] Preferably, determining the first-order resonant frequency corresponding to the in-service bolt with unknown original length based on the SNC method and the fused result includes:

[0028] The fused result is then subjected to FFT processing to obtain the signal spectrum;

[0029] The resonant frequency range is determined based on the spectrum of the signal;

[0030] The signals corresponding to the current resonant frequency range are superimposed and compressed to determine the first-order resonant frequency of the transverse wave and the first-order resonant frequency of the longitudinal wave.

[0031] Preferably, the expression for the calibration curve corresponding to a bolt of known original length is:

[0032] ;

[0033] in, Here, E is the first-order resonant frequency, E is Young's modulus, and A is the acoustoelastic coefficient. σ represents the transit time of a transverse or longitudinal wave under stress-free conditions, and σ is the axial stress on the bolt. The length of the main load-bearing part. This represents the original length of the bolt under stress-free conditions. The slope of the curve. This is the curve intercept.

[0034] Preferably, the expression for the calibration curve corresponding to the in-service bolt with unknown original length is:

[0035] ;

[0036] Among them, f L f is the first-order resonant frequency of the longitudinal wave. S It is the first-order resonant frequency of the transverse wave. The slope of the curve. This is the curve intercept.

[0037] An axial stress detection device for short bolts based on electromagnetic ultrasonic resonance, comprising:

[0038] The calibration module is used to determine the preset period and frequency of the excitation signal based on the material properties of the short bolt to be tested, and to calibrate the resonant frequency and stress to obtain the calibration curve.

[0039] The classification module is used to classify the length of the short bolts to be tested, and to obtain bolts with known original lengths and in-service bolts with unknown original lengths;

[0040] The single-wave extraction module is used to obtain the first-order resonant frequency of a bolt of known original length by using an electromagnetic ultrasonic transducer and the single-wave resonance method, based on a predetermined preset period and frequency.

[0041] The dual-wave extraction module is used to obtain the first-order resonant frequency of an in-service bolt of unknown original length by using an electromagnetic ultrasonic transducer and the dual-wave resonance method, based on a predetermined preset period and frequency.

[0042] A computer system is used to calculate the axial stress of the short bolt to be tested based on the calibration curve and the first-order resonant frequency.

[0043] Preferably, the electromagnetic ultrasonic transducer comprises:

[0044] Permanent magnet, iron-silicon alloy backplate, excitation coil and receiving coil;

[0045] The permanent magnet is used to provide a static magnetic field, the iron-silicon alloy back plate is used to enhance the magnetic field generated by the permanent magnet, the excitation coil is used to pass an excitation current to generate eddy currents on the surface of the short bolt to be tested, and the receiving coil is used to receive the echo signal.

[0046] Preferably, the excitation coil is a butterfly coil.

[0047] The present invention discloses the following technical effects:

[0048] This invention provides a method and apparatus for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance. The method includes: determining the preset period and frequency of the excitation signal according to the material properties of the short bolt to be tested, and calibrating the resonant frequency and stress to obtain a calibration curve; classifying the length of the short bolt to be tested into bolts with known original lengths and in-service bolts with unknown original lengths; obtaining the first-order resonant frequency corresponding to the bolt with known original length using an electromagnetic ultrasonic transducer and a single-wave resonance method based on the determined preset period and frequency; obtaining the first-order resonant frequency corresponding to the in-service bolt with unknown original length using an electromagnetic ultrasonic transducer and a double-wave resonance method based on the determined preset period and frequency; and calculating the axial stress of the corresponding short bolt to be tested according to the calibration curve and the first-order resonant frequency. This invention has the advantages of high precision and non-contact measurement, and can accurately calculate the axial stress of unknown short bolts, thereby quickly identifying potential structural safety hazards. By establishing a calibration curve between the resonant frequency and stress, and using flexible excitation signal settings, it can adapt to various material properties and reduce interference with equipment and the environment. Meanwhile, this method improves work efficiency, is applicable to a variety of complex environments, ensures the long-term stability and safety of engineering projects, overcomes the limitations of traditional contact measurement, and provides a more reliable and effective solution for structural inspection. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A flowchart of a method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance is provided for an embodiment of the present invention.

