A method and system for filtering harmonics of excitation signals of a residual stress measuring instrument

Abstract

The present invention discloses a method and system for filtering harmonics from the excitation signal of a residual stress measuring instrument. The method belongs to the field of signal processing technology. The method comprises decomposing the excitation signal of a magnetic anisotropic residual stress measuring instrument to obtain a multi-layer IMF function of the excitation signal, determining the IMF components containing harmonic components in each layer of the IMF components, obtaining the Fourier spectrum of the IMF components containing harmonics, determining the data length and sampling frequency of the IMF components containing harmonics, determining the frequency resolution of the IMF components containing harmonics based on the data length and sampling frequency, determining the harmonic frequency and frequency resolution to be eliminated based on actual conditions, obtaining a spectrum index corresponding to the harmonic frequency to be eliminated, setting the spectrum value at the spectrum index position to zero, and performing an inverse Fourier transform to obtain the IMF components after filtering out the harmonics. The method can effectively filter out the harmonic components in the excitation signal and improve the accuracy of the measurement results.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and more particularly to a method and system for filtering harmonics of an excitation signal of a residual stress measuring instrument. Background Art

[0002] For magnetic measuring instruments, applying excitation signals of different frequencies to the object to be tested will result in different magnetization parameters (such as remanence and coercivity) of the object to be tested. This can cause deviations in the instrument's interpretation of the test signal, thereby affecting the accuracy of the test results. Therefore, ensuring the uniformity of the excitation signal frequency is crucial to ensuring the accuracy of the test results.

[0003] However, in practical applications, due to the uneven quality of signal generators in magnetic measuring instruments, the presence of nonlinear elements in the circuit, and deficiencies in circuit design, the excitation signal is often accompanied by harmonic components. However, the presence of these harmonics causes the frequency spectrum of the excitation signal to become complex, which significantly interferes with the accurate identification of the detection signal. Therefore, filtering out the harmonic signals has become an urgent task.

[0004] Currently, commonly used filtering techniques include adjustable filters and wavelet transforms. However, adjustable filters are not an ideal choice due to their limited adjustment range, large internal instrument space requirements, high power consumption, and increased instrument cost. On the other hand, although wavelet transforms are widely used in signal processing, their results are significantly affected by the choice of wavelet basis and window function, resulting in a certain degree of uncertainty and low universality.

[0005] In summary, due to the uncertainty and instability of the processing results, the existing harmonic filtering technology cannot provide the instrument probe with excitation signals of different frequency values ​​in a timely and rapid manner according to the needs of on-site detection, which affects the accuracy of the measurement results. Summary of the Invention

[0006] In response to the problems existing in the above-mentioned fields, the present invention proposes a method and system for filtering harmonics of the excitation signal of a residual stress measuring instrument, which can reduce the uncertainty and instability caused by methods such as adjustable filters and wavelet transforms, thereby improving the accuracy and reliability of the measurement results.

[0007] To solve the above technical problems, the present invention discloses a method for filtering harmonics of an excitation signal of a residual stress measuring instrument, comprising the following steps:

[0008] Decompose the excitation signal of the magnetic anisotropic residual stress measuring instrument to obtain the multi-layer IMF function of the excitation signal;

[0009] According to the multi-layer IMF function of the excitation signal, the IMF components containing harmonic components in each layer of IMF components are determined;

[0010] Obtain the Fourier spectrum of the IMF component containing harmonics, determine the data length and sampling frequency of the IMF component containing harmonics; determine the frequency resolution of the IMF component containing harmonics based on the data length and sampling frequency; determine the harmonic frequency and frequency resolution to be eliminated based on actual conditions, obtain the spectrum index corresponding to the harmonic frequency to be eliminated, set the spectrum value at the spectrum index position to zero, and obtain the IMF component after filtering out the harmonics through inverse Fourier transform.

[0011] Preferably, determining the data length and sampling frequency of the IMF component containing harmonics specifically includes:

[0012] Perform Fourier transform on the IMF components containing harmonics, calculate the Fourier spectrum of the IMF components containing harmonics by fast Fourier transform algorithm, and determine the data length and sampling frequency f s ;

[0013] The IMF components containing harmonics are obtained as x[n], n=0, 1, ..., N-1, where N is the data length;

[0014] X[k]=FFT(x[n]), k=0,1,…,N-1, calculated by the Fast Fourier Transform algorithm;

[0015] The result X[k] of the fast Fourier transform algorithm is a complex number whose magnitude represents the intensity of the frequency component and whose phase represents the phase of the frequency component.

