A method and system for tracing stray currents in a new energy system

Abstract

This invention discloses a method and system for tracing stray currents in new energy systems, belonging to the technical field of stray current technology in new energy systems. The method for tracing stray currents in new energy systems provided by this invention generates spectral fingerprints by using Fast Fourier Transform (FFT) on high-frequency stray current component A and low-frequency stray current component B, and extracts long-time and short-time fingerprints by wavelet decomposition on very low-frequency stray current component C. This establishes a stray current spectral fingerprint database, a stray current long-time fingerprint database, and a stray current short-time fingerprint database, forming a multi-dimensional feature database. Based on correlation judgment results and combined with external meteorological data, this method achieves rapid tracing of known interference sources and continuously improves the multi-dimensional feature database through an automatic update mechanism. Furthermore, incorporating external meteorological data into the judgment process can distinguish between stray currents caused by natural factors and equipment failures, improving the accuracy of tracing.

Description

Technical Field

[0001] This invention relates to the field of stray current technology in new energy systems, specifically to a method and system for tracing the source of stray current in new energy systems. Background Technology

[0002] While the new energy industry is booming, stray current problems are becoming increasingly prominent, posing a significant challenge to the safe and stable development of new energy systems. As the scale of new energy systems continues to expand, these stray currents pose numerous serious hazards to the power grid. For example, stray currents accelerate the corrosion of the power grid grounding network, shorten its service life, and increase safety risks in grid operation; they can also cause DC bias in transformers, affecting their normal operation, reducing power quality, and in severe cases, even leading to transformer damage and significant economic losses.

[0003] However, stray currents in new energy systems originate from diverse sources, and their parameters vary depending on the source, which is closely related to operating conditions and site conditions. Existing technologies can only detect the presence or absence of stray currents, such as polarization current detection methods originating from the metal corrosion detection industry. When faced with complex power grid environments and new energy systems, they cannot quickly determine the specific source of stray currents, making it difficult to conduct targeted hazard assessments and risk management. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for tracing the source of stray current in a new energy system, so as to overcome the problem in the prior art that the specific source of stray current cannot be quickly determined, resulting in a lack of precise targeting in hazard assessment and control.

[0005] The present invention solves the above-mentioned technical problems through the following technical solution:

[0006] This invention provides a method for tracing the source of stray current in a new energy system, comprising the following steps:

[0007] S1. Obtain the high-frequency stray current component A, low-frequency stray current component B, and very low-frequency stray current component C at the preset sampling points of the new energy system.

[0008] S2. Use Fast Fourier Transform to extract the spectral fingerprints of high-frequency stray current component A and low-frequency stray current component B, and store them in a pre-built stray current spectral fingerprint library; use wavelet decomposition to extract the long-time fingerprint and short-time fingerprint of very low-frequency stray current component C, and store them in a pre-built stray current long-time fingerprint library and stray current short-time fingerprint library, respectively.

[0009] S3. Obtain the high-frequency stray current component a, low-frequency stray current component b, and very low-frequency stray current component c of the current data to be traced. Generate the spectral fingerprint, long-term fingerprint, and short-term fingerprint to be traced. Determine the correlation between the spectral fingerprint and the stray current spectral fingerprint database, the long-term fingerprint and the stray current long-term fingerprint database, and the short-term fingerprint and the stray current short-term fingerprint database, respectively. Based on the determination results and combined with external meteorological data, complete the source tracing or update of the stray current spectral fingerprint database, the stray current long-term fingerprint database, and the stray current short-term fingerprint database.

[0010] A further improvement of this invention lies in determining the correlation between the source-tracing spectral fingerprint and the stray current spectral fingerprint database, and based on the determination result, combined with external meteorological data, completing the source tracing or stray current spectral fingerprint database update, specifically including the following steps:

[0011] S311. Extract the spectral component with the largest amplitude from the spectral fingerprint to be traced;

[0012] S312. Using frequency as an indicator, determine whether a dominant frequency fingerprint exists in the stray current spectral fingerprint database. The dominant frequency fingerprint is a spectral fingerprint whose frequency difference with the spectral component with the largest amplitude is ≤ ±3%. If the determination result is negative, proceed to step S313. If the determination result is positive, extract the second largest amplitude spectral component from the spectral fingerprint to be traced and the dominant frequency fingerprint. Using frequency as an indicator, determine whether the frequency difference between the second largest amplitude spectral component in the spectral fingerprint to be traced and the second largest amplitude spectral component in the dominant frequency fingerprint is ≤ ±5%. If the determination result is positive, the spectral fingerprint to be traced and the dominant frequency fingerprint in the stray current spectral fingerprint database have the same cause of generation, and the tracing is completed. If the determination result is negative, proceed to step S313.

