Electric meter missort detection method, system and device

Abstract

The invention provides an electric meter missort detection method, system and device, and the method comprises the steps: collecting a frequency spectrum signal of a target electric meter box, and carrying out the preprocessing of the collected frequency spectrum signal; extracting missort signal features of the target electric meter box; according to the time domain coefficient of the CWD, carrying out compression rearrangement on the frequency spectrum variable of the frequency spectrum signal of the target electric meter box along the frequency axis direction according to the mapping relation between the instantaneous frequency and the instantaneous amplitude of the signal, and calculating a corresponding electric meter box missort score according to the average negative logarithm of the logarithmic regression predicted value of the time frequency coefficient of each frame of the frequency spectrum signal of the target electric meter box; and comparing the electric meter box missort score with a preset threshold value, judging whether the current target electric meter box frequency spectrum signal is abnormal or not, and when the current target electric meter box frequency spectrum signal is abnormal, judging that an electric meter missort behavior exists. According to the invention, high-precision intelligent detection of missort of the multi-meter-position electric meter box is realized, powerful theoretical support is provided for detection work, and the detection precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system automation, and in particular to a method, system and device for detecting a series of electric meter failures. Background Art

[0002] The electric meter is an instrument for measuring the user's electricity consumption. In order to facilitate management, most power companies currently use multi-meter meter boxes for electricity metering in user-concentrated areas. Each meter box usually contains 3 or more meters, and can reach up to 15 meters. The electric meters in the multi-meter meter box are mostly connected in parallel. In most cases, the wires are laid from the switch behind the meter box to the corresponding meter. Due to the cross-distribution of the lines in the meter box, it is easy for a section of the line to go in the wrong direction, resulting in a phenomenon of inter-household communication in the multi-meter meter box. Once the phenomenon of inter-household communication occurs, it will not only cause a short circuit fault in the multi-meter meter box, but also cause meter measurement errors. Therefore, it is very necessary to detect the inter-household communication of the multi-meter meter box. The initial detection method was manual detection, in which specialized technicians went to the site to inspect the multi-meter meter box. This method not only has a relatively low detection efficiency, but also has poor real-time performance, and cannot achieve real-time detection of inter-household communication of the multi-meter meter box. With the innovation and development of modern science and technology, intelligent detection methods have gradually replaced manual detection methods, which has reduced the detection workload to a certain extent and improved the detection efficiency. However, the current detection methods have serious problems of false detection and missed detection in actual applications and can no longer meet actual needs. Summary of the invention

[0003] The purpose of the present invention is to provide a method, system and device for detecting the inter-household connection of an electric meter, so as to solve the technical problem of how to realize the high-precision detection of the inter-household connection of an electric meter box.

[0004] On the one hand, a method for detecting a cross-connection of an electric meter is provided, comprising:

[0005] Collecting spectrum signals of the target electric meter box and preprocessing the collected spectrum signals;

[0006] Perform CWD time-frequency transformation on the preprocessed spectrum signal to extract the inter-household signal characteristics of the target meter box;

[0007] According to the mapping relationship between the instantaneous frequency and the instantaneous amplitude of the signal along the frequency axis according to the time domain coefficient of CWD, the spectrum variables of the spectrum signal of the target meter box are compressed and rearranged, and the corresponding meter box inter-household score is calculated according to the average negative logarithm of the logistic regression prediction value of the time-frequency coefficient of each frame of the spectrum signal of the target meter box;

[0008] The meter box inter-user score is compared with a preset threshold value, and whether the current target meter box spectrum signal is abnormal is determined according to the comparison result. When the current target meter box spectrum signal is abnormal, it is determined that there is meter inter-user behavior.

