Noise suppression method and system for acoustic-magnetic positioning of underground cable fault

Abstract

The invention discloses a noise suppression method and system for acoustic-magnetic positioning of underground cable faults. The method comprises the following steps: step 1, reading acoustic-magnetic signals; step 2, performing wavelet transform filtering on the obtained sound signal with noise; step 3, estimating an average amplitude spectrum of the noise in a non-effective sound segment of the noise-containing sound signal after wavelet transform filtering; 4, acquiring a power spectrum of the effective sound signal by adopting a spectrum subtraction algorithm; step 5, utilizing the property that the effective sound signal is not related to the noise to perform self-correlation processing on the sound signal with the noise so as to obtain a self-correlation frame sequence; and step 6, estimating the difference between the propagation time and the period of the acoustic magnetic signal, reducing the range of the effective sound signal, and generating an enhanced sound signal output through inverse fast Fourier transform. The noise suppression method and system for acoustic magnetic positioning of the underground cable fault are fast in calculation, high in accuracy and good in denoising effect.

Description

Technical Field

[0001] The invention belongs to the technical field of underground cable fault locating, and in particular relates to a noise suppression method and system for acoustic and magnetic locating of underground cable faults. Background Art

[0002] As power cables are used for a long time, the probability of failure will become higher and higher. Because they are buried underground, it is difficult to quickly determine the exact location of the fault point. Failure to repair the fault and restore power supply in time will cause great inconvenience to people's lives and huge economic losses to industrial production. Therefore, efficiently and accurately locating the fault point of the power cable has important social and practical significance.

[0003] Among the cable fault location methods, the acoustic-magnetic synchronization method is an effective method. The principle of the acoustic-magnetic synchronization method is to use the detection time difference caused by the different propagation speeds of the discharge magnetic field signal and the sound signal to locate the position of the cable fault point. The discharge magnetic field signal is obtained through the induction coil using the principle of electromagnetic induction, which is not easily interfered with and can be reliably identified. The sound signal is generated by the vibration caused by the impact discharge and is easily affected by the surrounding environmental noise, making the discharge sound signal difficult to detect and identify, resulting in inaccurate fault location.

[0004] The enhancement effect of spectral subtraction in non-stationary noise environment and low signal-to-noise ratio is not satisfactory due to the influence of sound signal distortion and residual noise. Therefore, the estimated noise can be updated based on the discharge magnetic field signal and autocorrelation effective sound sequence recognition, thereby improving the denoising performance.

[0005] To this end, we proposed a noise suppression method for acoustic and magnetic positioning of underground cable faults to solve the above problems. Summary of the invention

[0006] The present invention mainly aims at the deficiencies of the existing denoising technology in acoustic and magnetic locating of underground cable faults, and provides a noise suppression method and system for acoustic and magnetic locating of underground cable faults with fast calculation, high accuracy and good denoising effect.

[0007] A technical solution to achieve the above purpose is: a noise suppression method for underground cable fault acoustic and magnetic positioning, the specific steps are as follows:

[0008] Step 1, reading the acoustic magnetic signal, the acoustic magnetic signal comes from the acoustic magnetic synchronous signal with noise captured by the acoustic magnetic sensor during underground cable fault detection;

[0009] Step 2, performing wavelet transform filtering on the obtained noisy sound signal;

[0010] Step 3, estimating the average amplitude spectrum of the noise in the effective sound segment of the noisy sound signal after wavelet transform filtering;

[0011] Step 4, using a spectrum subtraction algorithm to obtain the power spectrum of the effective sound signal;

[0012] Step 5, using the property that the effective sound signal and the noise are uncorrelated, autocorrelation processing is performed on the noisy sound signal to obtain an autocorrelation frame sequence;

[0013] Step 6, estimate the propagation time and period difference of the acoustic and magnetic signals, narrow the range of the effective sound signal, and generate an enhanced sound signal output through inverse fast Fourier transform.

[0014] Furthermore, in step 1, the source of the acoustic-magnetic signal is to connect an impact generator to the terminal head of the faulty cable at one end of the faulty cable. The impact generator sends an impact signal to one end of the faulty cable and generates a flashover impact discharge at the fault point. The acoustic-magnetic signal generated by the discharge will diffuse and propagate in the covering soil above. Then, the acoustic-magnetic sensor probe is inserted into the ground. The acoustic-magnetic sensor captures the magnetic signal and acoustic signal generated by the discharge at the fault point as well as other environmental noise.

