A method for detecting residual stress based on chirp electromagnetic surface wave

Abstract

The application discloses a residual stress detection method based on chirp electromagnetic surface acoustic wave. Although the existing variable interval coil structure can expand the working frequency band of the sensor, the excitation ability of the low frequency component is limited, and it is difficult to meet the stress detection requirements of large thickness components. The electromagnetic acoustic sensor structure is designed, the width of the coil is continuously changed along the ultrasonic surface wave propagation direction, the wide frequency excitation of the chirp surface wave signal is realized, especially the excitation and propagation effect of the low frequency component is enhanced, so that the signal-to-noise ratio and detection accuracy after the signal matched filtering are significantly improved. The sensor utilizes the characteristics that the penetration depth of the ultrasonic surface wave changes with the frequency, realizes the layered detection of the multi-level residual stress in the material through the chirp signal, can cover multiple frequency ranges to adapt to components with different thicknesses, thereby avoiding the problem that the traditional fixed frequency or variable interval sensor needs to be frequently replaced. The method has high detection accuracy, wide detection range and high detection efficiency, and is suitable for nondestructive residual stress evaluation of metal components and engineering structure health monitoring.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing technology, specifically to a residual stress detection method based on Chirp electromagnetic surface wave (EMIS) tomography. This invention falls within the interdisciplinary field of EMIS sensor manufacturing and ultrasonic surface wave stress tomography, and is primarily applied to the nondestructive testing of residual stress in metal components, particularly suitable for measuring stress distribution in thick metal components used in aerospace, high-end equipment manufacturing, energy, and transportation industries. It offers advantages such as a wide detection range, high accuracy, and excellent efficiency. Background Technology

[0002] With the rapid development of high-end equipment manufacturing, high demands are placed on the performance and reliability of key components. However, residual stress is inevitably generated during the manufacturing and processing of components. Residual stress refers to the stress that is not released within the material during external loads, processing, or heat treatment. Residual stress has a significant impact on the mechanical properties, fatigue performance, and service life of materials. Therefore, accurately detecting the distribution of residual stress within materials is of great engineering significance. Traditional residual stress detection methods, such as X-ray diffraction, blind hole methods, and cutting methods, can provide relatively accurate stress measurements, but they usually have drawbacks such as damaging samples, complex operation, or only providing surface information. Furthermore, they are less effective at detecting deep stress distribution in thick materials.

[0003] In recent years, ultrasonic surface wave (SSW) testing technology has become an important means of non-destructive testing of residual stress due to its advantages such as high sensitivity, non-contact testing capability, and applicability to materials of varying thicknesses. However, traditional SSW methods generally employ fixed-frequency excitation, and their sensor structures can only operate at a single frequency, making it difficult to simultaneously acquire stress information from both shallow and deep layers. When it is necessary to detect the residual stress distribution at different depths, multiple measurements must be performed by changing the sensor or adjusting the excitation frequency. This is not only complex and inefficient, but also introduces systematic errors due to frequency switching and changes in coupling conditions, thus affecting the accuracy of the stress detection results.

[0004] Compared to fixed-frequency excitation signals, broadband chirp signals can cover a wider frequency range, thereby exciting surface waves of multiple wavelengths and enabling simultaneous detection of different depth layers of materials. This continuously variable frequency excitation method not only improves the ability to acquire deep information but also enhances signal matching efficiency and detection sensitivity, resulting in clearer echo signals and a significantly improved overall signal-to-noise ratio. Broadband excitation can effectively improve the accuracy and efficiency of residual stress detection, making it particularly suitable for analyzing the internal stress distribution of thick components.

[0005] However, under wideband excitation conditions, traditional electromagnetic acoustic sensors typically employ coil structures with equal linewidth or fixed spacing. Their electromagnetic coupling characteristics are more sensitive to high-frequency components, while the excitation efficiency of low-frequency sound waves is low, resulting in insufficient deep penetration capability and severe attenuation of low-frequency signal components, which limits the detection depth and resolution of thick materials.

