Electric leakage detection method based on wavelet transform harmonic suppression

Abstract

The invention discloses an electric leakage detection method based on wavelet transform harmonic suppression, and the method comprises the steps: synchronously collecting a current signal through a dual-channel differential sampling circuit, and obtaining two paths of differential voltages; after common-mode interference is eliminated through differential amplification and impedance matching, low-pass filtering and digital conversion are carried out; extracting signal multi-scale energy characteristics by using wavelet decomposition, and setting a threshold value to separate leakage low-frequency components from high-frequency noise; a pure electric leakage signal is obtained through inverse wavelet reconstruction, and a high-speed switching circuit compares a preset threshold value to trigger a protection action. According to the scheme, differential sampling, impedance matching and wavelet analysis technologies are combined, the signal-to-noise ratio and the detection precision are effectively improved, the problems that in the prior art, electric leakage signals are prone to interference and slow in response speed are solved, and the reliability and the real-time performance of electric leakage protection are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of leakage detection, and in particular relates to a leakage detection method based on wavelet transform harmonic suppression. Background Art

[0002] In power systems, leakage protection is crucial for ensuring personal safety and stable equipment operation. Existing leakage detection methods typically use single-channel current transformers or sampling resistors for signal acquisition, and determine the presence of leakage signals through simple filtering and comparison circuits. While simple and cost-effective, these methods are susceptible to common-mode interference and high-frequency noise in complex electromagnetic environments, resulting in reduced detection accuracy and even misjudgment or missed detections.

[0003] Furthermore, traditional solutions often rely on hardware filters for signal preprocessing, making it difficult to effectively separate leakage signals from harmonic noise. This is especially true in the presence of nonlinear loads or multi-frequency interference, significantly reducing the signal-to-noise ratio, impacting the accuracy and timeliness of protection. Consequently, a major technical challenge facing existing technologies is the difficulty in achieving high-precision extraction and rapid response to leakage signals in complex electromagnetic environments, resulting in insufficient reliability for leakage protection. Summary of the Invention

[0004] The purpose of the present invention is to provide a leakage detection method based on wavelet transform harmonic suppression, improve the stability of signal acquisition through a dual-channel differential sampling circuit, and effectively suppress common-mode interference by combining impedance matching and differential amplification technology to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention adopts the following technical solution: a leakage detection method based on wavelet transform harmonic suppression, comprising the following steps: The current signal is synchronously collected through a dual-channel differential sampling circuit to obtain two differential voltage signals; the differential voltage signal is input into a differential amplifier, common-mode interference is eliminated through a matching impedance network, and an amplified differential voltage is output; the amplified differential voltage is low-pass filtered to suppress high-frequency noise and then sent to the ADC module for digital conversion; the digitally converted signal is subjected to multi-scale analysis using wavelet decomposition to extract the energy distribution characteristics of each frequency band; a threshold is set based on the energy distribution characteristics to separate the low-frequency component and high-frequency noise component related to leakage; the low-frequency component is inversely reconstructed by wavelet to obtain a leakage signal, which is compared with a preset threshold through a high-speed switching circuit to trigger a protection action.

[0006] Preferably, the current signal is synchronously collected by a dual-channel differential sampling circuit to obtain two differential voltage signals, including: High-impedance sampling resistors are connected to both ends of the AC loop to convert the current signal into a voltage signal; the voltage signal is connected to an instrument amplification structure composed of an operational amplifier to output the first and second channel voltages; the first and second channel voltages are phase-calibrated, and the difference between the two channels is calculated to form a differential voltage signal.

[0007] Preferably, the differential voltage signal is input into a differential amplifier, common-mode interference is eliminated through a matching impedance network, and an amplified differential voltage is output, comprising: A four-port impedance matching network is constructed and connected to the positive and negative input terminals of the differential voltage signal; the resistance parameters in the impedance matching network are adjusted to make the equivalent impedances on both sides of the input terminal consistent; the matched signal is sent to an operational amplifier for amplification, and an amplified differential voltage signal is output.