[0016] Preferably, obtaining the spectrum index corresponding to the harmonic frequency to be eliminated comprises the following steps:

[0017] According to the data length and sampling frequency, the frequency resolution is calculated as:

[0018]

[0019] In actual situation, the frequency of the mth harmonic to be eliminated is determined to be f h =m×f0, where f0 is the fundamental frequency;

[0020] Determine the mth harmonic frequency to be eliminated based on the frequency resolution and actual conditions, and calculate the spectrum index corresponding to the harmonic frequency to be eliminated as follows:

[0021]

[0022] Preferably, obtaining the IMF component after filtering out harmonics by inverse Fourier transform comprises the following steps:

[0023] Take X[kh ]=0, when the harmonic frequency does not exactly correspond to an index, the spectrum index k needs to be h The spectrum value of the position is set to zero to obtain the processed spectrum X[k];

[0024] Perform inverse fast Fourier transform on the processed spectrum X[k], convert the processed spectrum X[k] back to the time domain, and obtain the IMF component after filtering out harmonics:

[0025] y[n]=IFFT(X[k]).

[0026] Preferably, the method further comprises reconstructing the signal of the IMF component obtained after filtering out the harmonics and other IMF components not containing harmonic components to obtain the excitation signal after eliminating the harmonics.

[0027] Preferably, the multi-layer IMF function for obtaining the excitation signal comprises the following steps:

[0028] Collect the excitation signal x(t) of the magnetic anisotropy residual stress measuring instrument;

[0029] Identify all local maximum and local minimum points of the excitation signal x(t);

[0030] All local maximum and minimum points are fitted by cubic spline interpolation to obtain the upper envelope e max (t) and the lower envelope e min (t);

[0031] Calculate the average of the upper and lower envelopes:

[0032]

[0033] Subtract the average of the upper and lower envelopes from the excitation signal x(t) to obtain the signal h(t):

[0034] h(t)=x(t)-m(t)

[0035] The excitation signal x(t) is decomposed into multiple IMF components using the empirical mode decomposition method.

[0036] Each IMF component satisfies two conditions: first, the number of extreme points and the number of zero crossings in the entire data segment must be equal or differ by at most one; second, at any point, the average value of the envelope defined by the local maximum and local minimum points is zero;

[0037] Determine whether the signal h(t) meets the conditions of the IMF component. If it does not meet the conditions of the IMF component, use h(t) as a new signal and repeat the above steps until a signal that meets the conditions of the IMF component is obtained, which is recorded as Ci (t), C i (t) is the i-th IMF component;

[0038] Subtracting the resulting IMF component from the excitation signal yields a residual signal r(t) = x(t) - C i (t);

[0039] Update the new signal h(t) through the residual signal r(t) and continue the above decomposition process. Stop the decomposition when the residual signal is a monotonic function.

[0040] The excitation signal x(t) is expressed as the sum of multiple IMF components and a residual component:

[0041]

[0042] Where n is the number of IMF components, r n (t) is the trend term;

[0043] Discard trend items.

[0044] Preferably, determining the IMF components containing harmonic components in the IMF components of each layer comprises the following steps:

[0045] According to the multi-layer IMF function of the excitation signal, the Hilbert transform is performed on the IMF components of each layer to obtain the Hilbert transform value of the IMF components of each layer, and the Hilbert transform value of the IMF components of each layer is calculated to obtain the instantaneous amplitude and instantaneous frequency of the IMF components of each layer;

[0046] According to the instantaneous amplitude and instantaneous frequency of each layer of IMF components, the Hilbert spectrum of each layer of IMF components is plotted; according to the Hilbert spectrum of each layer of IMF components, the IMF components containing harmonic components and the IMF components not containing harmonic components are determined.

[0047] Preferably, obtaining the instantaneous amplitude and instantaneous frequency of the IMF components of each layer specifically includes:

[0048] Perform Hilbert transform on each layer of IMF components to obtain the Hilbert transform value H[C i (t)];

[0049] Definition C i The Hilbert transform of (t) is:

[0050]

[0051] Where PV represents the Cauchy principal value integral, and τ represents the integral variable;

[0052] Through the Hilbert transform value, construct Ci The analytical signal of (t) is:

[0053]

[0054] in, represents the instantaneous amplitude, represents the instantaneous phase, Indicates the instantaneous frequency.

