Clock calibration method based on synchronous data

Abstract

The invention discloses a clock calibration method based on synchronous data, and the method is characterized in that the method comprises the following steps: 1, a plurality of proofreading groups are configured in advance, and each proofreading group comprises more than two pieces of power equipment; each proofreading group selects one comparison device; 2, detecting harmonic signals in a power grid in real time by each power device, and sending time information of feature points in the harmonic signals to a comparison device when the harmonic signals exceeding a threshold value are detected; 3, the comparison device receives the time information sent by the power devices in the same proofreading group, determines the time points of the feature points according to the distribution condition of comparison time, and sends the time points to the power devices, and the power devices calibrate clock signals of the power devices according to the time points; and all power equipment can be automatically calibrated by using the harmonic waves randomly appearing on the power grid, so that the calibration cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of clock calibration, and in particular to a clock calibration method based on synchronization data. Background Art

[0002] In the power system, there are many detection devices and sensors, such as concentrators and smart meters. These devices need to accurately detect various power information, so the internal clock information needs to be accurate. However, this type of power equipment has a low cost, and the internal clock unit will be offset due to time. Currently, the clock of each smart meter is manually calibrated, which is difficult and inefficient. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a clock calibration method based on synchronization data to solve the technical problems of the prior art in performing clock calibration on power equipment such as smart meters, such as high calibration difficulty and low calibration efficiency.

[0004] Technical solution of the present invention:

[0005] A clock calibration method based on synchronization data comprises the following steps:

[0006] Step 1: Pre-configure several proofreading groups, each of which includes more than two power devices; each proofreading group selects a comparison device;

[0007] Step 2: Each power device detects the harmonic signal in the power grid in real time. When a harmonic signal exceeding a threshold is detected, the time information of the characteristic point in the harmonic signal is sent to the comparison device;

[0008] Step 3: The comparison device receives the time information sent by each power device in the same calibration group, determines the time point of the feature point according to the distribution of the comparison time, and sends the time point to each power device. Each power device calibrates its own clock signal according to the time point.

[0009] In the technical solution provided by the present application, in order to solve the problem that each power device cannot calibrate the clock signal, the harmonic signal that appears irregularly on the power grid is used as a marking signal. All power devices in the calibration group will receive the harmonic signal at the same time, so each power device sends the time point when it receives the harmonic signal, and then summarizes it in the comparison device. It can be foreseen that if the time point sent by a certain power device is different from the other time points, the clock signal of the power device must be offset. In this way, in the technical solution provided by the present application, all power devices can be automatically calibrated by using the harmonics that appear randomly on the power grid, reducing the calibration cost.

[0010] Furthermore, step 1 includes the following steps:

[0011] Step 11: Determine the selection range of the calibration group according to the harmonic signal loss range in the power grid;

[0012] Step 12: Divide the entire power grid into several selection intervals according to the selection range;

[0013] Step 13: If there are more than 5 power devices in the selected interval, all the power devices in the selected interval are used as a calibration group; otherwise, the power devices in the selected interval are matched with surrounding power devices until there are more than 5 to form a calibration group;

[0014] Step 14: Perform communication test on the power devices between the calibration groups, and use the power device with the best communication performance with the other power devices in the calibration group as the comparison device.

[0015] In the technical solution provided in the present application, according to the diffusion law of harmonic signals, power equipment within the range of relatively small harmonic signal loss is selected as the proofreading group as much as possible. Therefore, when harmonic signals appear within this range, each power equipment can receive a relatively stable harmonic signal, thereby facilitating the clock correction of the power equipment of the proofreading component.