[0006] To address the aforementioned issues, this invention proposes a residual stress detection method based on chirp electromagnetic surface waves. By employing a variable linewidth coil design and combining wideband excitation with signal hierarchical reconstruction, it is possible to detect the residual stress distribution at different depths in thick components. Summary of the Invention

[0007] This invention proposes a residual stress detection method based on chirp electromagnetic surface waves for non-destructive testing of thick metal components. This method utilizes broadband chirp signal excitation, sensor structure optimization, and layered signal reconstruction to analyze the residual stress distribution along the thickness direction of the component and obtain quantitative stress tomography results. The specific steps are as follows:

[0008] Step 1: Sensor Coil Design. To improve the excitation efficiency of broadband signals, the sensor excitation coil is designed so that its linewidth gradually varies along the surface wave propagation direction. Assume the chirp signal is a linearly frequency-modulated signal:

[0009]

[0010] Where: f0 — starting frequency

[0011] B - Bandwidth

[0012] T - Pulse Width

[0013] The signal at the adjacent zero crossing point t i With t i+1 Integrals between

[0014]

[0015] Used to determine the line width w of the i-th coil element i :

[0016]

[0017] Step 2: Broadband chirp signal excitation. An ultrasonic transducer is used to excite a broadband chirp signal on the surface of the material under test, generating surface waves covering the frequency range f0-f1. The empirical relationship between the surface wave penetration depth z and the frequency f is:

[0018]

[0019] Where v is the surface wave velocity. This relationship ensures that different frequency components correspond to different depths in the material, enabling multi-layer information acquisition.

[0020] Step 3: Signal Reconstruction and System Analysis. After the received signal is acquired by the signal processing module, its amplitude-frequency and phase response characteristics are analyzed. The propagation process of ultrasonic surface waves in the component is regarded as a linear time-invariant system, whose transmission characteristics remain constant under different chirp excitation conditions. Since the chirp signal is a continuous linear frequency-modulated form in the time domain, it can be equivalent to a superposition of a series of single-frequency excitations in the frequency domain. Therefore, its effect on each frequency component can be regarded as a uniformly covered multi-frequency parallel excitation process.

[0021] In signal analysis, the amplitude-frequency response and phase response functions of the system can be obtained by comparing the amplitude-frequency characteristics of the excitation signal and the received signal. These response functions reflect the propagation characteristics of the material at different frequency components, including physical information such as energy attenuation, phase delay, and dispersion effects. Utilizing the equivalent excitation effect of the chirp signal on frequency components over a wide bandwidth, the overall received signal can be reconstructed using a signal reconstruction algorithm, resulting in response signals for several corresponding single-frequency components. Each single-frequency response signal corresponds to the propagation result of the material at a specific frequency, thus allowing the reconstruction of single-frequency received signals at multiple depths.

[0022] Step 4: Stress Tomography. Based on the frequency-depth correspondence, the material is divided into N detection layers along the thickness direction, each with a thickness Δz. By reconstructing the time-domain signals of each frequency component, the transit time t of the surface wave in each layer is obtained, and the wave velocity of that layer is calculated. Combined with the pre-calibrated wave velocity-stress relationship, the wave velocity of each layer can be converted into stress values, thus obtaining the layered stress distribution along the thickness direction.

[0023] The above one or more technical solutions have the following beneficial results:

[0024] (1) The low-frequency components of the surface wave signal are enhanced, improving the deep detection capability. By matching the variable linewidth coil structure with the chirp signal, this invention can enhance the excitation effect of the low-frequency components. Since low-frequency surface waves have a greater penetration depth, they can effectively obtain deep stress information inside the component, thereby improving the deep residual stress detection capability of thick materials.

[0025] (2) Improved overall signal-to-noise ratio and detection accuracy. The variable linewidth coil structure can achieve efficient excitation of frequency components under wideband chirp signal excitation, making the echo signal amplitude uniform and the signal energy sufficient. Combined with matched filtering, it can significantly improve the overall signal-to-noise ratio of the received signal, enhance signal recognition capability and stress calculation accuracy.

[0026] (3) Improved surface wave stress detection efficiency. The broadband chirp signal covers multiple frequency components, which can simultaneously excite shallow and deep surface waves, enabling multi-layer detection with a single excitation, thus greatly improving detection efficiency. This method is applicable to components of different thicknesses and materials, and can achieve efficient and non-destructive measurement of residual stress distribution without frequent sensor replacement. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the variable linewidth coil structure of the present invention.