[0055] Preferably, the determination of the IMF components containing harmonic components and the IMF components not containing harmonic components is to take time and instantaneous frequency as two dimensions, and determine the IMF components containing harmonic components and the IMF components not containing harmonic components by drawing the Hilbert spectrum of each layer of IMF components.

[0056] Preferably, the present invention further comprises an excitation signal harmonic filtering system of a residual stress measuring instrument, comprising:

[0057] A signal decomposition module is used to decompose the excitation signal of the magnetic anisotropic residual stress measuring instrument to obtain a multi-layer IMF function of the excitation signal;

[0058] A signal harmonic component determination module is used to determine the IMF components containing harmonic components in each layer of IMF components based on the multi-layer IMF function of the excitation signal;

[0059] The harmonic filtering module is used to obtain the Fourier spectrum of the IMF component containing harmonics, determine the data length and sampling frequency of the IMF component containing harmonics; determine the frequency resolution of the IMF component containing harmonics based on the data length and sampling frequency; determine the harmonic frequency and frequency resolution to be eliminated based on actual conditions, obtain the spectrum index corresponding to the harmonic frequency to be eliminated, and set the spectrum value at the spectrum index position to zero, and obtain the IMF component after filtering out the harmonics through inverse Fourier transform.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] The present invention proposes a method for filtering harmonics of an excitation signal. The method decomposes the read excitation signal to obtain a multi-layer IMF function, which can decompose a complex signal into a series of components called intrinsic mode functions (IMFs), each of which has different frequency and time scale characteristics. According to the Fourier spectrum of the IMF components containing harmonics, the data length and sampling frequency of the IMF components containing harmonics are determined, the frequency resolution of the IMF components containing harmonics is determined, and the frequency and amplitude information of the signal changing over time is obtained. This overcomes some limitations of traditional time-frequency analysis methods (such as Fourier transform), can adaptively process nonlinear and non-stationary signals, more accurately reveal the time-frequency characteristics of the signal, and achieve an accurate description of the time-frequency characteristics of the signal. According to the actual needs of the site, by determining the spectrum index corresponding to the harmonic frequency to be eliminated and setting the spectrum value at the spectrum index position to zero, an inverse Fourier transform is performed to convert the processed spectrum back to the time domain to ensure that the harmonics are effectively filtered out, and the IMF components after filtering out the harmonics are obtained, while also achieving real-time dynamic adjustment of the required frequency components. Furthermore, a pure excitation signal after filtering out harmonics can be obtained through signal reconstruction, which can be used for subsequent magnetic anisotropic residual stress measurement to improve the measurement accuracy and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of the method for filtering harmonics in the excitation signal of the residual stress measuring instrument proposed by the present invention;

[0063] Figure 2 An excitation signal of a magnetic anisotropy residual stress measuring instrument provided in an embodiment of the present invention;

[0064] Figure 3 The result of performing EMD decomposition on the excitation signal provided by the embodiment of the present invention;

[0065] Figure 4 The Hilbert spectrum of each layer of IMF components provided by the embodiment of the present invention is drawn;

[0066] Figure 5 The embodiment of the present invention provides an excitation signal with harmonic components filtered out. DETAILED DESCRIPTION

[0067] The following is a combination of the embodiments of the present invention Figure 1-Figure 5 , the technical solutions in the embodiments of the present invention are clearly and completely described. It should be understood that the terms used in the present invention are only used to describe specific implementation methods and are not intended to limit the present invention.

[0068] In order to accurately obtain the excitation signal with a single frequency component, and to provide the instrument probe with excitation signals of different single frequency values ​​in a timely and rapid manner according to the needs of on-site detection, such as Figure 1 As shown, the present invention proposes a method for filtering harmonics of an excitation signal of a residual stress measuring instrument.

[0069] This method aims to improve the accuracy and reliability of measurement results by designing a more stable and universal filtering method to reduce the uncertainty and instability of processing results caused by the selection of filters, wavelet bases, and window functions. This method effectively filters out harmonic components in the excitation signal, ensuring the authenticity and accuracy of the measured signal, addressing the specific needs of magnetic anisotropic residual stress measurement instruments. This method improves the overall performance of the measuring instrument, broadens its application prospects in the field of magnetic measurement, and provides a more precise measurement method for related industries.