[0016] In order to avoid the disorder of the power grid, harmonic control is needed, so the occurrence time of high-frequency harmonics in the power grid is unpredictable. At the same time, there are many normal harmonics in the power grid. Therefore, if normal harmonics or signals slightly higher than normal harmonics are sent to the comparison device as marking information, the rest of the power equipment may not be able to collect these harmonic information, so clock calibration will not be possible. To this end, this application provides the following technical solutions:

[0017] Furthermore, step 2 includes the following steps:

[0018] Step 21: Each power device collects harmonic signals in the power grid in real time. When the frequency and intensity of the harmonic signals exceed the preset threshold, the power signals within the period are collected;

[0019] Step 22: extracting characteristic points in the power signal and time points corresponding to the characteristic points;

[0020] Step 23: Send the feature points and time points as marking information to the comparison device.

[0021] In the technical solution provided by this application, the collected harmonic signals are filtered according to the frequency and intensity of the harmonic signals, so the characteristic points and time points are key points with high recognition characteristics. Within the scope of the proofreading group, all power equipment can receive relatively high information. In addition, this solution does not send the received signal directly to the comparison device, so the communication bandwidth can be reduced, and each power device can share the computing task, reducing the performance requirements of the comparison device.

[0022] Although each power device can receive a relatively obvious harmonic signal, the relatively small harmonic signals are different. Therefore, the distortion of the harmonics in the original signal received by each power device is not consistent. If the peak value of the received signal is directly used as the feature point, the signal will present different distortion characteristics due to the different low-frequency harmonic signals superimposed by the high-frequency harmonic signal. For this purpose, the present application provides the following technical solutions:

[0023] Furthermore, in step 22, the power signal is converted into a spectrum by Fourier transform, and characteristic points and time points corresponding to the characteristic points are extracted based on the spectrum.

[0024] In the technical solution provided in the present application, the signal is decomposed by Fourier transform, so that the signal can be converted from the time domain to the frequency domain. The feature points found in this way will pay more attention to the frequency domain information, and the frequency domain information is the main feature of high-order harmonics. This can avoid the influence of low-order harmonics on high-order harmonics and ensure that the feature points found will not be affected by low-order harmonics.

[0025] Fourier transform can only transform the signal from time domain to frequency domain, which will lead to the loss of time information. After finding the feature point, it is impossible to determine the time of the feature point. To this end, the present application provides the following technical solutions:

[0026] Further, step 22 includes the following steps:

[0027] Step 221: Divide the power signal x(t) into multiple signal frames x with equal spacing n (t), n represents the index of the signal frame, and x represents the amplitude of the power signal;

[0028] Step 222: Apply window function w(t) to each signal frame to perform windowing to obtain a processed frame x n (t)`;

[0029] x n (t)`=x n (t)·w(t);

[0030] Step 223: For each x n(t)` Perform Fourier transform to obtain the spectrum X of the processing frame n (w);

[0031] Step 224: Calculate the amplitude spectrum A based on the frequency spectrum n (w);

[0032] A n (w) = IX n (w) Ⅰ;

[0033] Step 225: Convert spectrum X n (w) arranging in time order to generate a time-frequency spectrum diagram;

[0034] S(n, k) = A n [k], where n is the frame rate index, k is the frequency index, and A n [k] = IX n [k]Ⅰ,ⅠX n [k]Ⅰ is an array of complex numbers, each element of which corresponds to a spectrum at a specific frequency;

[0035] Step 226: extracting characteristic points of the high-frequency harmonic signal and corresponding time points from the time-frequency spectrum diagram.

[0036] In the technical solution provided in the present application, after Fourier transform of the power signal, the signal can be decomposed, and signals of different frequencies in the signal can be extracted, and then a time-frequency spectrum diagram can be formed by spectrum superposition. Therefore, the frequency distribution can be found on the time axis at this time, thereby ensuring the accuracy of feature point extraction.

[0037] Because when high-order harmonic signals appear, power equipment will also be affected by low-order harmonics, so it is often impossible to accurately locate the exact reference position in the high-order harmonics. To address this problem, this application provides the following technical solutions:

[0038] Step 226 includes the following steps:

[0039] Step 2261: pre-set the high frequency range;

[0040] Step 2262: Select the amplitude peak point of each frequency in the high frequency range and use it as a feature point and time point.