[0008] Preferably, the amplified differential voltage is subjected to low-pass filtering to suppress high-frequency noise and then fed into an ADC module for digital conversion, including: A second-order active low-pass filter structure consisting of an operational amplifier and an RC network is used to connect the amplified differential voltage signal; the cutoff frequency of the low-pass filter structure is set, and the filtered analog voltage signal is input into the ADC module, and a digital signal is output after completing the analog-to-digital conversion.

[0009] Preferably, wavelet decomposition is used to perform multi-scale analysis on the digitally converted signal to extract the energy distribution characteristics of each frequency band, including: The digital signal is subjected to multi-layer wavelet decomposition, and the energy distribution of the corresponding frequency band is calculated for the approximate coefficients and detail coefficients obtained from each layer of decomposition. The energy distribution of each layer is compared to determine the frequency band range where the energy is concentrated; the wavelet coefficients are retained according to the energy distribution results, and the coefficients of the remaining frequency bands are set to zero to complete the signal screening.

[0010] Preferably, setting a threshold value according to the energy distribution characteristics to separate the low-frequency component and high-frequency noise component related to leakage includes: The average energy level is calculated based on the energy of each frequency band, and a dynamic threshold is set based on the average energy; frequency bands above the dynamic threshold are determined to be low-frequency components for leakage analysis; frequency bands below or equal to the threshold are classified as high-frequency noise components, and their coefficients are set to zero.

[0011] Preferably, performing inverse wavelet reconstruction on the low-frequency component includes: The wavelet coefficients corresponding to the low-frequency band are collected, the wavelet coefficients are merged layer by layer according to the original decomposition path, and a reconstruction filter bank is applied; an inverse wavelet transform operation is performed to restore the original time domain signal and output the reconstructed signal waveform.

[0012] Preferably, the high-speed switching circuit compares the leakage signal with a preset threshold value to trigger a protection action, including: The reconstructed signal is input into one end of the voltage comparator, and the other end is connected to a preset reference voltage. When the signal exceeds the preset reference voltage, the comparator outputs a trigger signal; the trigger signal is sent to the high-speed switch drive circuit to control the switching device to be turned on or off; after the circuit breaker operation is performed, the line current is detected to see if it has dropped to a safe range. If it does not meet the standard, the disconnected state is maintained.

[0013] Preferably, the method further comprises: during the protection action, isolating the fault circuit by adopting an emulated IT power supply mode to suppress transient voltage shocks.

[0014] Preferably, the method further comprises: dynamically adjusting the bias voltage of the differential sampling circuit according to the amplitude and phase information of the leakage signal.

[0015] Technical effects and advantages of the present invention: The leakage detection method based on wavelet transform harmonic suppression proposed in the present invention has the following advantages over the prior art: This invention improves signal acquisition stability through a dual-channel differential sampling circuit, effectively suppressing common-mode interference by combining impedance matching and differential amplification techniques. It further incorporates wavelet transforms to achieve multi-scale signal analysis and noise separation, thereby improving the signal-to-noise ratio of leakage signals. Finally, a high-speed switching circuit enables rapid protection. This solution effectively addresses existing issues such as leakage signals being susceptible to interference, low detection accuracy, and slow response speed, significantly improving the accuracy and real-time nature of leakage detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The figure is a flow chart of the leakage detection method based on wavelet transform harmonic suppression of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0018] The present invention provides Figure 1 The present invention provides a leakage detection method based on wavelet transform harmonic suppression, comprising the following steps: Step 1: Synchronously collect the current signal through a dual-channel differential sampling circuit to obtain two differential voltage signals; specifically, the following steps are included: High-impedance sampling resistors are connected to both ends of the AC loop, and Ohm's law (U=I×R) is used to convert the current signal into a voltage signal. This achieves safe and stable conversion of current signals to voltage signals, avoiding the interference and loss caused by direct current measurement.