[0070] The method specifically comprises the following steps:

[0071] S1: Signal acquisition and preprocessing

[0072] A high-performance sensor is used to collect or read the excitation signal x(t) of the magnetic anisotropy residual stress measuring instrument.

[0073] S2: Empirical Mode Decomposition (EMD)

[0074] EMD is an adaptive data processing method particularly suitable for analyzing nonlinear and nonstationary signals. Its basic idea is to decompose a complex signal into a series of components called Intrinsic Mode Functions (IMFs). Each IMF component has different frequency and time scale characteristics and meets certain conditions.

[0075] Use EMD to decompose the excitation signal x(t) into multiple IMF components, including:

[0076] Identify all local maximum and local minimum points of the excitation signal x(t);

[0077] All local maximum and minimum points are fitted by cubic spline interpolation to obtain the upper envelope e max (t) and the lower envelope e min (t);

[0078] Calculate the average of the upper and lower envelopes:

[0079]

[0080] Subtract the average of the upper and lower envelopes from the excitation signal x(t) to obtain the signal h(t):

[0081] h(t)=x(t)-m(t)

[0082] Each IMF component must meet two conditions: first, the number of extreme points and zero-crossing points must be equal or differ by at most one within the entire data segment; second, at any point, the average value of the envelope defined by the local maximum and local minimum points must be zero. These conditions ensure that the IMF component has the characteristics of a local, single-frequency component, which can better reflect the local characteristics and variation patterns of the signal.

[0083] Determine whether the signal h(t) meets the conditions of the IMF component. If it does not meet the conditions of the IMF component, use h(t) as a new signal and repeat the above steps until a signal that meets the conditions of the IMF component is obtained, which is recorded as C i (t), C i (t) is the i-th IMF component;

[0084] Subtracting the resulting IMF component from the excitation signal yields a residual signal r(t) = x(t) - C i (t);

[0085] Take the remaining signal as the new signal and continue the above decomposition process. Stop the decomposition when the remaining signal is a monotonic function.

[0086] Through such a decomposition process, the excitation signal x(t) is expressed as the sum of multiple IMF components and a residual component:

[0087]

[0088] Where n is the number of IMF components, r n (t) is the trend term, which is discarded.

[0089] The excitation signal x(t) is decomposed, and the complex excitation signal is decomposed into a series of IMF components with different frequencies and time scales.

[0090] S3: Hilbert-Huang Transform (HHT) Analysis

[0091] HHT is a time-frequency analysis method consisting of two main components: empirical mode decomposition (EMD) and the Hilbert transform (HT). It overcomes some of the limitations of traditional time-frequency analysis methods (such as the Fourier transform), can adaptively process nonlinear and non-stationary signals, and more accurately reveal the signal's time-frequency characteristics.

[0092] The Hilbert transform processes each IMF component to obtain its analytical signal, and then obtains the instantaneous frequency and instantaneous amplitude.

[0093] Specifically, according to the multi-layer IMF function, a Hilbert transform is performed on the IMF components of each layer to obtain the Hilbert transform values ​​of the IMF components of each layer; and according to the Hilbert transform values ​​of the IMF components of each layer, the instantaneous amplitude and instantaneous frequency of the IMF components of each layer are calculated.

[0094] Perform Hilbert transform on each layer of IMF components to obtain the Hilbert transform value H[C i (t)].

[0095] Definition C i The Hilbert transform of (t) is:

[0096]

[0097] Where PV represents the Cauchy principal value integral, and τ represents the integral variable.

[0098] Through the Hilbert transform value, construct C i The analytical signal of (t) is:

[0099]

[0100] in, represents the instantaneous amplitude, represents the instantaneous phase, Indicates the instantaneous frequency.

[0101] In this way, the frequency and amplitude information of the signal changing with time can be obtained, achieving an accurate description of the time-frequency characteristics of the signal.

[0102] Taking time and instantaneous frequency as two dimensions and using color to represent instantaneous amplitude, the Hilbert spectrum of each layer of IMF components is plotted; based on the Hilbert spectrum of each layer of IMF components, the IMF components containing harmonic components and the IMF components not containing harmonic components are determined.