[0009] Preferably, the preprocessing of the collected spectrum signal includes:

[0010] The spectrum signal of the target meter box collected is denoised using the following formula:

[0011] x=S(n)+B(n)

[0012] Wherein, S(n) represents the voltage spectrum signal of the multi-meter electric meter box; B(n) represents the noise signal; x represents the voltage spectrum signal of the multi-meter electric meter box;

[0013] The voltage spectrum signal of the multi-meter meter box is transformed into the frequency domain according to the following formula, and the statistical characteristics of the signal are perceived in the signal frequency domain:

[0014] X(x)=|X(x)|e i,za

[0015] Among them, X(x) represents the polar coordinates of the voltage spectrum signal x of the multi-meter meter box obtained by short-time Fourier transform; e represents the voltage phase of the multi-meter meter box; i represents the imaginary part; z represents the signal discrete frequency index; a represents the frame index.

[0016] Preferably, the method further comprises enhancing the noise signal according to the following formula:

[0017]

[0018] Among them, x * It represents the voltage spectrum signal of the enhanced multi-meter electric meter box; Represents the estimated magnitude spectrum in polar coordinates of the voltage spectrum signal of a multi-station electricity meter box.

[0019] Preferably, the extracting of the inter-household signal characteristics of the target electric meter box includes:

[0020] The processed voltage signal is transformed into CWD time-frequency signal according to the following formula:

[0021]

[0022] in, represents the voltage signal of the multi-meter meter box after time domain transformation; t represents time delay; b represents angular frequency variable; ε represents attenuation factor; c ... u represents the time variable; u represents the time domain; s represents the signal-to-noise ratio of the voltage signal of the multi-meter meter box;

[0023] The target meter box's cross-household echo signal is obtained as follows:

[0024]

[0025] Among them, x(t) represents the echo signal of the voltage spectrum of the multi-meter meter box; a(t) represents the instantaneous amplitude of the voltage signal; v represents the instantaneous phase of the voltage signal; λ represents the wavelength of the voltage spectrum signal of the multi-meter meter box; k represents the instantaneous speed of the voltage signal of the multi-meter meter box.

[0026] Preferably, compressing and rearranging the spectrum variables of the spectrum signal of the target electric meter box includes:

[0027] The spectrum variables of the spectrum signal of the multi-meter meter box are compressed and rearranged by the following formula:

[0028]

[0029] Wherein, T represents the time-frequency coefficient of the spectrum signal of the multi-meter electric meter box with synchronous compression transformation; h represents the Dirac function; κ(t) represents the instantaneous frequency estimation value of the voltage signal of the multi-meter electric meter box.

[0030] Preferably, the corresponding meter box inter-household score is calculated according to the following formula:

[0031]

[0032] Wherein, A(x) represents the household-to-household score of the multi-meter-position electric meter box; C represents the number of frames of the spectrum signal of the multi-meter-position electric meter box.

[0033] Preferably, comparing the meter box inter-household score with a preset threshold value comprises:

[0034]

[0035] Among them, δ(t) represents the comparison result of multi-meter meter box inter-household; 0 means that the multi-meter meter box is not inter-household; Indicates the threshold for identifying inter-tenant users in the meter box; 1 indicates that the multi-meter meter box is inter-tenant user.

[0036] Preferably, it also includes:

[0037] If the comparison result shows that the multi-meter meter box is a serial household, it is determined that the spectrum signal of the current target meter box is abnormal;

[0038] If the comparison result shows that the multi-meter meter box is not connected to other households, it is determined that there is no abnormality in the spectrum signal of the current target meter box.

[0039] On the other hand, a detection system for electric meter inter-user connection is also provided, which is used to implement the detection method for electric meter inter-user connection, comprising:

[0040] A signal acquisition module is used to acquire the spectrum signal of the target electric meter box and pre-process the acquired spectrum signal;

[0041] The feature extraction module is used to perform CWD time-frequency transformation on the preprocessed spectrum signal to extract the inter-household signal characteristics of the target meter box;

[0042] The scoring module is used to compress and rearrange the spectrum variables of the spectrum signal of the target meter box according to the mapping relationship between the instantaneous frequency and the instantaneous amplitude of the signal along the frequency axis according to the time domain coefficient of CWD, and calculate the corresponding meter box inter-household score according to the average negative logarithm of the logistic regression prediction value of the time-frequency coefficient of each frame of the spectrum signal of the target meter box;

[0043] The determination module is used to compare the meter box inter-user score with a preset threshold value, and determine whether the current target meter box spectrum signal is abnormal according to the comparison result. When the current target meter box spectrum signal is abnormal, it is determined that there is meter inter-user behavior.