[0019] The voltage signal is connected to the instrumentation amplifier structure composed of an operational amplifier to output the first and second channel voltages V1 and V2. The instrumentation amplifier is a differential amplifier structure with high common-mode rejection ratio and high input impedance. Its basic gain relationship is: , where G is the magnification, 、 In this step, the voltages obtained by the two sampling resistors are respectively connected to the two input terminals of the instrumentation amplifier, which outputs the corresponding first channel voltage V1 and second channel voltage V2.

[0020] Phase calibration is performed on V1 and V2 to ensure that they meet the phase consistency condition, that is, |φ(V1)-φ(V2)|<θ, where θ is the preset angle threshold. This eliminates the error caused by phase mismatch between channels and improves the accuracy and stability of differential signal extraction.

[0021] Calculating the differential voltage , and serves as the input signal for subsequent processing. This method can effectively cancel out common-mode noise (such as electromagnetic induction and ground noise) in the two channels, retaining the true differential-mode signal, that is, the effective signal reflecting the leakage characteristics.

[0022] Step 2: Input the differential voltage signal into a differential amplifier, eliminate common-mode interference through a matching impedance network, and output an amplified differential voltage; specifically, the following steps are included: A four-port impedance matching network consisting of precision resistors is constructed, where two resistors are connected to the positive input terminal and the other two are connected to the negative input terminal to adjust the equivalent impedance of the input path.

[0023] Adjust the resistance values ​​in the impedance matching network so that the equivalent impedance at the input terminal satisfies = , to achieve input balance; : Positive input impedance; : Negative input impedance; This process is completed based on the calculation relationship between parallel and series resistance. For example, for two parallel resistors R1 and R2, the equivalent impedance is: By rationally designing the resistance of each branch, the and consistency.

[0024] The matched signal is sent to the operational amplifier of the differential structure, and the amplification factor is ,in is the feedback resistor, is the input resistance; this structure can effectively amplify the voltage difference between the two input terminals while suppressing common-mode noise.

[0025] Output amplified differential voltage , as the basic signal for the next stage of processing. : The voltage value of the positive input terminal of the differential amplifier; : The voltage at the negative input of the differential amplifier. This formula reflects the core function of the differential amplifier: to amplify only the voltage difference between the two inputs, while ignoring the common-mode component that varies with them.

[0026] Step 3: Low-pass filtering the amplified differential voltage to suppress high-frequency noise and then send it to the ADC module for digital conversion; specifically, the following steps are included: A second-order active low-pass filter structure consisting of an operational amplifier and an RC (resistance-capacitance) network is used to process the amplified differential voltage signal output by the previous stage. This structure not only has good amplitude-frequency response characteristics, but also provides a certain voltage gain, while effectively suppressing high-frequency noise components and preserving the useful signal frequency band.

[0027] Set the cutoff frequency of the low-pass filter structure , where R is the resistance value and C is the capacitance value. By properly selecting the values ​​of R and C, the filter can be set to start attenuating high-frequency components at a specific frequency point, thereby effectively suppressing noise outside the target frequency band.

[0028] The filtered analog voltage signal is input into a 12-bit successive approximation ADC module to complete quantization processing; the filtered analog voltage signal can be converted into a digital quantity with 4096 levels of resolution.

[0029] Output digital signal: , as input data for subsequent wavelet analysis; : Filtered voltage. This achieves standardized conversion from analog signals to standard digital formats, allowing the signal to be directly processed by subsequent wavelet transform algorithms, improving the consistency and operability of data analysis.

[0030] Step 4: Use wavelet decomposition to perform multi-scale analysis on the digitally converted signal to extract the energy distribution characteristics of each frequency band; specifically, the following steps are included: Selecting orthogonal wavelet basis functions, perform a three-layer discrete wavelet decomposition on the digitized signal x(n); decomposing the input signal x(n) into approximate coefficients and detail coefficients at multiple scales. This step uses a three-layer decomposition structure to divide the original signal into multiple subbands, from low to high frequency, enabling detailed characterization of the signal at different time-frequency scales.

[0031] The approximate coefficients obtained by decomposing each layer and detail coefficient , calculate the energy of the corresponding frequency band ; This energy calculation method reflects the intensity distribution of the signal within the frequency band.