[0103] S4: Direct zeroing method to eliminate harmonics

[0104] Perform Fourier transform (FFT) on the IMF component containing harmonics to obtain the corresponding Fourier spectrum. Based on the corresponding Fourier spectrum, determine the spectrum index corresponding to the harmonic frequency to be eliminated, and set the spectrum value at the spectrum index position to zero. Then, perform inverse Fourier transform to obtain the IMF component after filtering out the harmonics.

[0105] 1) Perform Fourier transform on the IMF components containing harmonics to determine the data length and sampling frequency:

[0106] Let the IMF component containing harmonics be x[n], n=0,1,…,N-1, where N is the number of data points;

[0107] Determine the sampling frequency f s ;

[0108] Use the Fast Fourier Transform algorithm to calculate X[k] = FFT(x[n]), k = 0, 1, ..., N-1;

[0109] When X[k] is a complex number, its amplitude represents the intensity of the sampling frequency component, and its phase represents the phase of the sampling frequency component.

[0110] 2) Determine the spectrum index corresponding to the harmonic frequency to be eliminated

[0111] Calculate the frequency resolution:

[0112]

[0113] Determine the mth harmonic frequency f to be eliminated according to the actual situation h =m×f0, f0 is the fundamental frequency;

[0114] Calculate the spectrum index:

[0115]

[0116] Among them, k h is the spectrum index corresponding to the mth harmonic frequency to be eliminated.

[0117] 3) Set the spectrum value corresponding to the harmonic frequency to zero

[0118] Let X[k h ]=0, when the harmonic frequency does not exactly correspond to an index, the spectrum index k needs to be h The spectrum value of the position is set to zero to obtain the processed spectrum X[k] to ensure that the harmonics are effectively eliminated.

[0119] 4) Perform inverse Fourier transform

[0120] Perform inverse fast Fourier transform on the processed spectrum X[k], convert the processed spectrum X[k] back to the time domain, and obtain the IMF component after filtering out harmonics:

[0121] y[n]=IFFT(X[k])

[0122] S5: Reconstruct the harmonic-filtered IMF components and the harmonic-free IMF to obtain a pure, harmonic-free excitation signal. This signal can be used for subsequent magnetic anisotropic residual stress measurements to improve measurement accuracy and reliability.

[0123] The present invention also proposes an excitation signal harmonic filtering system for a residual stress measuring instrument, comprising:

[0124] A signal decomposition module is used to decompose the excitation signal of the magnetic anisotropic residual stress measuring instrument to obtain a multi-layer IMF function of the excitation signal;

[0125] A signal harmonic component determination module is used to determine the IMF components containing harmonic components in each layer of IMF components based on the multi-layer IMF function of the excitation signal;

[0126] The harmonic filtering module is used to obtain the Fourier spectrum of the IMF component containing harmonics, determine the data length and sampling frequency of the IMF component containing harmonics; determine the frequency resolution of the IMF component containing harmonics based on the data length and sampling frequency; determine the harmonic frequency and frequency resolution to be eliminated based on actual conditions, obtain the spectrum index corresponding to the harmonic frequency to be eliminated, and set the spectrum value at the spectrum index position to zero, and obtain the IMF component after filtering out the harmonics through inverse Fourier transform.

[0127] The proposed method for filtering harmonics from the excitation signal of a residual stress measuring instrument decomposes the read excitation signal to obtain a multi-layer IMF function. This method can decompose complex signals into a series of intrinsic mode function (IMF) components, each with distinct frequency and time-scale characteristics. Based on the Fourier spectrum of the acquired harmonic IMF components, the data length and sampling frequency of the harmonic IMF components are determined, as is the frequency resolution of the harmonic IMF components. This method obtains information about the frequency and amplitude of the signal over time, overcoming some of the limitations of traditional time-frequency analysis methods (such as Fourier transforms). It can adaptively process nonlinear and non-stationary signals, more accurately revealing the signal's time-frequency characteristics and achieving a precise description of the signal's time-frequency characteristics. By determining the spectral index corresponding to the harmonic frequency to be eliminated and setting the spectral value at that spectral index position to zero, an inverse Fourier transform is then performed to convert the processed spectrum back to the time domain to ensure that the harmonics are effectively filtered out. The IMF component after filtering out the harmonics is obtained. Further signal reconstruction can produce a pure, harmonically filtered excitation signal. This signal can be used for subsequent magnetic anisotropic residual stress measurements to improve measurement accuracy and reliability. According to actual field needs, the IMF function containing the corresponding frequency components is retained to achieve harmonic filtering, while also enabling real-time dynamic adjustment of the required frequency components.