[0044] On the other hand, a device for detecting inter-household electric meter connection is also provided, and the target electric meter box is detected by the method described.

[0045] In summary, the implementation of the embodiments of the present invention has the following beneficial effects:

[0046] The method, system and device for detecting inter-household connection of electric meters provided by the present invention automatically detect inter-household connection of electric meter boxes by setting technical parameters such as transmission frequency, period and range of wireless pulse signals of voltage sensors based on voltage spectrum characteristics of multi-meter-position electric meter boxes according to actual conditions; high-precision intelligent detection of inter-household connection of multi-meter-position electric meter boxes is achieved, which provides strong theoretical support for the detection work and improves the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying creative labor, other drawings obtained based on these drawings still belong to the scope of the present invention.

[0048] Figure 1 The present invention is a schematic diagram of the main process of a method for detecting a series of user connections of an electric meter according to an embodiment of the present invention.

[0049] Figure 2 This is a spectrum signal diagram of an electric meter box in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0051] like Figure 1 FIG. 1 is a schematic diagram of an embodiment of a method for detecting a cross-connection of an electric meter provided by the present invention. In this embodiment, the method comprises the following steps:

[0052] Step S1, collect the spectrum signal of the target meter box and pre-process the collected spectrum signal; understandably, use the audio enhancement method to denoise the collected spectrum signal of the meter box; use the short-time Fourier transform technology to transform the voltage spectrum signal of the multi-meter meter box into the frequency domain, and then enhance the noise signal.

[0053] In one embodiment, the preprocessing of the collected spectrum signal includes denoising the collected spectrum signal of the target electric meter box by the following formula:

[0054] x=S(n)+B(n)

[0055] Wherein, S(n) represents the voltage spectrum signal of the multi-meter electric meter box; B(n) represents the noise signal; x represents the voltage spectrum signal of the multi-meter electric meter box;

[0056] The short-time Fourier transform technology is used to transform the voltage spectrum signal x of the multi-meter meter box into the frequency domain, and the statistical characteristics of the signal are perceived in the signal frequency domain. The voltage spectrum signal of the multi-meter meter box is transformed into the frequency domain according to the following formula, and the statistical characteristics of the signal are perceived in the signal frequency domain:

[0057] X(x)=|X(x)|e i,za

[0058] Among them, X(x) represents the polar coordinates of the voltage spectrum signal x of the multi-meter meter box obtained by short-time Fourier transform; e represents the voltage phase of the multi-meter meter box; i represents the imaginary part; z represents the signal discrete frequency index; a represents the frame index.

[0059] The phase information of the pure spectrum signal can be replaced by the phase information of the noisy signal, so the noise signal is enhanced as follows:

[0060]

[0061] Among them, x * It represents the voltage spectrum signal of the enhanced multi-meter electric meter box; Represents the estimated magnitude spectrum in polar coordinates of the voltage spectrum signal of a multi-station electricity meter box.

[0062] Normally, when a multi-meter electric meter box has a cross-household phenomenon, the voltage signal of the electric meter box circuit will mutate and become unstable, the signal amplitude will suddenly increase, the frequency will increase, and the signal spectrum characteristics will be significantly different from the normal state. Therefore, starting from the voltage spectrum characteristics of the multi-meter electric meter box, the cross-household detection of the electric meter box is carried out automatically. According to the demand, wireless sensing technology is used to capture the voltage spectrum signal of the multi-meter electric meter box. A model IHFA-A4F8 voltage sensor is connected to the power bus of the multi-meter electric meter box. According to the actual situation, the technical parameters such as the transmission frequency, period, and range of the wireless pulse signal of the voltage sensor are set. The voltage sensor consists of a signal transmitter and a signal receiver. The signal transmitter transmits a wireless pulse signal to the target. The wireless pulse signal propagates forward along the line. When it contacts the target, a reflection signal will be generated. The reflection signal is finally received by the signal receiver. The signal is the voltage spectrum signal of the multi-meter electric meter box.