[0032] Compare the energy distribution of detail coefficients of each layer, determine the frequency band with concentrated energy, and mark it as the frequency band to be processed; specifically: the energy of each layer obtained in the previous step A horizontal comparison is performed to identify frequency bands with significantly higher energy than other layers. These frequency bands are considered to contain the main components of the useful leakage signal and are therefore marked as "bands to be processed." This process is based on the principle of energy comparison and does not require a fixed threshold, making it highly adaptive.

[0033] Based on the labeling results, the wavelet coefficients of specific frequency bands are retained, while the coefficients of other frequency bands are set to zero to complete the signal component screening. By selectively retaining the wavelet coefficients of key frequency bands, the signal-to-noise ratio of the leakage signal is significantly improved, providing a purer and more reliable signal foundation for subsequent reconstruction and protection actions.

[0034] Step 5: Setting a threshold value based on the energy distribution characteristics to separate the low-frequency component and high-frequency noise component related to leakage; specifically comprising the following steps: After completing the wavelet decomposition and extracting the energy of each layer's detail coefficients, the energy values ​​corresponding to the three frequency bands (layer 1 to layer 3) are converted to Sum them up and take the average value μ to characterize the overall energy level of the signal: ; By calculating the average energy μ, a reference benchmark reflecting the overall signal strength can be established.

[0035] The dynamic threshold T is set to μ*k, where k is the empirical proportional factor, ranging from 1.2 to 1.8. This method enables the threshold to be adaptively adjusted as the input signal energy changes, making it suitable for leakage detection scenarios under different working conditions.

[0036] For each layer’s detail coefficient energy , and compare it with the dynamic threshold T. If the energy of a layer If it is greater than T, it is considered that the frequency band contains more effective information and belongs to the low-frequency component. The corresponding approximate coefficient and detail coefficient are retained as the basis for subsequent leakage characteristic analysis.

[0037] Frequency bands with energy less than or equal to T are classified as high-frequency noise components, and the corresponding coefficients are set to zero. By removing high-frequency noise components, the signal-to-noise ratio of the leakage signal is significantly improved, and the interference of redundant information on subsequent processing is reduced.

[0038] Step 6: Perform inverse wavelet reconstruction on the low-frequency component to obtain a leakage signal; specifically, the following steps are included: After completing the wavelet decomposition and energy analysis, the wavelet detail coefficients and approximate coefficients corresponding to the low-frequency components are collected; the wavelet coefficients are merged layer by layer according to the original decomposition path, and a preset reconstruction filter group (a combination of low-pass and high-pass filters) is applied to gradually restore the time resolution of the signal.

[0039] Execute the inverse wavelet transform formula , where CA is the three-layer approximation coefficient, is the detail coefficient of the jth layer; this formula converts the wavelet coefficients in the multi-scale domain back to the original time domain signal through a step-by-step synthesis method.

[0040] Output the reconstructed time domain signal waveform , as the basis for subsequent protection judgment. It is a complete time domain waveform that reflects the main features related to leakage in the original current signal.

[0041] Step 7: Compare the leakage signal with the preset threshold through the high-speed switching circuit to trigger the protection action; specifically, the following steps are included: The leakage signal after wavelet reconstruction is measured by voltage comparator With a preset reference voltage The comparator has two input terminals, one of which is connected to , the other end is connected to a fixed or adjustable reference voltage .when Exceed When the comparator output state changes.

[0042] when At When the comparator outputs a high-level signal Trig = 1, indicating that the protection condition is met; otherwise, it remains at a low level. This binary output can be used to drive the subsequent control circuit.

[0043] The Trig=1 signal is sent to the high-speed switch drive circuit to control the MOSFET device to turn on or off; as an electronic switch, MOSFET has the advantages of fast response, long life, and no sparks.

[0044] After the MOSFET is disconnected, the system continues to monitor the residual current in the main circuit to detect whether the line current drops to a safe range. If not, the disconnected state is maintained to prevent the device from being powered on again and causing secondary risks.