[0051] Figure 2A flowchart illustrating the detection details provided in this embodiment of the invention;

[0052] Figure 3 A schematic diagram of the electromagnetic ultrasonic bolt axial stress detection system provided in an embodiment of the present invention;

[0053] Figure 4 Typical excitation signals and their spectrum diagrams are provided for embodiments of the present invention;

[0054] Figure 5 A resonant principle diagram provided for an embodiment of the present invention;

[0055] Figure 6 This is a flowchart of echo signal processing provided in an embodiment of the present invention;

[0056] Figure 7 This is a schematic diagram of the SNC method provided in an embodiment of the present invention;

[0057] Figure 8 This is a schematic diagram of the experimental data results curves of the resonance method provided in an embodiment of the present invention, wherein, Figure 8 (a) shows the curve corresponding to the single-wave resonance method. Figure 8 (b) is the curve corresponding to the two-wave resonance method. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] like Figure 1 As shown, this invention provides a method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance, comprising:

[0061] Step 100: Based on the material properties of the short bolt to be tested, determine the preset period and frequency of the excitation signal and calibrate the resonant frequency and stress to obtain the calibration curve;

[0062] Step 200: Classify the lengths of the short bolts to be tested to obtain bolts with known original lengths and in-service bolts with unknown original lengths;

[0063] Step 300: Based on the predetermined preset period and frequency, the first-order resonant frequency corresponding to the bolt with a known original length is obtained by using an electromagnetic ultrasonic transducer and the single-wave resonance method.

[0064] Step 400: Based on the predetermined preset period and frequency, the first-order resonant frequency corresponding to the in-service bolt with unknown original length is obtained by using an electromagnetic ultrasonic transducer and the dual-wave resonance method.

[0065] Step 500: Calculate the axial stress of the short bolt to be tested based on the calibration curve and the first-order resonant frequency.

[0066] Furthermore, such as Figure 2 As shown, the process of obtaining the first-order resonant frequency corresponding to a bolt of known original length using an electromagnetic ultrasonic transducer and the single-wave resonance method, based on a predetermined preset period and frequency, includes:

[0067] The electromagnetic ultrasonic transducer is installed on the surface of a bolt of known original length and activated based on a preset period and frequency to emit a target excitation signal for detection and obtain an echo signal.

[0068] The echo signal was subjected to Fourier transform analysis to extract the resonant frequency;

[0069] The corresponding first-order resonant frequency is determined based on the resonant frequency.

[0070] Specifically, the electromagnetic ultrasonic transducer based on Lorentz force consists of a permanent magnet, an iron-silicon alloy backplate, an excitation coil, and a receiving coil. The permanent magnet is used to provide a static magnetic field, the backplate is used to enhance the magnetic field generated by the permanent magnet, the excitation coil mainly functions to pass an excitation current to generate eddy currents on the surface of the bolt specimen, and the receiving coil mainly functions to receive the echo signal.

[0071] Electromagnetic ultrasonic transducers based on the Lorentz force can generate body waves, i.e., transverse or longitudinal waves, within a specimen. The excitation process begins by passing an alternating current through a coil. The non-zero conductivity surface of the specimen induces eddy currents under the influence of a dynamic magnetic field, with the direction opposite to the direction of the current flowing through the coil. Under the influence of a static magnetic field generated by a permanent magnet, electrons on the specimen surface are deflected and vibrate due to the Lorentz force, colliding with atoms and generating vibrations. These continuous collisions between atoms propagate their energy within the specimen as mechanical waves (body waves). The receiving process is the reverse of the excitation process. Electrons, under the influence of the Coulomb force, vibrate at the same speed as the atoms. The moving electrons are deflected by the Lorentz force in the static magnetic field, forming eddy currents. The alternating eddy current field generates a dynamic magnetic field, and the coil induces an electrical signal under the influence of this dynamic magnetic field, thus achieving reception.

[0072] A butterfly coil is used as the excitation coil. The butterfly coil consists of a central main lobe and two side lobes. Its main working region is the main lobe. The current direction in the main lobe region is the same, meaning the current in the central conductor is in the same direction. Therefore, the energy of the main lobe is more concentrated, while the energy of the side lobes is relatively weaker. Taking a transverse wave as an example, the sound field directivity function of the main lobe of the helical coil in polar coordinates is:

[0073] ;

[0074] In the formula, a is the half-field of the main lobe region, f is the excitation current frequency, f(θ) is a function related to the transverse and longitudinal wave velocities, and k s This represents the wave number of the transverse wave. The radiated sound field excited by the butterfly coil electromagnetic ultrasonic transducer is relatively concentrated along the central axis.

[0075] A chirped signal (linear frequency modulated signal) is used as the excitation current signal. A chirped signal is a signal whose frequency varies with time. Due to its wide bandwidth and good autocorrelation properties, it is widely used in radar, sonar, communication, and ultrasonic testing. A linear chirped signal is used as the excitation current signal:

[0076] ;

[0077] In the formula, For amplitude, The initial frequency, For the frequency sweep rate, The initial phase is determined by using a long-period chirped signal as the excitation current signal, which ensures that the ultrasonic wave resonates within a certain frequency band, thereby enabling the extraction of the resonant frequency of the electromagnetic ultrasonic echo signal.