[0041] In the solution provided in the present application, when selecting the high frequency range, the peak point of each frequency in the high frequency range is selected, so it is ensured that the selected signal can be extracted from all power equipment, ensuring the accuracy of subsequent matching.

[0042] Generally speaking, the clock units of power equipment usually have clock offsets due to accumulated errors. For example, some clock units are 1 second fast, and some clock units are 1 second slow. However, there are also some clock units with too large measurement errors, which may have an error of 1 second per minute. If such clock units are only calibrated, errors will exist again after a period of time. For this purpose, this application provides the following technical solutions:

[0043] Furthermore, step 3 includes the following steps:

[0044] Step 31: Collect the characteristic points and time points sent by each power device;

[0045] Step 32: Compare the frequency and amplitude of the characteristic point of each power device to determine whether the characteristic point uploaded by the power device is derived from the detection result of the same high-order harmonic signal. If yes, continue with the subsequent steps; if not, terminate;

[0046] Step 33: Compare the time points corresponding to the characteristic points of each power device, select the correct time points and characteristic points according to the distribution pattern of the characteristic points, and send the time points and characteristic points to the remaining power devices for clock correction.

[0047] In the technical solution provided by the present application, since there are multiple characteristic points and the time of each characteristic point is clearly marked, it is not only possible to calibrate the time, but also the error, thereby increasing the accuracy of clock calibration.

[0048] There is no regularity in the appearance of high-order harmonics in the power grid. If two high-order harmonics appear in the power grid in a short period of time, the power equipment will be affected when using this as a clock calibration relationship, resulting in calibration errors. To this end, this application provides the following technical solutions:

[0049] Furthermore, in step 32, when any two comparison devices compare whether the transmitted feature points and time points are from the same high-order harmonic, the following steps are adopted:

[0050] Step 321: For the frequency order of all feature points, if the frequency order of the feature points is the same, then execute step 322, otherwise they do not belong to the same high-order harmonic;

[0051] Step 322: Compare the amplitudes of the feature points with the same frequency sequence. If the amplitude error is within a preset range, execute step 323; otherwise, they do not belong to the same high-order harmonic.

[0052] Step 323: If the error direction of the amplitude matches the diffusion direction of the higher-order harmonics, they belong to the same higher-order harmonics; otherwise, they do not belong to the same higher-order harmonics.

[0053] In the technical solution provided in the present application, whether it is the same high-order harmonic to be monitored is determined based on the frequency order of the features, the amplitude of the feature points, and the attenuation direction of the amplitude. Therefore, it has high accuracy and can avoid being affected when multiple high-order harmonic signal sources appear in a short period of time.

[0054] Furthermore, in step 21, each smart meter uses a bandpass filter to filter the received harmonic signal, and when the amplitude of the received harmonic signal exceeds a preset intensity threshold, the power signal of the period is collected.

[0055] In the technical solution provided in the present application, a bandpass filter is used to filter the received power signal, so that the power signal with special harmonics in the bandpass filter can be accurately detected.

[0056] Furthermore, in step 21, the cutoff frequency of the bandpass filter is adjusted based on the high frequency range in the time-frequency spectrum diagram.

[0057] In the technical solution provided in the present application, by adjusting the cutoff frequency of the bandpass filter, the influence of the frequency of low-order harmonics in the power signal on the high-order harmonics can be continuously reduced, thereby increasing the resolution of the high-frequency part of the time-frequency spectrum and increasing the accuracy of spectrum calibration.

[0058] Beneficial effects of the present invention:

[0059] The present invention pre-configures a number of calibration groups, each of which includes more than two power devices; each calibration group selects a comparison device; step 2: each power device detects the harmonic signal in the power grid in real time, and when a harmonic signal exceeding a threshold is detected, the time information of the characteristic point in the harmonic signal is sent to the comparison device. All power devices can be automatically calibrated using the harmonics that randomly appear on the power grid, thereby reducing the calibration cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the process of the present invention;

[0061] Figure 2 The spectrum diagram. DETAILED DESCRIPTION

[0062] refer to Figures 1-2 Embodiment 1: A clock calibration method based on synchronization data comprises the following steps:

[0063] Step 1: Pre-configure several proofreading groups, each of which includes more than two power devices; each proofreading group selects a comparison device.