[0063] Considering that the voltage sensor is easily interfered by external factors during the signal capture process and there are noise signals in the original signal, in order to ensure the accuracy of subsequent meter box inter-household detection, the audio enhancement method is used to denoise the collected meter box spectrum signal, and the x value is automatically calculated through the formula.

[0064] Step S2, perform CWD time-frequency transformation on the preprocessed spectrum signal to extract the inter-household signal characteristics of the target meter box; it can be understood that by analyzing the voltage spectrum signal of the multi-meter meter box, the inter-household signal characteristics of the multi-meter meter box are extracted, the processed voltage signal is subjected to CWD time-frequency transformation, and the echo signal is calculated. CWD, i.e., choi-williams distribution, is linear and has short-time Fourier transform. The short-time Fourier transform technology is used to transform the multi-meter meter box voltage spectrum signal x into the frequency domain, and the statistical characteristics of the signal are sensed in the signal frequency domain. The phase information of the pure spectrum signal can be replaced by the phase information of the noisy signal, and the enhanced multi-meter meter box voltage spectrum signal x is obtained by the formula. * . The voltage spectrum signal is transformed back to the time domain through inverse transformation, and then the denoised spectrum signal is formed by replacing the phase of the noise signal, so as to complete the acquisition and processing of the spectrum signal of the meter box. By analyzing the voltage spectrum signal of the multi-meter meter box, the characteristics of the multi-meter meter box inter-household signal are extracted, and the processed voltage signal is first subjected to CWD time-frequency transformation. For the multi-meter meter box inter-household signal, the transmission distance of the signal within the coherent accumulation time is less than or equal to a distance resolution unit, and the transmission period is set to 0.15s, so as to calculate the multi-meter meter box inter-household echo signal x(t).

[0065] In one embodiment, the extracting of the inter-household signal characteristics of the target electric meter box includes:

[0066] The processed voltage signal is transformed into CWD time-frequency signal according to the following formula:

[0067]

[0068] in, represents the voltage signal of the multi-meter meter box after time domain transformation; t represents time delay; b represents angular frequency variable; ε represents attenuation factor; c ... u represents the time variable; u represents the time domain; s represents the signal-to-noise ratio of the voltage signal of the multi-meter meter box;

[0069] The target meter box's cross-household echo signal is obtained as follows:

[0070]

[0071] Among them, x(t) represents the echo signal of the voltage spectrum of the multi-meter meter box; a(t) represents the instantaneous amplitude of the voltage signal; v represents the instantaneous phase of the voltage signal; λ represents the wavelength of the voltage spectrum signal of the multi-meter meter box; k represents the instantaneous speed of the voltage signal of the multi-meter meter box.