[0015] Furthermore, in step 2, the wavelet transform filtering method is to use wavelet transform to decompose the noisy sound signal into wavelet coefficients at different scales; select an appropriate threshold to identify the noise; set the noise coefficient to zero to remove the noise. Use the processed wavelet coefficients to reconstruct the signal through inverse wavelet transform to generate a preliminary denoised signal.

[0016] Furthermore, in step 4, the spectrum subtraction algorithm for obtaining the power spectrum of the effective sound signal is as follows:

[0017]

[0018] Among them, |X i (w)| 2 Indicates the effective sound signal x i (n) power spectrum; i represents the frame number; |Y i (w)| 2 Represents the noisy signal y i The power spectrum of (n); |D i (w)| 2 represents the noise signal d i (n) power spectrum.

[0019] Furthermore, the power spectrum of the noise in the effective sound segment is replaced by the estimation of the average amplitude spectrum of the non-effective sound segment. i (w)| 2 .

[0020] Furthermore, in step 5, using the property that the effective sound signal and the noise are uncorrelated, the autocorrelation function formula of the noisy sound signal is as follows:

[0021]

[0022] Wherein, N is the length of each frame after the signal is framed, and k is the delay amount.

[0023] Furthermore, in step 6, the propagation time difference t of the acoustic magnetic signal generated by the discharge 1 Determined by the detection range, the duration of the discharge sound signal t 2 , so the effective range of the discharge sound signal can be narrowed to the inner range of the electromagnetic signal 1 +t 2 .

[0024] Furthermore, in step 6, the enhanced sound signal generated by the inverse fast Fourier transform is determined by the power spectrum obtained by the spectrum subtraction algorithm, the phase spectrum of the noisy sound signal, the autocorrelation frame sequence, and the effective range.

[0025] Furthermore, the noise energy is mainly concentrated near the zero-order autocorrelation coefficient of the noisy sound signal frame, and the maximum value or the main-to-sub-peak ratio of the autocorrelation function between the noisy sound signal and the estimated noise is used to determine whether the current processed signal frame is a valid sound signal frame.

[0026] Another technical solution to achieve the above-mentioned purpose is: a noise suppression system for acoustic and magnetic positioning of underground cable faults, used to implement the noise suppression method in the acoustic and magnetic positioning of underground cable faults, including an electronic device and a computer-readable storage medium; the electronic device includes a processor, a memory and a communication bus, and the memory stores a computer-readable program that can be executed by the processor; the communication bus realizes the connection and communication between the processor and the memory; the processor is used to execute the computer-readable program; the computer-storable medium stores one or more programs, and the one or more programs can be executed by one or more processors.

[0027] Beneficial effects of the present invention:

[0028] 1. The noise suppression method for acoustic and magnetic positioning of underground cable faults constructed by the present invention aims to solve the problem that the discharge sound signal at the cable fault point is easily interfered by environmental noise and reduces the detection effect. The autocorrelation characteristics of the noisy sound signal and the discharge magnetic field signal are used for real-time noise estimation to suppress environmental noise, highlight the discharge sound signal and improve the signal-to-noise ratio. The method can effectively suppress environmental noise and improve the detection effect of the discharge sound signal.

[0029] 2. The enhanced sound signal finally generated by the present invention is jointly determined by the power spectrum obtained by the spectral subtraction algorithm, the phase spectrum of the noisy sound signal, the autocorrelation frame sequence, and the effective range. It makes full use of the noise information and the magnetic signal information in the underground cable fault acoustic and magnetic positioning technology. No new noise will be generated in the process of noise removal, and the processing effect is more stable.

[0030] 3. The present invention can effectively improve the signal-to-noise ratio of the sound signal waveform and easily identify the range of the effective sound signal by performing short-time autocorrelation processing on the noisy sound signal through wavelet transform filtering and windowing and framing. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Shown is a schematic diagram of the overall process of an embodiment of the present invention;

[0032] Figure 2 The display is a wavelet transform filtering result diagram of the simulated sampling data of an embodiment of the present invention;

[0033] Figure 3 Shown is a diagram of enhanced results obtained after processing of simulated sampling data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to better understand the technical solution of the present invention, the following is a detailed description through specific embodiments:

[0035] See also Figures 1 to 3 The drawings in this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0036] The noise suppression method for underground cable fault acoustic and magnetic positioning described in the present invention mainly removes the noise in the sound signal from the acoustic and magnetic synchronous signal with noise captured by the acoustic and magnetic sensor during underground cable fault detection to obtain an enhanced sound signal, which can be expressed by the following formula:

[0037] y i (n) = x i (n)+d i (n)

[0038] Among them, i represents the number of frames after the signal is divided into frames; n represents the time series; x i (n) indicates a valid sound signal; y i (n) represents a noisy signal; d i(n) represents the noise signal.