[0064] Step 1 includes the following steps:

[0065] Step 11: Determine the selection range of the calibration group according to the harmonic signal loss range in the power grid;

[0066] The loss range is a preset range, which is mainly set based on experience. If the loss of harmonic signals in the equipment in the power grid and the transmission environment of the power grid is small, the loss range is large, otherwise it is relatively small.

[0067] Step 12: Divide the entire power grid into several selection intervals according to the selection range.

[0068] Step 13: If there are more than 5 power devices in the selection interval, all the power devices in the selection interval are used as a calibration group; otherwise, the power devices in the selection interval are matched with surrounding power devices until there are more than 5 to form a calibration group.

[0069] When selecting power equipment, each power equipment must be able to directly receive harmonic signals from the power grid, and cannot obtain harmonic signals by forwarding. For example, smart meter A can receive harmonic signals, and smart meter B can be connected to smart meter A by signal, but smart meter B cannot obtain harmonic signals on the power grid through the forwarding of smart meter A, but needs to obtain harmonic signals through the household line connected to the power grid.

[0070] Step 14: Perform communication test on the power devices between the calibration groups, and use the power device with the best communication performance with the other power devices in the calibration group as the comparison device.

[0071] The communication test of power equipment mainly tests its signal strength and communication capability. If a certain power equipment cannot communicate stably with other power equipment, it is naturally impossible to calibrate the clock signal together. Therefore, when recommending a comparison device, it is necessary to consider the communication capability of the comparison device with all other power equipment. In this solution, the comparison device is just an information concentrator. If the comparison device needs to calibrate the clock with other power equipment, it is also necessary to calibrate the clock of the comparison device as an ordinary power device during the calibration.

[0072] Step 2: Each power device detects the harmonic signal in the power grid in real time. When a harmonic signal exceeding a threshold is detected, the time information of the characteristic point in the harmonic signal is sent to the comparison device.

[0073] Step 2 includes the following steps:

[0074] Step 21: Each power device collects harmonic signals in the power grid in real time. When the frequency and intensity of the harmonic signals exceed the preset threshold, the power signals within the period are collected;

[0075] In this scheme, the main purpose is to collect high-order harmonic signals in the power grid, which generally have a relatively high amplitude. After the high-order harmonic signal appears, it will affect the original low-order harmonic signal, causing the received signal to be distorted. In order to reduce the amount of calculation, in step 21, only the frequency and intensity of the signal are detected. Only after both items exceed the threshold, the power signal during this period will be collected for analysis.

[0076] Step 22: Extract characteristic points in the power signal and the time points corresponding to the characteristic points.

[0077] The power signal collected in step 22 is essentially the information converted by the power equipment when receiving the high-frequency harmonic signal. Because the high-order harmonic signal has a fast transmission rate in the power grid, within a certain range, the actual time when all power equipment receives the high-order harmonic signal is basically the same. At this time, it is only necessary to compare the time when all power equipment receives the high-order harmonic signal to determine whether there is an error in the clock of some power equipment.

[0078] When extracting feature points, it is necessary to accurately find the reference signal of high-order harmonics from the complex power signal. To this end, this application provides the following technical solutions:

[0079] In step 22, the power signal is converted into a spectrum by Fourier transform, and characteristic points and time points corresponding to the characteristic points are extracted based on the spectrum.