[0072] Step S3, according to the mapping relationship between the instantaneous frequency and the instantaneous amplitude of the signal along the frequency axis direction according to the time domain coefficient of CWD, the spectrum variables of the spectrum signal of the target meter box are compressed and rearranged, and the corresponding meter box inter-household score is calculated according to the average negative logarithm of the logistic regression prediction value of the time-frequency coefficient of each frame of the spectrum signal of the target meter box. It can be understood that due to the limitation of the inaccuracy principle, the energy concentration in the CWD time-frequency transformation process is poor, and the time-frequency ridges of the spectrum signal of the multi-meter meter box are not obvious enough, which is not conducive to the subsequent identification and detection of the inter-household behavior of the meter box. Therefore, the echo signal is synchronously compressed and transformed, and the CWD time-frequency transformation result is compressed according to the instantaneous frequency estimation value of each point on the echo signal, so as to achieve the effect of enhancing the time domain concentration of the spectrum signal. The partial derivative of the echo signal time is obtained to obtain the instantaneous frequency of the voltage signal of the multi-meter meter box. The time domain coefficient of CWD is compressed and rearranged along the frequency axis according to the mapping relationship between the instantaneous frequency and instantaneous amplitude of the signal, and the time-frequency coefficient T of the multi-meter meter box spectrum signal with synchronous compression transformation is obtained. The time-frequency coefficient of the spectrum signal of the multi-meter meter box is obtained by synchronous compression transformation of the signal. The time-frequency coefficient can reflect the time domain characteristics and frequency domain characteristics of the spectrum signal, which is used as the signal feature of the multi-meter meter box for subsequent intelligent identification and detection of the meter box.

[0073] In one embodiment, compressing and rearranging the spectrum variables of the spectrum signal of the target electric meter box includes:

[0074] The spectrum variables of the spectrum signal of the multi-meter meter box are compressed and rearranged by the following formula:

[0075]

[0076] Wherein, T represents the time-frequency coefficient of the spectrum signal of the multi-meter electric meter box with synchronous compression transformation; h represents the Dirac function; κ(t) represents the instantaneous frequency estimation value of the voltage signal of the multi-meter electric meter box.

[0077] Calculate the corresponding meter box inter-household score according to the following formula:

[0078]

[0079] Wherein, A(x) represents the household-to-household score of the multi-meter-position electric meter box; C represents the number of frames of the spectrum signal of the multi-meter-position electric meter box.

[0080] According to the extracted spectrum characteristics of multi-meter meter box signals, the behavior of inter-household communication of the meter box is identified and detected. The inter-household communication detection of the multi-meter meter box relies on the inter-household communication score A(x), which is calculated by the average negative logarithm of the logistic regression prediction value of the time-frequency coefficient of each frame of the spectrum signal of the multi-meter meter box.

[0081] Step S4, compare the meter box inter-user score with a preset threshold, and determine whether the current target meter box spectrum signal is abnormal according to the comparison result. If the current target meter box spectrum signal is abnormal, it is determined that there is meter inter-user behavior. It is understandable that a threshold is set according to the actual situation, and the multi-meter meter box inter-user score is compared with the threshold.

[0082] In one embodiment, comparing the meter box inter-household score with a preset threshold value includes:

[0083]

[0084] Among them, δ(t) represents the comparison result of multi-meter meter box inter-household; 0 means that the multi-meter meter box is not inter-household; Indicates the threshold for identifying inter-tenant users in the meter box; 1 indicates that the multi-meter meter box is inter-tenant user.

[0085] A threshold is set according to the actual situation, and the score of the multi-meter meter box is compared with the threshold. If it is less than the threshold, it means that the spectrum signal of the current multi-meter meter box is normal and there is no inter-user behavior; if it is greater than the threshold, it means that the spectrum signal of the current multi-meter meter box is abnormal and there is inter-user behavior. When δ(t) = 0, it means that the meter box is not inter-user; δ(t) = 1 means that the multi-meter meter box is inter-user.

[0086] In one embodiment, 40 multi-digit electric meter boxes are selected, the number of electric meter digits in the electric meter box ranges from 3 to 9, and the electric meter model is IYHFA-A4F8A. The design method is used to perform intelligent detection of multi-digit electric meter box series households. The sampling period of the signal collector is set to 0.15s, and the sampling frequency is set to 45.46Hz. A total of 700Byte of electric meter box voltage data information is collected, and its spectrum signal diagram is as follows Figure 2As shown. According to the analysis, the characteristics of inter-household connection were extracted, and the inter-household connection of the meter box was identified and detected. A total of 12 multi-meter meter boxes with inter-household connection were detected, and 28 meter boxes did not have the inter-household connection phenomenon.