[0039] The source of the acoustic magnetic signal is to connect the impact generator to the terminal head of the faulty cable at one end of the faulty cable. The impact generator sends an impact signal to one end of the faulty cable and generates a flashover impact discharge at the fault point. The acoustic magnetic signal generated by the discharge will spread in the covering soil above. Then the acoustic magnetic sensor probe is inserted into the ground. The acoustic magnetic sensor captures the magnetic signal and acoustic signal generated by the discharge at the fault point and other environmental noise. The present invention aims to reduce the noise of the noisy sound signal and restore the effective sound signal as much as possible.

[0040] The present invention first uses wavelet transform to decompose the noisy sound signal into wavelet coefficients at different scales; selects an appropriate threshold to identify the noise; and sets the noise coefficient to zero to remove the noise. Using the processed wavelet coefficients, the signal is reconstructed by inverse wavelet transform to generate a preliminary denoised signal. Since the effective sound signal and the noise signal are independent of each other, their statistical means are zero. The formula for the power spectrum of the effective sound signal is as follows:

[0041]

[0042] Among them, |X i (w)| 2 Indicates the effective sound signal x i The power spectrum of (n); |Y i (w)| 2 Represents the noisy signal y i The power spectrum of (n); |D i (w)| 2 represents the noise signal d i (n) power spectrum.

[0043] The key to this algorithm is to obtain the amplitude spectrum of the noise |D i (w)| 2 , but it cannot be obtained directly. It can be replaced by estimating the average amplitude spectrum of the non-effective sound segment. According to the characteristic that the human ear is insensitive to phase change, the phase spectrum of the noisy sound signal is used, combined with the obtained power spectrum and inverse fast Fourier transform to restore the enhanced effective sound time domain signal. In the specific operation, since the estimated noise is not the real noise at the corresponding moment, the result may be negative, but the amplitude spectrum or power spectrum should not be negative, so the negative result is set equal to zero.

[0044] The effective sound discrimination based on autocorrelation processing is to use the fact that the sound signal itself is correlated, while the sound signal and noise, and noise and noise are uncorrelated. By doing autocorrelation processing on the noisy signal, we can get the same autocorrelation frame sequence as the sound signal without noise, and use the autocorrelation coefficient as the feature of discriminating effective sound. The autocorrelation calculation function of the noisy sound signal is:

[0045]

[0046] Wherein, N is the length of each frame after the signal is framed, and k is the delay amount.

[0047] Using the property that the effective sound signal and the noise are uncorrelated, the autocorrelation function formula of the noisy sound signal after expansion and simplification is as follows:

[0048]

[0049] The noise energy is mainly concentrated near the zero-order autocorrelation coefficient of the noisy sound signal frame. The maximum value of the autocorrelation function between the noisy sound signal and the preceding estimated noise or the ratio of the main peak to the secondary peak is used to determine whether the current processing signal frame is a valid sound frame.

[0050] The propagation time difference t of the acoustic and magnetic signals generated by the discharge 1 Determined by the detection range, the duration of the discharge sound signal t 2 , so the effective range of the discharge sound signal can be reduced to the t 1 +t 2 within the range.

[0051] The enhanced sound signal generated by inverse fast Fourier transform is determined by the power spectrum obtained by the spectral subtraction algorithm, the phase spectrum of the noisy sound signal, the autocorrelation frame sequence, and the effective range.

[0052] The above technical solution is further described below with reference to the accompanying drawings and examples:

[0053] Obtain analog raw sampling data, including magnetic signals and noisy sound signals;

[0054] The wavelet transform filtering method for sampling noisy audio signals is used for preliminary noise reduction. The results are as follows: Figure 3 As shown;

[0055] The noise power spectrum of the segment without effective sound is calculated, and the power spectrum of the effective sound is calculated; through autocorrelation calculation and the effective sound range set according to the magnetic signal, an enhanced effective sound signal time domain diagram is generated.

[0056] Those skilled in the art should recognize that the above embodiments are only used to illustrate the present invention, and are not intended to limit the present invention. As long as they are within the spirit of the present invention, any changes or modifications to the above embodiments will fall within the scope of the claims of the present invention.