[0045] Step 8: During protection operation, use the simulated IT power supply mode to isolate the fault circuit and suppress transient voltage shocks. The specific steps include: After detecting the protection signal Trig=1, the primary side switch of the dual-winding isolation transformer is started; the secondary side of the transformer maintains the load power supply, while the primary side is disconnected from the main power supply, forming a floating grounding state; after the primary side is disconnected, the secondary side can still maintain power supply to the load for a period of time. At the same time, due to the lack of a grounding point, the system enters a "floating grounding" state, that is, the ground potential is not fixed, thereby suppressing the propagation of transient voltage surges.

[0046] Control ground resistance Connect to the neutral point to limit the rate of change of ground current ,in is the bus voltage, is the line inductance; : ground resistance; t: time; as t increases, the denominator increases, Reduce, thereby suppressing transient current impact. Monitor the residual voltage of the line in the floating state ,like Drop to safety threshold Below, reclosing is allowed.

[0047] Step nine: dynamically adjusting the bias voltage of the differential sampling circuit according to the amplitude and phase information of the leakage signal; specifically comprising the following steps: After the leakage signal is reconstructed by wavelet, the reconstructed signal is extracted using digital signal processing algorithms (such as peak detection and fast Fourier transform) Peak With the initial phase angle ; This information reflects the energy intensity and time starting position of the current signal.

[0048] according to and Calculating the offset adjustment ; This formula combines the amplitude and phase information of the signal to produce a bias adjustment that varies with the signal.

[0049] The ΔV is output to the reference input of the differential sampling circuit through the digital-to-analog converter and superimposed on the original bias voltage. Above this, a new bias control voltage is formed.

[0050] Updated bias voltage , used for level alignment during the next round of current signal acquisition. This voltage serves as the reference level for the next cycle of signal acquisition, ensuring that the differential sampling circuit remains in its optimal operating range. This implements a closed-loop adaptive bias voltage adjustment mechanism, significantly improving the system's responsiveness to signal changes and acquisition accuracy during continuous operation, enhancing the robustness and reliability of the entire leakage detection system.

[0051] In order to more clearly illustrate the operation steps of the above-mentioned leakage detection method based on wavelet transform harmonic suppression, the following will be further described in conjunction with an embodiment, as follows: Consider an industrial site with multiple nonlinear loads (such as inverters and rectifiers), resulting in significant harmonic interference in the power lines. A 220V AC circuit requires leakage detection and protection to prevent accidents caused by insulation aging or short circuits. The following process is used to collect, process, and initiate protection actions for leakage signals.

[0052] Step 1: Dual-channel differential sampling and signal preprocessing Current to voltage: Connect two high-impedance sampling resistors (R1=R2=1kΩ) at both ends of the AC circuit; Use Ohm's law U=I×R to convert the current flowing through the line into a voltage signal; Assuming there is a weak leakage current I=5mA in the current circuit, the corresponding voltage U=5V.

[0053] Instrumentation amplifier gain setting: Use an instrumentation amplifier and set the gain factor G to 10. The first channel voltage V1 is 50mV, and the second channel voltage V2 is 49mV (due to differences in transmission paths, there is a certain deviation). Phase calibration: The phase difference between the two channels is detected to be |φ(V1)-φ(V2)|=3°, which is less than the preset threshold θ=5°, meeting the consistency condition; Differential Voltage Calculation: Calculate the differential voltage =V1-V2=1mV; output this differential voltage as input for subsequent processing.

[0054] Step 2: Differential Amplification and Common-Mode Rejection Build an impedance matching network: Use four precision resistors to form a four-port structure. = =10kΩ; Difference Amplifier Configuration: Setting the Feedback Resistor =100kΩ, input resistance =10kΩ; amplification factor =100k / 10k=10; Output amplified differential voltage: input voltage =0.5V, =0.49V; output voltage after amplification =10×0.01V=0.1V; output to the next level filter module.