[0028] Figure 2 Comparison of the amplitude-frequency characteristics of excitation signals for different coil structures

[0029] Figure 3 Figure showing the residual stress distribution Detailed Implementation

[0030] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and examples. It should be understood that the following examples are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0031] The first step, in the sensor design phase, is to improve the excitation efficiency of broadband signals. A variable linewidth coil structure with a gradually varying linewidth along the surface wave propagation direction is proposed. Three sensor models—single coil, variable spacing, and variable linewidth structure—are established using COMSOL, and their received signal spectra are simulated under the same excitation conditions. The variable linewidth coil structure is shown below. Figure 1 As shown, the line width w of the i-th coil unit i The coil linewidth depends on the signal at the adjacent zero crossing point t. i With t i+1 The integral between them can be expressed as:

[0032]

[0033]

[0034] The particle vibration displacement was extracted in the simulation, and the amplitude-frequency response curve after Fourier transform was compared. The results are as follows: Figure 2 As shown, the variable linewidth design effectively improves the uniform response capability of wideband signals by changing the spatial sensitivity distribution of the coil, thus achieving higher wideband excitation efficiency.

[0035] A stable static magnetic field is established on the surface of a conductive material using a permanent magnet. A broadband chirp current is then passed through a variable-width loop coil, inducing eddy currents on the material surface. Because the coil linewidth gradually changes along the direction of surface wave propagation, the eddy current density and distribution also exhibit gradual spatial variation, allowing for optimized modulation of the Lorentz force at different locations and frequency components. This spatially varying Lorentz force distribution excites a broadband surface wave covering the frequency range f0 to f1 on the material surface, achieving simultaneous and efficient excitation of multiple frequency components. The variable-width structure, by controlling the local current density distribution of the coil, enhances high-frequency components in the narrow-linewidth region and strengthens the coupling of low-frequency components in the wide-linewidth region, thereby significantly improving the energy utilization of the excitation signal.

[0036] The second step is to consider the penetration depth of surface waves as a function of frequency. Studies have shown that the detection depth z and the center frequency f satisfy an empirical formula:

[0037]

[0038] The surface wave velocity in the aluminum alloy plate is 2948 m / s. The chirp signal with a frequency range of 200 kHz to 3 MHz corresponds to a detection depth z range of 0.1 to 10 mm.

[0039] The chirp signal can be represented as:

[0040]

[0041] Where: f0 — starting frequency — 200kHz

[0042] B – Bandwidth – 2.8MHz

[0043] T – Pulse Width – 10µs

[0044] The third step involves acquiring the received signal using two-dimensional nodes, then performing Fourier transforms on both the excitation and received signals to obtain their amplitude-frequency response curves. Based on the relationship between these curves, the overall system transfer function, encompassing the surface wave excitation, propagation, and reception processes, is extracted. Since this system can be considered a linear system, its transfer function remains unchanged for different excitation functions. The chirp signal exhibits linear frequency modulation in the time domain and can be considered a continuously variable excitation. Its frequency response function comprehensively reflects the propagation characteristics of the material at different frequency components, including energy attenuation, phase delay, and dispersion effects. Therefore, amplitude-frequency response curves for multiple single-frequency signals can be established and multiplied by the transfer function to obtain the received amplitude-frequency characteristics of the corresponding single-frequency signal. This process can be represented as follows:

[0045]

[0046] The fourth step involves establishing a COMSOL simulation model and designing a trapezoidal variable cross-section member. A uniaxial tensile load is applied to the member to obtain the stress distribution along the thickness direction at its midpoint. The average stress at each thickness is calculated using a step size of 0.5 mm. The transit time difference of each dominant frequency component is determined using a cross-correlation algorithm under both stress-free and gradient stress conditions. Combined with the wave velocity-stress relationship curve, the residual stress value at the corresponding thickness location is calculated. This yields the following results: Figure 3 The diagram shows the layered residual stress distribution along the thickness direction.