Claims

1. A noise suppression method for acoustic and magnetic positioning of underground cable faults, characterized in that: The specific steps are as follows: Step 1, reading the acoustic magnetic signal, the acoustic magnetic signal comes from the acoustic magnetic synchronous signal with noise captured by the acoustic magnetic sensor during underground cable fault detection; Step 2, performing wavelet transform filtering on the obtained noisy sound signal; Step 3, estimating the average amplitude spectrum of the noise in the effective sound segment of the noisy sound signal after wavelet transform filtering; Step 4, using a spectrum subtraction algorithm to obtain the power spectrum of the effective sound signal; Step 5, using the property that the effective sound signal and the noise are uncorrelated, autocorrelation processing is performed on the noisy sound signal to obtain an autocorrelation frame sequence; Step 6, estimate the propagation time and period difference of the acoustic and magnetic signals, narrow the range of the effective sound signal, and generate an enhanced sound signal output through inverse fast Fourier transform.

2. A noise suppression method for underground cable fault acoustic and magnetic positioning according to claim 1, characterized in that: In step 1, the source of the acoustic-magnetic signal is to connect the impact generator to the terminal head of the faulty cable at one end of the faulty cable. The impact generator sends an impact signal to one end of the faulty cable and generates a flashover impact discharge at the fault point. The acoustic-magnetic signal generated by the discharge will diffuse and propagate in the covering soil above. Then the acoustic-magnetic sensor probe is inserted into the ground. The acoustic-magnetic sensor captures the magnetic signal and acoustic signal and other environmental noise generated by the discharge at the fault point.

3. A noise suppression method for acoustic and magnetic location of underground cable faults according to claim 1, characterized in that: In step 2, the wavelet transform filtering method is to use wavelet transform to decompose the noisy sound signal into wavelet coefficients at different scales; select an appropriate threshold to identify the noise; The noise coefficients are set to zero to remove the noise. Using the processed wavelet coefficients, the signal is reconstructed by inverse wavelet transform to generate a preliminary denoised signal.

4. The noise suppression method for underground cable fault acoustic and magnetic location according to claim 1, characterized in that: In step 4, the spectrum subtraction algorithm for obtaining the power spectrum of the effective sound signal is as follows: Among them, |X i (w) 2 Indicates the effective sound signal x i (n) power spectrum; i represents the frame number; |Y i (w) 2 Represents the noisy signal y i The power spectrum of (n); |D i (w) 2 represents the noise signal d i (n) power spectrum.

5. A noise suppression method for acoustic and magnetic location of underground cable faults according to claim 4, characterized in that: The power spectrum D of the noise with effective sound segments is replaced by the estimation of the average amplitude spectrum of the non-effective sound segments. i (w) 2 .

6. A noise suppression method for underground cable fault acoustic and magnetic location according to claim 1, characterized in that: In step 5, using the property that the effective sound signal and the noise are uncorrelated, the autocorrelation function formula of the noisy sound signal is as follows: Wherein, N is the length of each frame after the signal is framed, and k is the delay amount.

7. The noise suppression method for acoustic and magnetic location of underground cable faults according to claim 1, characterized in that: In step 6, the propagation time difference t1 of the acoustic and magnetic signal generated by the discharge is determined by the detection range, and the discharge sound signal lasts for t2, so the effective range of the discharge sound signal can be narrowed to within t1+t2 of the electromagnetic signal.

8. The noise suppression method for acoustic and magnetic location of underground cable faults according to claim 1, characterized in that: In step 6, the enhanced sound signal generated by the inverse fast Fourier transform is determined by the power spectrum obtained by the spectrum subtraction algorithm, the phase spectrum of the noisy sound signal, the autocorrelation frame sequence, and the effective range.

9. A noise suppression method for acoustic and magnetic location of underground cable faults according to claim 6, characterized in that: The noise energy is mainly concentrated near the zero-order autocorrelation coefficient of the noisy sound signal frame. The maximum value of the autocorrelation function or the ratio of the main peak to the secondary peak between the noisy sound signal and the estimated noise is used to determine whether the current processing signal frame is a valid sound signal frame.

10. A noise suppression system for acoustic and magnetic location of underground cable faults, characterized in that: A method for suppressing noise in acoustic and magnetic positioning of underground cable faults according to any one of claims 1 to 9, comprising an electronic device and a computer-readable storage medium; the electronic device comprises a processor, a memory and a communication bus, the memory stores a computer-readable program executable by the processor; the communication bus realizes connection and communication between the processor and the memory; The processor is used to execute the computer-readable program; the computer-storable medium stores one or more programs, and the one or more programs can be executed by one or more processors.

Images