[0078] Furthermore, the principle of magnetic ultrasonic measurement of bolt axial stress is mainly based on Hooke's law and the acoustic elastic effect. The influence of axial stress on bolt characteristics is mainly reflected in its influence on bolt length and sound velocity in the bolt stress area.

[0079] According to Hooke's Law, within the elastic deformation range of a plastic material, stress is directly proportional to strain. When a bolt is subjected to axial stress, the length L of the main stress-bearing part is... e Change to L σ :

[0080] ;

[0081] In the formula, σ is the axial stress on the bolt, and E is Young's modulus.

[0082] According to the acoustoelastic effect, the propagation speed of sound waves in a solid is related to the stress it is subjected to. Therefore, the volume wave velocity in a bolt under axial stress is:

[0083] ;

[0084] In the formula, v0 is the propagation velocity of transverse or longitudinal waves under stress-free conditions, v σ Let σ be the propagation velocity of the transverse or longitudinal wave under stress σ, A be the acoustoelastic coefficient, and t0 be the transit time of the transverse or longitudinal wave under no stress.

[0085] The relationship between transverse and longitudinal wave velocities and stress is as follows:

[0086] ;

[0087] ;

[0088] In the formula, ρ is the material density, and v L For the longitudinal wave velocity, v S Let λ be the transverse wave velocity, μ and λ be the second-order elastic coefficients, and l, m, and n be the third-order elastic coefficients.

[0089] Electromagnetic ultrasonic resonance measurement of bolt axial stress is mainly divided into single-wave resonance method and dual-wave resonance method. Single-wave resonance uses transverse or longitudinal waves for detection and requires the original length of the bolt to be known, making it more suitable for controlling bolt preload. Dual-wave resonance method uses a combination of transverse and longitudinal waves for measurement, does not require the original length of the bolt to be known, and is more suitable for axial stress detection of bolts in service.

[0090] The condition for ultrasonic waves to resonate in a bolt is that the incident and reflected waves are in phase inside the specimen and their amplitudes are superimposed. Ultrasonic resonance will enhance the energy, and its first-order resonant frequency is:

[0091] ;

[0092] In the formula, t is the ultrasonic wave transit time.

[0093] Based on the Fourier series expansion formula and neglecting higher-order terms, the theoretical formula for measuring the axial stress of bolts using the single-wave resonance method can be obtained:

[0094] ;

[0095] For bolts of the same material and specification, the axial stress and resonant frequency are linearly related. In actual testing, it is necessary to calibrate k1 and b1 and measure the ultrasonic transit time t0 of the bolt under stress-free conditions, which requires knowing the original length of the bolt and the wave velocity of the sound wave in the material.

[0096] The two-wave method establishes the relationship between the ratio of the difference to the sum of the resonant frequencies of the transverse and longitudinal waves and the axial stress. The theoretical formula for the two-wave method is:

[0097] ;

[0098] In the formula, f L f is the first-order resonant frequency of the longitudinal wave. S It is the first-order resonant frequency of the transverse wave.

[0099] In actual testing, k2 and b2 also need to be calibrated. k2 and b2 depend only on the material parameters and are independent of the original bolt length. Therefore, the k2 and b2 of bolts made of the same material are the same. The difference between the axial stress and the resonant frequencies of the longitudinal and transverse waves under stress is linearly related to the sum of the resonant frequencies.

[0100] The noise of the echo signal is suppressed by combining EMD (Empirical Mode Decomposition) with wavelet denoising, and the useful signal is extracted. The spectrum of the echo signal and its first resonant frequency are analyzed by frequency domain analysis and SNC (n-time compression superposition) method.

[0101] Furthermore, the step of performing Fourier transform analysis on the echo signal to extract the resonant frequency includes:

[0102] Empirical mode decomposition (EMD) is performed on the echo signal to obtain multiple intrinsic mode functions;

[0103] The instantaneous frequency and instantaneous amplitude of each intrinsic mode function are extracted using the Hilbert transform;

[0104] The intrinsic mode function is denoised based on the instantaneous frequency and instantaneous amplitude to obtain a denoising function;

[0105] The resonant frequency is obtained by performing inverse wavelet transform and Fourier transform on the denoising function.

[0106] Specifically, the single-wave resonance method combines EMD with wavelet denoising to suppress harmonics and noise in the echo signal. EMD is a technique for processing complex signals, revealing the signal's intrinsic structure through iterative adaptive decomposition into intrinsic mode functions (IMFs). A crucial part of the EMD algorithm is the extraction of instantaneous frequencies. In EMD, the intrinsic mode components (IMFs) of a signal are represented as a signal with time-varying amplitude and instantaneous frequency. The extraction of instantaneous frequencies is typically achieved through Hilbert transform (HT).