[0047] The results show that the method for detecting residual stress distribution in thick components based on broadband chirp surface wave signals, combined with variable linewidth coil design, broadband signal excitation, and multi-frequency signal reconstruction method, can achieve quantitative characterization of residual stress distribution along the thickness direction in thick components.

Claims

1. A method for detecting residual stress based on Chirp electromagnetic surface waves, characterized in that, By simultaneously optimizing the time-domain characteristics of the chirp excitation signal and the coil geometric parameters, an electromagnetic acoustic sensor structure with a coil linewidth that continuously varies along the propagation direction of ultrasonic surface waves is designed, thereby achieving enhanced excitation of low-frequency components and efficient excitation of broadband surface waves. Based on the characteristic that the penetration depth of ultrasonic surface waves varies with frequency, multilayer tomography is used to detect stress information at different depths within a material. This method includes the following steps: Step 1: Sensor Coil Design; Design the excitation coil of the electromagnetic acoustic sensor, making the linewidth of the coil gradually change along the surface wave propagation direction; Integrate the chirp excitation signal over the time interval between adjacent zero crossings, calculate the linewidth of each coil element based on the integration result, and thus establish a linewidth distribution model to achieve coordinated excitation of low-frequency and high-frequency components; Assume the chirp signal is a linear frequency modulated signal: ； Where: f0 — starting frequency; B – Bandwidth; T—Pulse width; The signal at the adjacent zero crossing point t i With t i+1 Integrals between; ； Used to determine the line width w of the i-th coil unit i : ； Step 2: Broadband chirp signal excitation; Based on the characteristic that the penetration depth of ultrasonic surface waves is proportional to the frequency, an appropriate frequency range is selected to generate a broadband linearly modulated chirp excitation signal. This causes the sensor to generate ultrasonic surface wave signals covering multiple frequency components on the surface of the conductive material being measured, enabling detection at different depths. The empirical relationship between the surface wave penetration depth z and the frequency f is: ； Where v is the surface wave velocity; Step 3: Signal Reconstruction and System Analysis; The surface wave propagation process is considered as a linear time-invariant system, whose transfer function remains unchanged under different excitation conditions; By analyzing the amplitude-frequency and phase response characteristics of the excitation signal and the received signal, the propagation characteristics of each frequency component are extracted; The equivalent excitation effect of the chirp signal on different frequency components in a wide frequency range is used to realize the separation and reconstruction of each single frequency response, providing basic data for subsequent residual stress tomography detection; Step 4: Stress tomography; By utilizing the wave velocity variation patterns corresponding to different frequency components, a mapping relationship between frequency and detection depth is established. The stress distribution at different depths of the material is then to be tomographically analyzed and features extracted to obtain information on the layered distribution of residual stress.

2. The residual stress detection method based on chirp electromagnetic surface waves according to claim 1, characterized in that, This structure can enhance the energy output of low-frequency components under broadband excitation conditions, improve the penetration ability and signal stability of surface waves in thick materials, and achieve higher detection sensitivity and resolution in deep regions.

3. The residual stress detection method based on Chirp electromagnetic surface waves according to claim 1, characterized in that, The chirp excitation signal is a linear frequency modulated signal, and its frequency range is determined according to the acoustic characteristics and thickness parameters of the material under test, so as to cover the main propagation frequency band of surface waves, thereby realizing the acquisition of multi-layer stress information from shallow to deep layers.

4. The residual stress detection method based on chirp electromagnetic surface waves according to claim 1, characterized in that, Based on the linear frequency modulation characteristics of chirp signals and the linear time-invariant characteristics of surface wave propagation systems, the independent separation and time-domain reconstruction of different frequency components are achieved by analyzing the amplitude-frequency and phase responses of the excitation and received signals.

5. The residual stress detection method based on chirp electromagnetic surface waves according to claim 1, characterized in that, Tomographic detection is based on the characteristic that frequency determines the propagation wavelength, and wavelength determines the effective detection depth. By covering different single-wavelength detection layers with multi-frequency signals, multi-frequency information fusion is achieved. Combined with the approximately linear relationship between stress and wave velocity in the material, the stress distribution along the thickness direction of the component is obtained by inversion. This provides a quantitative basis for the non-destructive assessment of residual stress inside the material.

Images