[0087] If the comparison result shows that the multi-meter meter box is a series of households, the spectrum signal of the current target meter box is determined to be abnormal; if the comparison result shows that the multi-meter meter box is not a series of households, the spectrum signal of the current target meter box is determined to be normal.

[0088] The embodiment of the present invention further provides a system for detecting the inter-user connection of an electric meter, which is used to implement the method for detecting the inter-user connection of an electric meter, comprising:

[0089] A signal acquisition module is used to acquire the spectrum signal of the target electric meter box and pre-process the acquired spectrum signal;

[0090] The feature extraction module is used to perform CWD time-frequency transformation on the preprocessed spectrum signal to extract the inter-household signal characteristics of the target meter box;

[0091] The scoring module is used to compress and rearrange the spectrum variables of the spectrum signal of the target meter box according to the mapping relationship between the instantaneous frequency and the instantaneous amplitude of the signal along the frequency axis according to the time domain coefficient of CWD, and calculate the corresponding meter box inter-household score according to the average negative logarithm of the logistic regression prediction value of the time-frequency coefficient of each frame of the spectrum signal of the target meter box;

[0092] The determination module is used to compare the meter box inter-user score with a preset threshold value, and determine whether the current target meter box spectrum signal is abnormal according to the comparison result. When the current target meter box spectrum signal is abnormal, it is determined that there is meter inter-user behavior.

[0093] The embodiment of the present invention further provides a device for detecting inter-household connections of an electric meter, and detects a target electric meter box by using the method for detecting inter-household connections of an electric meter.

[0094] It should be noted that the system described in the above embodiment corresponds to the method described in the above embodiment. Therefore, the undetailed parts of the system described in the above embodiment can be obtained by referring to the contents of the method described in the above embodiment, and will not be repeated here.

[0095] In summary, the implementation of the embodiments of the present invention has the following beneficial effects:

[0096] The method, system and device for detecting inter-household connection of electric meters provided by the present invention automatically detect inter-household connection of electric meter boxes by setting technical parameters such as transmission frequency, period and range of wireless pulse signals of voltage sensors based on voltage spectrum characteristics of multi-meter-position electric meter boxes according to actual conditions; high-precision intelligent detection of inter-household connection of multi-meter-position electric meter boxes is achieved, which provides strong theoretical support for the detection work and improves the detection accuracy.

[0097] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for detecting inter-user connection of an electric meter, characterized in that: include: Collecting spectrum signals of the target electric meter box and preprocessing the collected spectrum signals; Perform CWD time-frequency transformation on the preprocessed spectrum signal to extract the inter-household signal characteristics of the target meter box; According to the mapping relationship between the instantaneous frequency and the instantaneous amplitude of the signal along the frequency axis according to the time domain coefficient of CWD, the spectrum variables of the spectrum signal of the target meter box are compressed and rearranged, and the corresponding meter box inter-household score is calculated according to the average negative logarithm of the logistic regression prediction value of the time-frequency coefficient of each frame of the spectrum signal of the target meter box; The meter box inter-user score is compared with a preset threshold value, and whether the current target meter box spectrum signal is abnormal is determined according to the comparison result. When the current target meter box spectrum signal is abnormal, it is determined that there is meter inter-user behavior.

2. The method according to claim 1, characterized in that The preprocessing of the collected spectrum signal includes: The spectrum signal of the target meter box collected is denoised using the following formula: x=S(n)+B(n) Wherein, S(n) represents the voltage spectrum signal of the multi-meter electric meter box; B(n) represents the noise signal; x represents the voltage spectrum signal of the multi-meter electric meter box; The voltage spectrum signal of the multi-meter meter box is transformed into the frequency domain according to the following formula, and the statistical characteristics of the signal are perceived in the signal frequency domain: X(x)=|X(x)|e i,za Among them, X(x) represents the polar coordinates of the voltage spectrum signal x of the multi-meter meter box obtained by short-time Fourier transform; e represents the voltage phase of the multi-meter meter box; i represents the imaginary part; z represents the signal discrete frequency index; a represents the frame index.