[0107] ;

[0108] In the formula, The original signal, The transformed signal, The conjugate signal of the original signal. This is the spectrum of the original signal.

[0109] EMD allows for time-frequency analysis of signals, and is particularly suitable for non-stationary signals. The core idea of ​​EMD is to decompose complex signals into a set of Integral Functions (IMFs) through a "filtering" process.

[0110] ;

[0111] In the formula, It is the amplitude of the time series. It is the instantaneous angular frequency. It is a phase.

[0112] Noise-dominated IMFs are removed by adjusting correlation coefficients and energy ratios. After EMD processing, wavelet denoising is used to threshold the remaining IMFs, enhancing detail preservation. Wavelet denoising employs a "decomposition-screening-reconstruction" method, suppressing high-frequency noise and harmonics by eliminating or shrinking small coefficients while retaining large coefficients. The core of wavelet transform is to decompose the signal into frequency components at different scales using a set of basis functions with "zoom" capabilities.

[0113] ;

[0114] In the formula, b is the scale parameter (controlling frequency resolution), and b is the translation parameter (controlling time positioning). For the mother wavelet function.

[0115] Thresholding is applied to the detail coefficients of each layer:

[0116] ;

[0117] In the formula, c is the retention / shrinkage treatment value. For the threshold, These are the processed coefficients. The core of this step is designing a reasonable threshold strategy, and the threshold is calculated as follows: , is the highest frequency detail coefficient, and N is the signal length.

[0118] By using the processed coefficients to perform inverse wavelet transform, and reconstructing the signal through a biorthogonal wavelet basis to ensure accurate signal recovery, noise suppression of the echo signal can be achieved.

[0119] Furthermore, the method of obtaining the first-order resonant frequency corresponding to an in-service bolt of unknown original length using the dual-wave resonance method includes:

[0120] The initial weights of the three methods in the two-wave resonance method are determined based on the resonant energy dominance criterion. The three methods include instantaneous frequency blind source separation (TF-BSS), resonant sparse dictionary learning (RSDL), and resonant subspace filtering (RSF).

[0121] The initial weights of the three methods are updated and adjusted accordingly using the separation effect feedback criterion, the reconstruction accuracy driving criterion, and the signal-to-noise ratio gain to obtain the final weights of the three methods.

[0122] The filtering results of each method are fused according to the final weights to obtain the fused result.

[0123] The first-order resonant frequency corresponding to the in-service bolt with unknown original length is determined based on the SNC method and the fused result.

[0124] Furthermore, the first-order resonant frequency corresponding to the in-service bolt with unknown original length is determined based on the SNC method and the fused result, including:

[0125] The fused result is then subjected to FFT processing to obtain the signal spectrum;

[0126] The resonant frequency range is determined based on the spectrum of the signal;

[0127] The signals corresponding to the current resonant frequency range are superimposed and compressed to determine the first-order resonant frequency of the transverse wave and the first-order resonant frequency of the longitudinal wave.

[0128] Specifically, the dual-wave resonance method uses an adaptive weighted resonance collaborative separation method to separate and suppress the signal echo signal. That is, the two-wave resonance signal is processed by weighted collaborative processing of three methods: frequency blind source separation (TF-BSS), resonant sparse dictionary learning (RSDL), and resonator space filtering (RSF). The signal is separated and filtered, the spectrum of transverse and longitudinal waves is extracted, and harmonics and noise are suppressed.

[0129] To determine the weights of the three methods, a high-resolution time-frequency map is first obtained using the Synchronous Squeeze Wavelet Transform (SST). SST performs time-frequency pre-enhancement, transforming the time-frequency representation of the continuous wavelet transform (CWT). Transformed into a high-resolution time-frequency representation through frequency reallocation :

[0130] ;

[0131] In the formula, Δf is the frequency resolution. It is a discrete scale.

[0132] The result Integrating over the frequency band yields the theoretical resonant band energy R of the longitudinal / transverse waves:

[0133] ;

[0134] In the formula, For the longitudinal wave resonant frequency band energy, This represents the energy of the transverse wave resonant frequency band.

[0135] The initial weights α1, α2, and α3 for the three methods are determined based on the resonant energy dominance criterion:

[0136] ;

[0137] The first method is to use a virtual multi-channel time-frequency BSS, which constructs virtual channels through time shifting:

[0138] ;

[0139] In the formula, d represents the spatial resolution.

[0140] Its weights are adjusted according to the separation effect feedback criterion. The cross-correlation coefficient ρ between the separated signals is calculated. If ρ > 0.7, the weight of α1 is reduced to achieve dynamic updating of α1.