[0080] Step 22 includes the following steps:

[0081] Step 221: Divide the power signal x(t) into multiple signal frames x with equal spacing n (t), n represents the index of the signal frame, and x represents the amplitude of the power signal;

[0082] Step 222: Apply window function w(t) to each signal frame to perform windowing to obtain a processed frame x n (t)`;

[0083] x n (t)`=x n (t)·w(t);

[0084] Step 223: For each x n (t)` Perform Fourier transform to obtain the spectrum X of the processing frame n (w);

[0085] Step 224: Calculate the amplitude spectrum A based on the frequency spectrum n (w);

[0086] A n (w) = IX n (w) Ⅰ;

[0087] Step 225: Convert spectrum X n (w) arranging in time order to generate a time-frequency spectrum diagram;

[0088] S(n, k) = A n [k], where n is the frame rate index, k is the frequency index, and A n [k] = IX n [k]Ⅰ,ⅠX n [k]Ⅰ is an array of complex numbers, each element of which corresponds to a spectrum at a specific frequency;

[0089] Step 226: extracting characteristic points of the high-frequency harmonic signal and corresponding time points from the time-frequency spectrum diagram.

[0090] like Figure 2 As shown, in the above scheme, the spectrum diagram of the signal is extracted. Relatively speaking, in the low-frequency signal, because there are a large number of low-order harmonics, the color at the bottom of the spectrum is dark, while the signal at the top is relatively small. Although the information at the top is small, it is all generated by high-order harmonic signal sources, so it can be used to extract the corresponding feature points.

[0091] Step 226 includes the following steps:

[0092] Step 2261: Pre-set the high frequency range.

[0093] The high frequency range is a preset value, which is mainly used to determine which signals are sudden high frequency signals.

[0094] Step 2262: Select the amplitude peak point of each frequency in the high frequency range and use it as a feature point and time point.

[0095] When selecting the amplitude peak point of each frequency, it is mainly used to obtain effective feature points. If the amplitude of the peak point is too small, it cannot be selected to avoid the influence of noise.

[0096] Step 23: Send the feature points and time points as marking information to the comparison device.

[0097] Step 3: The comparison device receives the time information sent by each power device in the same calibration group, determines the time point of the characteristic point according to the distribution of the comparison time, and sends the time point to each comparison device. Each comparison device calibrates its own clock signal according to the time point.

[0098] Step 3 includes the following steps:

[0099] Step 31: Collect the characteristic points and time points sent by each power device;

[0100] Step 32: Compare the frequency and amplitude of the characteristic point of each power device to determine whether the characteristic point uploaded by the power device is derived from the detection result of the same high-order harmonic signal. If yes, continue with the subsequent steps; if not, terminate;

[0101] In step 32, when any two comparison devices compare the transmitted characteristic points and time points to see whether they are from the same high-order harmonic, the following steps are adopted:

[0102] Step 321: For the frequency order of all feature points, if the frequency order of the feature points is the same, then execute step 322, otherwise they do not belong to the same high-order harmonic;

[0103] Step 322: Compare the amplitudes of the feature points with the same frequency sequence. If the amplitude error is within a preset range, execute step 323; otherwise, they do not belong to the same high-order harmonic.

[0104] Step 323: If the error direction of the amplitude matches the diffusion direction of the higher-order harmonics, they belong to the same higher-order harmonics; otherwise, they do not belong to the same higher-order harmonics.

[0105] In step 321 to step 323, three conditions are used to determine whether the power signals received by the two power devices are observations of the same high-order harmonic signal source. In this way, by adopting the above scheme, it is possible to infer that the power signals received by those power devices are observations of the same harmonic signal source, and then these power devices are clock-calibrated together. The reason why the present application adopts the above order for screening is that the amount of comparison can be reduced successively.