3. The method according to claim 2, characterized in that It also includes enhancing the noise signal according to the following formula: Among them, x * It represents the voltage spectrum signal of the enhanced multi-meter electric meter box; Represents the estimated magnitude spectrum in polar coordinates of the voltage spectrum signal of a multi-station electricity meter box.

4. The method according to claim 3, characterized in that The extracting of the inter-household signal characteristics of the target electric meter box includes: The processed voltage signal is transformed into CWD time-frequency signal according to the following formula: in, represents the voltage signal of the multi-meter meter box after time domain transformation; t represents time delay; b represents angular frequency variable; ε represents attenuation factor; c ... u represents the time variable; u represents the time domain; s represents the signal-to-noise ratio of the voltage signal of the multi-meter meter box; The target meter box's cross-household echo signal is obtained as follows: Among them, x(t) represents the echo signal of the voltage spectrum of the multi-meter meter box; a(t) represents the instantaneous amplitude of the voltage signal; v represents the instantaneous phase of the voltage signal; λ represents the wavelength of the voltage spectrum signal of the multi-meter meter box; k represents the instantaneous speed of the voltage signal of the multi-meter meter box.

5. The method according to claim 4, characterized in that The compressing and rearranging the spectrum variables of the spectrum signal of the target electric meter box includes: The spectrum variables of the spectrum signal of the multi-meter meter box are compressed and rearranged by the following formula: Wherein, T represents the time-frequency coefficient of the spectrum signal of the multi-meter electric meter box with synchronous compression transformation; h represents the Dirac function; κ(t) represents the instantaneous frequency estimation value of the voltage signal of the multi-meter electric meter box.

6. The method according to claim 5, characterized in that Calculate the corresponding meter box inter-household score according to the following formula: Wherein, A(x) represents the household-to-household score of the multi-meter-position electric meter box; C represents the number of frames of the spectrum signal of the multi-meter-position electric meter box.

7. The method according to claim 6, characterized in that The comparing the meter box inter-household score with a preset threshold value includes: Among them, δ(t) represents the comparison result of multi-meter meter box inter-household; 0 means that the multi-meter meter box is not inter-household; Indicates the threshold for identifying inter-tenant users in the meter box; 1 indicates that the multi-meter meter box is an inter-tenant user.

8. The method according to claim 7, characterized in that Also includes, If the comparison result shows that the multi-meter meter box is a serial household, it is determined that the spectrum signal of the current target meter box is abnormal; If the comparison result shows that the multi-meter meter box is not connected to other households, it is determined that there is no abnormality in the spectrum signal of the current target meter box.

9. A system for detecting inter-user electricity meter connection, used to implement the method according to any one of claims 1 to 8, characterized in that: include, A signal acquisition module is used to acquire the spectrum signal of the target electric meter box and pre-process the acquired spectrum signal; The feature extraction module is used to perform CWD time-frequency transformation on the preprocessed spectrum signal to extract the inter-household signal characteristics of the target meter box; The scoring module is used to compress and rearrange the spectrum variables of the spectrum signal of the target meter box according to the mapping relationship between the instantaneous frequency and the instantaneous amplitude of the signal along the frequency axis according to the time domain coefficient of CWD, and calculate the corresponding meter box inter-household score according to the average negative logarithm of the logistic regression prediction value of the time-frequency coefficient of each frame of the spectrum signal of the target meter box; The determination module is used to compare the meter box inter-user score with a preset threshold value, and determine whether the current target meter box spectrum signal is abnormal according to the comparison result. When the current target meter box spectrum signal is abnormal, it is determined that there is meter inter-user behavior.

10. A device for detecting a series of electric meter connections, characterized in that: The target electric meter box is detected by the method described in any one of claims 1 to 8.

Images