[0141] ;

[0142] The second method uses RSDL to first construct a resonant atom library. , For longitudinal wave atomic libraries, For transverse wave atom library:

[0143] ;

[0144] Its weights are adjusted according to the reconstruction accuracy-driven criterion, and the residual energy ratio is calculated. .like If the value is less than 0.1, then the weight of α2 is increased to achieve dynamic updates of α2.

[0145] ;

[0146] The third method uses the RSF approach, first constructing the Hankel matrix:

[0147] ;

[0148] In the formula, L is the delay embedding dimension. The signal is then projected onto the P / S resonator space using the oblique projection operator:

[0149]

[0150] In the formula, r is the rank determined based on the number of resonant frequencies.

[0151] Its weights are dynamically adjusted based on the subspace signal-to-noise ratio gain G:

[0152] ;

[0153] The three methods are combined, and the final weighted fused signal is obtained according to their respective weights. The concept of closed-loop feedback is introduced to back-calculate the contribution of each method and recalculate the weights until the correlation coefficients of the transverse and longitudinal waves drop below 0.2. This separates the transverse and longitudinal waves and filters out noise and harmonics.

[0154] To improve the robustness and accuracy of the detection, the first-order resonant frequency of the ultrasonic wave was extracted as an indicator to evaluate the axial stress of the bolt. The noise-suppressed echo signal was processed using a combination of Fourier transform and SNC (Sequential Noise Resonance), and its spectrum was analyzed to extract its first-order resonant frequency.

[0155] The spectrum of the signal is obtained using Fourier transform:

[0156] ;

[0157] in, It is the spectrum of the signal. It is a time-domain signal. It refers to frequency.

[0158] A segment of the spectrum is extracted, and its resonant order is estimated using the SNC method. The SNC method is a method for obtaining the first-order resonant frequency position based on the amplitude spectrum curve of the specimen. First, an FFT is performed on the echo signal to obtain the signal spectrum. Then, the n~n+k order resonant frequency positions are found. This segment of the spectrum is extracted and compressed by a factor of n, n+1, up to n+k. The maximum value of the resulting spectrum curve is then combined, and this maximum value is the first-order resonant frequency.

[0159] ;

[0160] The single-wave method calculates the bolt axial stress value using the first-order resonant frequency of the transverse or longitudinal wave, while the double-wave method calculates the bolt axial stress value using the ratio of the difference between the first-order resonant frequencies of the transverse and longitudinal waves to their sum.

[0161] The single-wave method is mainly used for preload control of bolts of known original length during assembly, while the double-wave method is mainly used for axial stress detection of bolts in service. By combining noise suppression techniques based on EMD and wavelet denoising with spectral analysis techniques based on Fourier transform and SNC, the axial stress of short bolts can be efficiently evaluated, their preload controlled, or bolt failure determined.

[0162] This embodiment also provides a short bolt axial stress detection device based on electromagnetic ultrasonic resonance, including:

[0163] The calibration module is used to determine the preset period and frequency of the excitation signal based on the material properties of the short bolt to be tested, and to calibrate the resonant frequency and stress to obtain the calibration curve.

[0164] The classification module is used to classify the length of the short bolts to be tested, and to obtain bolts with known original lengths and in-service bolts with unknown original lengths;

[0165] The single-wave extraction module is used to obtain the first-order resonant frequency of a bolt of known original length by using an electromagnetic ultrasonic transducer and the single-wave resonance method, based on a predetermined preset period and frequency.

[0166] The dual-wave extraction module is used to obtain the first-order resonant frequency of an in-service bolt of unknown original length by using an electromagnetic ultrasonic transducer and the dual-wave resonance method, based on a predetermined preset period and frequency.

[0167] A computer system is used to calculate the axial stress of the short bolt to be tested based on the calibration curve and the first-order resonant frequency.

[0168] Furthermore, such as Figure 3 As shown, the electromagnetic ultrasonic transducer includes:

[0169] Permanent magnet, iron-silicon alloy backplate, excitation coil and receiving coil;

[0170] The permanent magnet is used to provide a static magnetic field, the iron-silicon alloy back plate is used to enhance the magnetic field generated by the permanent magnet, the excitation coil is used to pass an excitation current to generate eddy currents on the surface of the short bolt to be tested, and the receiving coil is used to receive the echo signal.

[0171] Furthermore, the excitation coil is a butterfly coil.