[0128] The method for filtering harmonics of the excitation signal of the residual stress measuring instrument proposed in the present invention has the following advantages:

[0129] 1. By accurately filtering out harmonic interference in the excitation signal, the measurement error is effectively reduced, making the magnetic anisotropic residual stress measurement results more accurate and reliable, providing more valuable data support for material performance evaluation and engineering structure analysis.

[0130] 2. This invention incorporates multiple advanced technologies, enabling flexible adjustment of harmonic filtering strategies based on diverse measurement scenarios and requirements. This approach achieves optimal harmonic filtering across complex signal environments and diverse measurement tasks, significantly enhancing its applicability and versatility.

[0131] 3. During harmonic filtering, we prioritize the protection of the excitation signal's fundamental wave and useful components, minimizing the impact on signal integrity. This ensures the sensitivity and stability of the measurement system, enabling the measurement results to truly reflect the material's magnetic anisotropic residual stress state.

[0132] 4. The adaptive nature of EMD and HHT techniques enables the method to adaptively process excitation signals of various types and characteristics, without requiring a prior understanding of the specific characteristics of the signals. This method effectively handles both variations in signal frequency and the complexity of harmonic components, demonstrating strong practical value and reliability.

[0133] Example

[0134] In order to verify the feasibility of the method proposed in the present invention, the excitation signal harmonic reconstruction method of the residual stress measuring instrument proposed in the present invention is verified through the following examples.

[0135] The excitation signal of the magnetic anisotropy residual stress measuring instrument read in this embodiment is as follows: Figure 2 As shown, the excitation signal is decomposed by EMD to obtain 7 IMF components and a trend term, as shown in Figure 3 As shown. Draw the Hilbert spectrum of each layer of IMF components, as shown Figure 4 As shown, the Hilbert spectra of the 7 IMF components correspond to Figure 4 As shown in Figures (a) to (g).

[0136] Depend on Figure 4 As can be seen in Figure (a), the Hilbert spectrum of the IMF1 component clearly shows a harmonic component of 300Hz.

[0137] The IMF1 component is processed by direct zeroing method to eliminate the harmonic components. Then the signal is reconstructed to obtain a new excitation signal as follows Figure 5 shown.

[0138] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

[0139] In addition, unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the art to which the present invention belongs. All documents mentioned in this specification are incorporated by reference to disclose and describe the methods related to the documents. In the event of any conflict with any incorporated document, the content of this specification shall prevail.