[0106] Step 322 and step 323 are actually the contents of synchronous detection. The amplitude error value mainly refers to whether there is an error in the amplitude of the signal of the same frequency. As for the error direction, it is mainly to judge whether the occurrence of the error is reasonable. For example, the signal source is 100 meters away from power equipment A and only 50 meters away from power equipment B, but the amplitude of the signal received by power equipment B is higher than that of power equipment A, which is obviously not in line with common sense. The position of the signal source can be determined by fuzzy matching based on the relationship between the power equipment. For example, A is on the left side of B, and C is on the left side of B, while the amplitude of the power signal received by the three power equipments A, B, and C is not synchronously increased or decreased, but there is not much regularity. Therefore, in practice, it is only necessary to have enough power equipment and determine it based on the amplitude strength and the change of amplitude to roughly determine the approximate direction of the high-order harmonic signal source.

[0107] Step 33: Compare the time points corresponding to the characteristic points of each power device, select the correct time points and characteristic points according to the distribution law of the characteristic points, and send the time points and characteristic points to the remaining power devices for clock correction. In this way, all power devices in the same comparison group can perform clock correction together, and then the clock signals of all power devices in the same comparison group can be synchronized within a reasonable range.

[0108] The clock calibration method will not be elaborated here. Clock calibration is an existing technical means and can be directly corrected when the reference time is known.

[0109] Embodiment 2: Embodiment 2 differs from Embodiment 1 in that a bandpass filter is used in step 21 to filter the received power signal.

[0110] Specifically: Further, in step 21, each smart meter uses a bandpass filter to filter the received harmonic signal, and when the amplitude of the received harmonic signal exceeds a preset intensity threshold, the power signal of the period is collected.

[0111] Furthermore, in step 21, the cutoff frequency of the bandpass filter is adjusted based on the high frequency range in the time-frequency spectrum diagram.

[0112] In step 21, after the power signal is extracted, the signal processing method in step 22 is first used to extract the time-frequency spectrum. After the spectrum is extracted, the selection range of the high-order frequency is determined, and then the filtering range of the bandpass filter is adjusted to filter the previously received power signal. After the filtering is successful, the signal processing method in step 22 and subsequent steps are executed. In this way, the influence of low-order harmonics in the second extracted time-frequency spectrum will be smaller.

[0113] Example 3: Example 3 introduces Beidou timing technology based on Example 1, and uses the accurate atomic clock in the Beidou satellite to perform clock calibration.

[0114] Step 3 also includes step 34:

[0115] When the adjustment value of the clock signal of the power equipment exceeds the preset threshold, the power equipment sends a synchronization timing request to the Beidou satellite, and the Beidou satellite sends timing information to the power equipment.

[0116] In this way, each power device in this scheme does not exchange information with the Beidou satellite according to specific instructions, but sends a request to the Beidou satellite when the difference in clock calibration is too large. Therefore, the frequency of clock calibration can be reduced and the clock calibration bandwidth can be reduced. However, each comparison group can introduce a correct clock information when there is a large difference, thereby increasing the accuracy of clock calibration.

Claims

1. A clock calibration method based on synchronization data, characterized in that: The method comprises: Step 1: Pre-configure several proofreading groups, each of which includes more than two power devices; each proofreading group selects a comparison device; Step 2: Each power device detects the harmonic signal in the power grid in real time. When a harmonic signal exceeding a threshold is detected, the time information of the characteristic point in the harmonic signal is sent to the comparison device; Step 3: The comparison device receives the time information sent by each power device in the same calibration group, determines the time point of the feature point according to the distribution of the comparison time, and sends the time point to each power device. Each power device calibrates its own clock signal according to the time point.

2. A clock calibration method based on synchronization data according to claim 1, characterized in that: The implementation method of step 1 includes: Step 11: Determine the selection range of the calibration group according to the harmonic signal loss range in the power grid; Step 12: Divide the entire power grid into several selection intervals according to the selection range; Step 13: If there are more than 5 power devices in the selected interval, all the power devices in the selected interval are used as a calibration group; otherwise, the power devices in the selected interval are matched with surrounding power devices until there are more than 5 to form a calibration group; Step 14: Perform communication test on the power devices between the calibration groups, and use the power device with the best communication performance with the other power devices in the calibration group as the comparison device.