[0172] Specifically, such as Figure 4-5 As shown, the electromagnetic ultrasonic high-voltage excitation system can generate high-voltage, high-current chirped excitation signals. The specific excitation signal characteristics are determined based on the structural parameters of the test piece. The flight time of the ultrasonic wave under stress-free conditions is estimated based on the bolt length, Young's modulus, and density. The number of chirped signal cycles is set to ensure the duration is not less than half the flight time of the ultrasonic wave in the bolt. The start and stop frequencies of the chirped signal can also be set, and its bandwidth must be greater than three times the theoretical first-order resonant frequency of the bolt specimen under stress-free conditions. The main function of the impedance matching module is to convert the non-purely resistive transducer coil impedance to a purely resistive impedance, thereby improving the transducer's efficiency in receiving the electromagnetic ultrasonic excitation source signal, reducing energy loss, and achieving efficient signal transmission. The excitation and receiving coils are separated to reduce interference from the excitation signal to the receiving end, isolating the excitation and receiving. The system's receiving and excitation work simultaneously to ensure real-time signal transmission. Excitation and reception share a single electromagnetic ultrasonic resonant transducer, which mainly consists of the following three parts:

[0173] Neodymium iron boron (NdFeB) permanent magnets: NdFeB is currently the strongest permanent magnet material, possessing advantages such as high remanence, high coercivity, and high energy product, and it also exhibits strong stability. The main function of the permanent magnet is to provide a strong static magnetic field, ensuring that the surface of the specimen experiences a sufficiently large Lorentz force. Its magnetization direction is along the bolt height direction.

[0174] Ferrosilicon alloy backplate: Ferrosilicon alloy has the advantages of high magnetic permeability, low loss and low cost. It can enhance the magnetic induction intensity of the test specimen surface corresponding to the coil, increase the return energy and improve the transduction efficiency.

[0175] FPC coil: A flexible coil, chosen for its light weight, thinness, and good toughness as the coil of the electromagnetic ultrasonic resonant transducer, allowing for better contact with the specimen surface. A butterfly coil is used as the excitation coil. When an excitation current is applied to the excitation coil, a corresponding alternating magnetic field is generated, thereby exciting eddy currents on the specimen surface; the receiving coil generates an induced voltage to achieve reception.

[0176] When exciting an ultrasonic signal, the electromagnetic ultrasonic high-voltage excitation system applies an excitation current signal to the excitation coil, generating an alternating electric field, which in turn produces an alternating magnetic field. Eddy currents are induced on the surface of the specimen, and the electrons inside are subjected to Lorentz force under the influence of the static magnetic field generated by the permanent magnet, resulting in periodic strain. This periodic strain excites an ultrasonic volume wave signal, allowing the ultrasonic wave to propagate within the bolt specimen. The ultrasonic wave is reflected after reaching the bottom. Due to the large number of excitation cycles and the long excitation time, the wave reflected from the bottom interferes with the downward-propagating excitation waveform. Only specific frequency signal components are amplified, while other frequency signal components are attenuated, resulting in a resonance phenomenon. When receiving the ultrasonic signal, the ultrasonic wave generates eddy currents as it propagates within the bolt specimen and returns to the specimen surface, thus generating an alternating magnetic field. This induces a receiving voltage signal in the receiving coil, completing the reception of the electromagnetic ultrasonic echo signal. The wave velocity changes as the ultrasonic wave passes through the stressed area of ​​the bolt, increasing the transit time and consequently decreasing the resonant frequency. Therefore, the resonant frequency can be used to detect the magnitude of the axial stress in the bolt.

[0177] To prevent large-amplitude excitation signals from damaging the receiving circuit, the received signal needs to be limited to avoid damage to the subsequent conditioning circuitry. The received electromagnetic ultrasonic echo signal typically has a small amplitude and needs to be amplified by a low-noise preamplifier (LNA) to improve the signal-to-noise ratio of the echo, facilitating subsequent signal acquisition and processing. The amplified signal is then sampled by an ADC before being transmitted to the computer system for processing.

[0178] The computer is responsible for system control and signal processing, including setting excitation signal parameters, configuring transducer signal processing parameters, and displaying and storing the final results. Ultimately, it visualizes the axial stress values ​​and can determine whether the bolt preload during assembly meets requirements and whether any bolts in service have failed.

[0179] Furthermore, let's gain a deeper understanding of how to apply the aforementioned device:

[0180] First, based on the characteristics of the bolted material (such as length, elastic modulus, density, etc.), select an appropriate excitation signal frequency and duration. The excitation signal frequency band should cover at least three resonant frequencies, and the duration of the excitation signal must be greater than 0.5 times the transit time of the ultrasonic wave in the bolt under stress-free conditions to ensure that the ultrasonic wave can resonate in the bolt specimen. Furthermore, the curve of axial stress versus resonant frequency in the bolt under stress-free conditions needs to be calibrated.