[0055] Step 3: Low-pass filtering and ADC digitization Design a low-pass filter: Use an RC second-order active low-pass structure and set the cutoff frequency ; Take R = 10kΩ, C = 10nF, then =1.6kHz, which can effectively filter out high-frequency noise; Analog signal quantization: using a 12-bit successive approximation ADC module with a reference voltage range of 0-3.3V; the filtered voltage =0.8V; digitized result D=(0.8-0)*(4095 / (3.3-0))=992; output digital signal D=992 for wavelet analysis.

[0056] Step 4: Wavelet decomposition to extract energy features Select wavelet basis function: select Daubechiesdb4 wavelet and perform three-layer discrete wavelet decomposition on the signal x(n); Calculate the energy of each layer: First layer detail coefficient energy =200; Second layer detail coefficient energy =150; The third layer detail coefficient energy =100; Mark the frequency band with concentrated energy: calculate the average energy μ = (200 + 150 + 100) / 3 = 150; set the dynamic threshold T = μ × k = 150 × 1.5 = 225; if the comparison result is: > T, retain the first layer coefficients; set the rest to zero.

[0057] Step 5: Inverse wavelet reconstruction to obtain pure leakage signal Collect and merge wavelet coefficients: only retain the first layer detail coefficients and the three-layer approximation coefficient CA; Perform inverse wavelet transform: Formula: = (CA,CD_1,0,0); get the reconstructed signal x_recon, which is a clear leakage signal waveform; Output time domain signal: =The amplitude is about 0.7V, which reflects the actual leakage component.

[0058] Step 6: Compare and trigger protection actions Set comparison threshold: preset reference voltage =0.5V; Comparative judgment: =0.7V> , comparator output Trig=1; Control MOSFET to disconnect: trigger the high-speed switch drive circuit, MOSFET turns on, and the main circuit power supply is disconnected; Monitor residual current: Check whether the residual current drops to a safe range after power failure =10mA; if the standard is not met, the device will remain disconnected to prevent secondary risks.

[0059] Step 7: Floating Ground and Transient Voltage Suppression Start the isolation transformer: the primary side is disconnected, the secondary side continues to supply power, and enters the floating grounding state; Connecting grounding resistor: Access =100Ω, bus voltage = 310V, line inductance =1mH; Transient current change rate: ; When t=1ms, =310 / (0.001+100*0.001)=310 / 0.101=3069A / s; significantly lower than the impact current when no grounding resistance is added, effectively suppressing transient voltage.

[0060] Reclosing allowed: Residual pressure Down to =50V or less, confirm that the fault has been eliminated and allow power to be restored.

[0061] Step 8: Adaptive adjustment of bias voltage Extracting peak and phase: from reconstructed signal Extract peak values =0.7V, initial phase angle = 45°; Calculate the offset adjustment: =0.35×sin(45°)=0.247V; Update bias voltage: original bias voltage =1.65V; The new bias voltage = 1.65 + 0.247 = 1.897 V; this is used for the next round of sampling to improve sampling accuracy.

[0062] The complete execution of the above steps demonstrates that this method has high precision, strong anti-interference ability, and a fast response mechanism. It is suitable for leakage detection scenarios in complex electromagnetic environments and has good engineering application prospects.

[0063] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A leakage detection method based on wavelet transform harmonic suppression, characterized in that: The following steps are involved: The current signal is synchronously collected through a dual-channel differential sampling circuit to obtain two differential voltage signals; Inputting the differential voltage signal into a differential amplifier, eliminating common-mode interference through a matching impedance network, and outputting an amplified differential voltage; The amplified differential voltage is subjected to low-pass filtering to suppress high-frequency noise and then sent to an ADC module for digital conversion; The digitally converted signal is analyzed at multiple scales using wavelet decomposition to extract the energy distribution characteristics of each frequency band. Setting a threshold value according to the energy distribution characteristics to separate the low-frequency component and high-frequency noise component related to leakage; The low-frequency component is subjected to inverse wavelet reconstruction to obtain a leakage signal, which is compared with a preset threshold value through a high-speed switching circuit to trigger a protection action.