3. The clock calibration method based on synchronization data according to claim 1, characterized in that: The implementation method of step 2 includes: Step 21: Each power device collects harmonic signals in the power grid in real time. When the frequency and intensity of the harmonic signals exceed the preset threshold, the power signals within the period are collected; Step 22: extracting characteristic points in the power signal and time points corresponding to the characteristic points; Step 23: Send the feature points and time points as marking information to the comparison device.

4. The clock calibration method based on synchronization data according to claim 4, characterized in that: The method for obtaining the characteristic points in the power signal and the time points corresponding to the characteristic points is: converting the power signal into a spectrum using Fourier transform, and extracting the characteristic points and the time points corresponding to the characteristic points based on the spectrum.

5. The clock calibration method based on synchronization data according to claim 4, characterized in that: The implementation method of step 22 includes: Step 221: Divide the power signal x(t) into multiple signal frames x with equal spacing n (t), n represents the index of the signal frame, and x represents the amplitude of the power signal; Step 222: Apply window function w(t) to each signal frame to perform windowing to obtain a processed frame x n (t)`; x n (t)`=x n (t)·w(t); Step 223: For each x n (t)` Perform Fourier transform to obtain the spectrum X of the processing frame n (w); Step 224: Calculate the amplitude spectrum A based on the frequency spectrum n (w); A n (w)=ⅠX n (w)Ⅰ; Step 225: Convert spectrum X n (w) arranging in time order to generate a time-frequency spectrum diagram; S(n, k) = A n [k], where n is the frame rate index, k is the frequency index, and A n [k] = IX n [k]Ⅰ,ⅠX n [k]Ⅰ is an array of complex numbers, each element of which corresponds to a spectrum at a specific frequency; Step 226: extracting characteristic points of the high-frequency harmonic signal and corresponding time points from the time-frequency spectrum diagram.

6. The clock calibration method based on synchronization data according to claim 5, characterized in that: Step 226 includes: Step 2261: pre-set the high frequency range; Step 2262: Select the amplitude peak point of each frequency in the high frequency range and use it as a feature point and time point.

7. The clock calibration method based on synchronization data according to claim 1, characterized in that: Step 3 includes: Step 31: Collect the characteristic points and time points sent by each power device; Step 32: Compare the frequency and amplitude of the characteristic point of each power device to determine whether the characteristic point uploaded by the power device is derived from the detection result of the same high-order harmonic signal. If yes, continue with the subsequent steps; if not, terminate; Step 33: Compare the time points corresponding to the characteristic points of each power device, select the correct time points and characteristic points according to the distribution pattern of the characteristic points, and send the time points and characteristic points to the remaining power devices for clock correction.

8. The clock calibration method based on synchronization data according to claim 1, characterized in that: The implementation method of step 32 includes: Step 321: For the frequency order of all feature points, if the frequency order of the feature points is the same, then execute step 322, otherwise they do not belong to the same high-order harmonic; Step 322: Compare the amplitudes of the feature points with the same frequency sequence. If the amplitude error is within a preset range, execute step 323; otherwise, they do not belong to the same high-order harmonic. Step 323: If the error direction of the amplitude matches the diffusion direction of the higher-order harmonics, they belong to the same higher-order harmonics; otherwise, they do not belong to the same higher-order harmonics.

9. The clock calibration method based on synchronization data according to claim 3, characterized in that: In step 21, each smart meter uses a bandpass filter to filter the received harmonic signal. When the amplitude of the received harmonic signal exceeds the preset intensity threshold, the power signal of the period is collected, and the receiving frequency of the bandpass filter is less than the preset frequency threshold.

10. The clock calibration method based on synchronization data according to claim 7, characterized in that: Step 3 also includes: Step 34: When the adjustment value of the clock signal of the power equipment exceeds the preset threshold, the power equipment sends a synchronization timing request to the Beidou satellite, and the Beidou satellite sends timing information to the power equipment.

Images