[0181] For controlling the preload of bolts with known original lengths, a single-wave resonance method is used for detection. A transverse or longitudinal wave electromagnetic ultrasonic resonant transducer is used to detect the bolt preload; for in-service measurement of bolt axial stress, a combination of transverse and longitudinal wave transducers is used.

[0182] The electromagnetic ultrasonic resonant probe consists of a neodymium iron boron permanent magnet, an iron-silicon alloy backplate, an FPC excitation and receiving coil, and a housing. It is placed on the surface of the test material and does not require a coupling agent. It is only necessary to ensure that the lift-off distance between the probe and the test piece is as small as possible to ensure the transducer's transduction efficiency.

[0183] Turn on the electromagnetic ultrasonic high-voltage excitation system, pass the excitation current through the FPC excitation coil of the transducer, and excite it once every certain period of time with an excitation interval of 100 ms to ensure the continuity of detection;

[0184] The probe receives the echo signal reflected from the bolt being tested. The echo signal is an electromagnetic ultrasonic resonant signal, containing the resonant frequency and amplitude information of the ultrasonic waves. The FPC receiving coil transmits the echo signal to the subsequent signal conditioning circuit. After being processed by the limiting circuit and the preamplifier module, the echo signal undergoes analog-to-digital conversion and is transmitted to the computer system for further processing.

[0185] Computer systems process signals, such as Figure 6As shown. The single-wave resonance method first decomposes the echo signal into a set of IMFs using EMD, initially separating noise and signal, and removing the excitation and overshoot regions in the echo, retaining the modal components of the echo signal, thus obtaining the signal after preliminary denoising. Then, wavelet transform is performed on the signal, thresholding is applied to the coefficients of each level of detail, and then inverse wavelet transform is performed on the processed coefficients to reconstruct the echo signal, obtaining the echo signal after harmonic and noise suppression. The dual-wave resonance method first calculates the weights of the three methods TF-BSS, RSDL, and RSF. The weight coefficients of the TF-BSS method are adjusted according to the cross-correlation coefficient ρ, the weight coefficients of the RSDL method are adjusted according to the residual energy ratio, and the weight coefficients of the RSF method are adjusted according to the subspace signal-to-noise ratio gain G. Then, the three methods are calculated in parallel, and the results are weighted, fused, and iterated to separate the transverse and longitudinal waves and suppress noise and harmonics. Next, FFT is used and half of its spectrum is truncated to obtain the actual spectrum of the echo signal. Then, the frequency band range corresponding to the excitation signal in the spectrum is truncated, and the truncated spectrum is processed using the SNC method, as shown. Figure 7 As shown, by compressing and superimposing them, the frequency corresponding to the maximum amplitude is the first-order ultrasonic resonant frequency of the bolt under the current stress state.

[0186] Finally, the first-order resonant frequency data obtained from the transducer is substituted into the pre-calibrated expression for the relationship between the resonant frequency and the axial stress to obtain the axial stress value of the bolt under the current working conditions.

[0187] Furthermore, such as Figure 8 As shown, Figure 8 (a) shows the calibration curve corresponding to the single-wave resonance method. It can be seen from the figure that there is a significant linear relationship between the first-order resonant frequency of the transverse or longitudinal wave and the axial stress of the bolt, R. 2 =0.99, and its first-order resonant frequency shows a decreasing trend with the increase of axial stress, indicating the feasibility of using the single-wave resonance method to measure the axial stress of bolts. Figure 8 (b) shows the calibration curve corresponding to the dual-wave resonance method. It can be seen from the figure that the ratio of the difference to the sum of the first-order resonant frequencies of the transverse and longitudinal waves has a significant linear relationship with the axial stress of the bolt. 2 =0.99, and this ratio shows a decreasing trend as the axial stress increases, indicating the feasibility of using the dual-wave resonance method to measure the axial stress of bolts.

[0188] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0189] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance, characterized in that, include: Based on the material properties of the short bolt to be tested, the preset period and frequency of the excitation signal are determined, and the resonant frequency and stress are calibrated to obtain the calibration curve. The lengths of the short bolts to be tested are classified to obtain bolts with known original lengths and in-service bolts with unknown original lengths; Based on the predetermined preset period and frequency, the first-order resonant frequency of the bolt with a known original length is obtained by using an electromagnetic ultrasonic transducer and the single-wave resonance method. Based on a predetermined preset period and frequency, the first-order resonant frequency of an in-service bolt of unknown original length is obtained by using an electromagnetic ultrasonic transducer and the dual-wave resonance method. The axial stress of the short bolt to be tested is calculated based on the calibration curve and the first-order resonant frequency.