2. The leakage detection method based on wavelet transform harmonic suppression according to claim 1 is characterized in that: The current signal is synchronously collected through a dual-channel differential sampling circuit to obtain two differential voltage signals, including: Connect high-impedance sampling resistors at both ends of the AC loop to convert the current signal into a voltage signal; Connecting the voltage signal to an instrument amplification structure composed of an operational amplifier to output first and second channel voltages; Phase calibration is performed on the first and second channel voltages, and a difference between the two channels is calculated to form a differential voltage signal.

3. The leakage detection method based on wavelet transform harmonic suppression according to claim 1, characterized in that: Inputting the differential voltage signal into a differential amplifier, eliminating common-mode interference through a matching impedance network, and outputting an amplified differential voltage, including: Constructing a four-port impedance matching network and connecting it to the positive and negative input terminals of the differential voltage signal; Adjusting the resistance parameters in the impedance matching network to make the equivalent impedances on both sides of the input end consistent; The matched signal is sent to the operational amplifier for amplification, and an amplified differential voltage signal is output.

4. The leakage detection method based on wavelet transform harmonic suppression according to claim 1, characterized in that: The amplified differential voltage is subjected to low-pass filtering to suppress high-frequency noise and then sent to the ADC module for digital conversion, including: A second-order active low-pass filter structure composed of an operational amplifier and an RC network is used to receive the amplified differential voltage signal; The cutoff frequency of the low-pass filter structure is set, the filtered analog voltage signal is input to the ADC module, and a digital signal is output after analog-to-digital conversion is completed.

5. The leakage detection method based on wavelet transform harmonic suppression according to claim 4 is characterized in that: Wavelet decomposition is used to perform multi-scale analysis on the digitally converted signal to extract the energy distribution characteristics of each frequency band, including: Performing multi-layer wavelet decomposition on the digital signal, calculating the energy distribution of the corresponding frequency band for the approximate coefficients and detail coefficients obtained from each layer of decomposition, comparing the energy distribution of each layer, and determining the frequency band range where the energy is concentrated; According to the energy distribution results, the wavelet coefficients are retained and the coefficients of the remaining frequency bands are set to zero to complete the signal screening.

6. The leakage detection method based on wavelet transform harmonic suppression according to claim 1, characterized in that: Setting a threshold value based on the energy distribution characteristics to separate the low-frequency component and high-frequency noise component related to leakage, including: Calculating an average energy level based on the energy of each frequency band, and setting a dynamic threshold based on the average energy; The frequency band above the dynamic threshold is determined as low-frequency component for leakage analysis; Frequency bands below or equal to the threshold are classified as high-frequency noise components, and their coefficients are set to zero.

7. The leakage detection method based on wavelet transform harmonic suppression according to claim 6, characterized in that: Performing inverse wavelet reconstruction on the low-frequency component includes: Collect the wavelet coefficients corresponding to the low-frequency band, merge the wavelet coefficients layer by layer according to the original decomposition path, and apply a reconstruction filter bank; Perform inverse wavelet transform operation to restore the original time domain signal and output the reconstructed signal waveform.

8. The leakage detection method based on wavelet transform harmonic suppression according to claim 1, characterized in that: The high-speed switching circuit compares the leakage signal with the preset threshold and triggers the protection action, including: The reconstructed signal is input into one end of the voltage comparator and the other end is connected to a preset reference voltage. When the signal exceeds the preset reference voltage, the comparator outputs a trigger signal. Sending the trigger signal to the high-speed switch drive circuit to control the switch device to be turned on or off; After the circuit breaker is executed, the circuit breaker checks whether the line current drops to a safe range. If it does not, the circuit breaker remains in the disconnected state.

9. The leakage detection method based on wavelet transform harmonic suppression according to claim 1, characterized in that: The method further includes: during the protection action, isolating the fault circuit by adopting an emulated IT power supply mode to suppress transient voltage impact.

10. The leakage detection method based on wavelet transform harmonic suppression according to claim 1, characterized in that: The method further includes dynamically adjusting a bias voltage of a differential sampling circuit according to the amplitude and phase information of the leakage signal.

Images