Claims

1. A method for filtering harmonics of an excitation signal of a residual stress measuring instrument, characterized in that: The following steps are involved: The excitation signal of the magnetic anisotropic residual stress measuring instrument is decomposed to obtain multiple IMF functions of the excitation signal. Based on the multiple IMF functions of the excitation signal, the IMF components containing harmonic components in each IMF component are determined: Acquisition of excitation signals for magnetic anisotropy residual stress measuring instruments x ( t ); Identify excitation signal x ( t ) all local maximum and local minimum points; All local maximum and minimum points are fitted by cubic spline interpolation to obtain the upper envelope e max ( t ) and the lower envelope e min ( t ); Calculate the upper envelope e max ( t ) and the lower envelope e min ( t )’s average value: From the excitation signal x ( t ) minus the upper envelope e max ( t ) and the lower envelope e min ( t ) to obtain the average value of the signal h ( t ): Using the empirical mode decomposition method, the excitation signal x ( t ) to decompose the signal into multiple IMF components; Each IMF component satisfies two conditions: first, in the entire data segment, the number of extreme points and the number of zero crossing points must be equal or differ by at most one; second, at any point, the upper envelope defined by the local maximum point and the local minimum point e max ( t ) and the lower envelope e min ( t ) has a mean of zero; Judgment signal h ( t ) meets the conditions of the IMF component. If the conditions of the IMF component are not met, h ( t ) as a new signal, repeat the above steps until a signal that meets the IMF component conditions is obtained, which is recorded as C i ( t ), C i ( t ) is the i IMF components; The resulting IMF component is subtracted from the excitation signal to obtain a residual signal ; Through the residual signal r ( t ) Update the new signal and continue the above decomposition process, and stop the decomposition when the remaining signal is a monotonic function; The excitation signal x ( t ) is expressed as the sum of multiple IMF components and a residual component: Where, n is the number of IMF components, r n ( t ) is the trend item; Discard trend items; According to the multiple IMF functions of the excitation signal, each IMF component is subjected to Hilbert transform to obtain the Hilbert transform value of each IMF component, and the Hilbert transform value of each IMF component is calculated to obtain the instantaneous amplitude and instantaneous frequency of each IMF component; According to the instantaneous amplitude and instantaneous frequency of each IMF component, the Hilbert spectrum of each IMF component is plotted; according to the Hilbert spectrum of each IMF component, the IMF components containing harmonic components and the IMF components not containing harmonic components are determined; Obtain the Fourier spectrum of the IMF component containing harmonics and determine the data length of the IMF component containing harmonics: Get the IMF components containing harmonics as x [ n ], n =0,1,…, N -1, where N is the data length; For IMF components containing harmonics x [ n ] Perform Fourier transform and calculate it by fast Fourier transform algorithm x [ n ]'s Fourier spectrum X [ k ]=FFT( x [ n ]), k =0,1,…, N -1; Where, N is the length of IMF component data, X [ k ] is a complex number, whose amplitude represents the intensity of the frequency component and phase represents the phase of the frequency component; According to the IMF component data length and the preset sampling frequency f s , determine the frequency resolution of the IMF component containing harmonics as: Determine the harmonic frequency to be eliminated and the frequency resolution of the IMF component according to the actual situation, and obtain the spectrum index corresponding to the harmonic frequency to be eliminated: Get the actual situation to determine the first m The subharmonic frequency is f h = m × f 0 ,in, f 0 is the fundamental frequency; Determine the first component to be eliminated based on the frequency resolution of the IMF component and the actual situation. m Subharmonic frequency, calculate the spectrum index corresponding to the harmonic frequency to be eliminated: The spectrum value at the spectrum index position is set to zero, and the IMF component after filtering out the harmonics is obtained through inverse Fourier transform: Pick X [ k h ]=0, when the harmonic frequency is not described as an index, the spectrum index k h The spectrum value of the position is set to zero to obtain the processed spectrum X [ k ]; After processing, the spectrum X [ k ] to perform inverse fast Fourier transform and transform the processed spectrum X [ k ] is converted back to the time domain to obtain the IMF component after filtering out harmonics: y [ n ]= IFFT ( X [ k ]) The IMF component after filtering out harmonics is reconstructed with other IMF components that do not contain harmonic components to obtain the excitation signal after eliminating harmonics.

2. The method for filtering harmonics of the excitation signal of the residual stress measuring instrument according to claim 1, characterized in that: The step of obtaining the instantaneous amplitude and instantaneous frequency of each IMF component specifically includes: Perform Hilbert transform on each IMF component to obtain the Hilbert transform value of each IMF component ; definition C i ( t ) is: Where, PV represents the Cauchy principal value integral, represents the integration variable, t for t time; By using the Hilbert transform value, we can construct C i ( t ) is: Where, represents the instantaneous amplitude, represents the instantaneous phase, j represents the imaginary unit, Represents the analysis signal The instantaneous phase.

3. The method for filtering harmonics of the excitation signal of the residual stress measuring instrument according to claim 1, characterized in that: The method of determining the IMF components containing harmonic components and the IMF components not containing harmonic components is to use time and instantaneous frequency as two dimensions, and determine the IMF components containing harmonic components and the IMF components not containing harmonic components by drawing the Hilbert spectrum of each layer of IMF components.