2. The method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance according to claim 1, characterized in that, The process of obtaining the first-order resonant frequency corresponding to a bolt of known original length using an electromagnetic ultrasonic transducer and a single-wave resonance method, based on a predetermined preset period and frequency, includes: The electromagnetic ultrasonic transducer is installed on the surface of a bolt of known original length and activated based on a preset period and frequency to emit a target excitation signal for detection and obtain an echo signal. The echo signal was subjected to Fourier transform analysis to extract the resonant frequency; The corresponding first-order resonant frequency is determined based on the resonant frequency.

3. The method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance according to claim 2, characterized in that, The step of performing Fourier transform analysis on the echo signal to extract the resonant frequency includes: Empirical mode decomposition (EMD) is performed on the echo signal to obtain multiple intrinsic mode functions; The instantaneous frequency and instantaneous amplitude of each intrinsic mode function are extracted using the Hilbert transform; The intrinsic mode function is denoised based on the instantaneous frequency and instantaneous amplitude to obtain a denoising function; The resonant frequency is obtained by performing inverse wavelet transform and Fourier transform on the denoising function.

4. The method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance according to claim 1, characterized in that, The method of obtaining the first-order resonant frequency corresponding to an in-service bolt of unknown original length using the two-wave resonance method includes: The initial weights of the three methods in the two-wave resonance method are determined based on the resonant energy dominance criterion. The three methods include instantaneous frequency blind source separation, resonant sparse dictionary learning, and resonant oscillator space filtering. The initial weights of the three methods are updated and adjusted accordingly using the separation effect feedback criterion, the reconstruction accuracy driving criterion, and the signal-to-noise ratio gain to obtain the final weights of the three methods. The filtering results of each method are fused according to the final weights to obtain the fused result. The first-order resonant frequency corresponding to the in-service bolt with unknown original length is determined based on the SNC method and the fused result.

5. The method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance according to claim 4, characterized in that, The first-order resonant frequency corresponding to the in-service bolt with unknown original length is determined based on the SNC method and the fused result, including: The fused result is then subjected to FFT processing to obtain the signal spectrum; The resonant frequency range is determined based on the spectrum of the signal; The signals corresponding to the current resonant frequency range are superimposed and compressed to determine the first-order resonant frequency of the transverse wave and the first-order resonant frequency of the longitudinal wave.

6. The method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance according to claim 2, characterized in that, The expression for the calibration curve corresponding to the original length of the bolt is: ； in, Here, E represents the first-order resonant frequency corresponding to a single wave, E is Young's modulus, and A is the acoustoelastic coefficient. σ represents the transit time of a transverse or longitudinal wave under stress-free conditions, and σ is the axial stress on the bolt. The length of the main load-bearing part. The original length of the bolt under stress-free conditions. The slope of the first curve is... This is the curve intercept.

7. The method for detecting axial stress in short bolts based on electromagnetic ultrasonic resonance according to claim 6, characterized in that, The expression for the calibration curve corresponding to an in-service bolt with an unknown original length is: ； Among them, f L f is the first-order resonant frequency of the longitudinal wave. S It is the first-order resonant frequency of the transverse wave. This represents the slope of the second curve.

8. A short bolt axial stress detection device based on electromagnetic ultrasonic resonance, applied to the method described in any one of claims 1-7, characterized in that, include: The calibration module is used to determine the preset period and frequency of the excitation signal based on the material properties of the short bolt to be tested, and to calibrate the resonant frequency and stress to obtain the calibration curve. The classification module is used to classify the length of the short bolts to be tested, and to obtain bolts with known original lengths and in-service bolts with unknown original lengths; The single-wave extraction module is used to obtain the first-order resonant frequency of a bolt of known original length by using an electromagnetic ultrasonic transducer and the single-wave resonance method, based on a predetermined preset period and frequency. The dual-wave extraction module is used to obtain the first-order resonant frequency of an in-service bolt of unknown original length by using an electromagnetic ultrasonic transducer and the dual-wave resonance method, based on a predetermined preset period and frequency. A computer system is used to calculate the axial stress of the short bolt to be tested based on the calibration curve and the first-order resonant frequency.

9. The axial stress detection device for short bolts based on electromagnetic ultrasonic resonance according to claim 8, characterized in that, The electromagnetic ultrasonic transducer includes: Permanent magnet, iron-silicon alloy backplate, excitation coil and receiving coil; The permanent magnet is used to provide a static magnetic field, the iron-silicon alloy back plate is used to enhance the magnetic field generated by the permanent magnet, the excitation coil is used to pass an excitation current to generate eddy currents on the surface of the short bolt to be tested, and the receiving coil is used to receive the echo signal.

10. The axial stress detection device for short bolts based on electromagnetic ultrasonic resonance according to claim 9, characterized in that, The excitation coil is a butterfly coil.

Images