4. A harmonic filtering system for an excitation signal of a residual stress measuring instrument, characterized in that: include: A signal decomposition module is used to decompose the excitation signal of the magnetic anisotropic residual stress measuring instrument to obtain multiple IMF functions of the excitation signal; A signal harmonic component determination module is used to determine the IMF components containing harmonic components in each IMF component based on multiple IMF functions of the excitation signal; Acquisition of excitation signals for magnetic anisotropy residual stress measuring instruments x ( t ); Identify excitation signal x ( t ) all local maximum and local minimum points; All local maximum and minimum points are fitted by cubic spline interpolation to obtain the upper envelope e max ( t ) and the lower envelope e min ( t ); Calculate the upper envelope e max ( t ) and the lower envelope e min ( t )’s average value: From the excitation signal x ( t ) minus the upper envelope e max ( t ) and the lower envelope e min ( t ) to obtain the average value of the signal h ( t ): Using the empirical mode decomposition method, the excitation signal x ( t ) to decompose the signal into multiple IMF components; Each IMF component satisfies two conditions: first, in the entire data segment, the number of extreme points and the number of zero crossing points must be equal or differ by at most one; second, at any point, the upper envelope defined by the local maximum point and the local minimum point e max ( t ) and the lower envelope e min ( t ) has a mean of zero; Judgment signal h ( t ) meets the conditions of the IMF component. If the conditions of the IMF component are not met, h ( t ) as a new signal, repeat the above steps until a signal that meets the IMF component conditions is obtained, which is recorded as C i ( t ), C i ( t ) is the i IMF components; The resulting IMF component is subtracted from the excitation signal to obtain a residual signal ; Through the residual signal r ( t ) Update the new signal and continue the above decomposition process, and stop the decomposition when the remaining signal is a monotonic function; The excitation signal x ( t ) is expressed as the sum of multiple IMF components and a residual component: Where, n is the number of IMF components, r n ( t ) is the trend item; Discard trend items; According to the multiple IMF functions of the excitation signal, each IMF component is subjected to Hilbert transform to obtain the Hilbert transform value of each IMF component, and the Hilbert transform value of each IMF component is calculated to obtain the instantaneous amplitude and instantaneous frequency of each IMF component; According to the instantaneous amplitude and instantaneous frequency of each IMF component, the Hilbert spectrum of each IMF component is plotted; according to the Hilbert spectrum of each IMF component, the IMF components containing harmonic components and the IMF components not containing harmonic components are determined; The harmonic filtering module is used to obtain the Fourier spectrum of the IMF component containing harmonics and determine the data length of the IMF component containing harmonics; Determine the frequency resolution of the IMF component containing harmonics based on the IMF component data length and the preset sampling frequency; Determine the harmonic frequency and frequency resolution to be eliminated based on the actual situation, obtain the spectrum index corresponding to the harmonic frequency to be eliminated, set the spectrum value at the spectrum index position to zero, and obtain the IMF component after filtering out the harmonics through inverse Fourier transform; The obtaining of the Fourier spectrum of the IMF component containing harmonics and determining the data length of the IMF component containing harmonics specifically includes: Get the IMF components containing harmonics as x [ n ], n =0,1,…, N -1, where N is the data length; For IMF components containing harmonics x [ n ] Perform Fourier transform and calculate it by fast Fourier transform algorithm x [ n ]'s Fourier spectrum X [ k ]=FFT( x [ n ]), k =0,1,…, N -1; Where, N is the length of IMF component data, X [ k ] is a complex number, whose amplitude represents the intensity of the frequency component and phase represents the phase of the frequency component; Obtaining the spectrum index corresponding to the harmonic frequency to be eliminated specifically includes: According to the IMF component data length and the preset sampling frequency f s , determine the frequency resolution of the IMF component containing harmonics as: Determine the harmonic frequency to be eliminated and the frequency resolution of the IMF component according to the actual situation, obtain the spectrum index corresponding to the harmonic frequency to be eliminated, and set the spectrum value at the spectrum index position to zero. Then, obtain the IMF component after filtering out the harmonics through inverse Fourier transform: Get the actual situation to determine the first m The subharmonic frequency is f h = m × f 0 ,in, f 0 is the fundamental frequency; Determine the first component to be eliminated based on the frequency resolution of the IMF component and the actual situation. m Subharmonic frequency, calculate the spectrum index corresponding to the harmonic frequency to be eliminated: The spectrum value at the spectrum index position is set to zero, and the IMF component after filtering out harmonics is obtained by inverse Fourier transform, specifically including: Pick X [ k h ]=0, when the harmonic frequency is not described as an index, the spectrum index k h The spectrum value of the position is set to zero to obtain the processed spectrum X [ k ]; After processing, the spectrum X [ k ] to perform inverse fast Fourier transform and transform the processed spectrum X [ k ] is converted back to the time domain to obtain the IMF component after filtering out harmonics: y [ n ]= IFFT ( X [ k ]) The IMF component after filtering out harmonics is reconstructed with other IMF components that do not contain harmonic components to obtain the excitation signal after eliminating harmonics.

Images