[0013] S313. Obtain external meteorological data for the same time period as the spectrum fingerprint to be traced, plot the waveform relationship between the spectrum component with the largest amplitude in the spectrum fingerprint to be traced and the external meteorological data. If the waveforms are related, the spectrum fingerprint to be traced is caused by the external meteorological data, and the tracing is completed. Update the spectrum fingerprint to be traced to the stray current spectrum fingerprint database.

[0014] Determine the correlation between the long-term fingerprint of the source to be traced and the long-term fingerprint database of stray currents. Based on the determination result and combined with external meteorological data, complete the update of the source tracing or stray current long-term fingerprint database. The specific steps include:

[0015] S321. Based on the time period and region of the long-term fingerprint to be traced, determine whether there is a long-term fingerprint in the stray current long-term fingerprint database that is in the same time period and region. If the determination result is yes, the long-term fingerprint to be traced and the long-term fingerprint in the stray current long-term fingerprint database that is in the same time period and region have the same cause of generation, and the tracing is completed; if the determination result is no, proceed to step S322.

[0016] S322. Obtain the external meteorological data for the same time period as the long-term fingerprint to be traced, and draw the waveform relationship between the long-term fingerprint to be traced and the external meteorological data. If the waveforms are related, the long-term fingerprint to be traced is caused by the external meteorological data, and the tracing is completed. Update the long-term fingerprint to be traced to the stray current long-term fingerprint database.

[0017] Determine the correlation between the short-time fingerprint of the source to be traced and the stray current short-time fingerprint database. Based on the determination result and combined with external meteorological data, complete the source tracing or stray current short-time fingerprint database update. The specific steps include:

[0018] S331. Based on the time period and region of the short-time fingerprint to be traced, determine whether there is a short-time fingerprint in the stray current short-time fingerprint database that is in the same time period and region. If the determination result is yes, the short-time fingerprint to be traced and the short-time fingerprint in the stray current short-time fingerprint database that is in the same time period and region have the same cause of generation, and the tracing is completed; if the determination result is no, proceed to step S332.

[0019] S332. Obtain the external meteorological data for the same time period as the short-time fingerprint to be traced, and draw the waveform relationship diagram between the short-time fingerprint to be traced and the external meteorological data. If the waveforms are related, the short-time fingerprint to be traced is caused by the external meteorological data, and the tracing is completed; and update the short-time fingerprint to be traced to the stray current short-time fingerprint database.

[0020] A further improvement of the present invention is that the criteria for determining whether there are long-term fingerprints in the same time period and the same region in the stray current long-term fingerprint database are as follows: the time period of the long-term fingerprint in the stray current long-term fingerprint database differs from the time period of the long-term fingerprint to be traced by less than one minute; and the location of the long-term fingerprint and the location of the long-term fingerprint to be traced differs from the location of the long-term fingerprint to be traced by less than 50 km.

[0021] The criteria for determining whether short-time fingerprints from the same time period and region exist in the stray current short-time fingerprint database are as follows: the time period of the short-time fingerprint in the stray current short-time fingerprint database differs from the time period of the short-time fingerprint to be traced by less than one second; and the region of the short-time fingerprint differs from the region of the short-time fingerprint to be traced by less than 10 km.

[0022] A further improvement of the present invention is that the external meteorological data includes at least solar radiation intensity and wind speed.

[0023] A further improvement of this invention is that the extraction of long-time and short-time fingerprints from the very low frequency stray current component C using wavelet decomposition specifically includes the following steps:

[0024] S21. The very low frequency stray current component C is decomposed into 8 components by using the db4 wavelet in a 7-segment multi-resolution decomposition.

[0025] S22. Let C1 be the decomposition component with the lowest resolution. Extract the time nodes of all local peaks and local valleys from the waveform of decomposition component C1. Sort the time nodes in ascending order according to their chronological order. Based on the ascending order, construct the first sequence S = [s1, s2, ..., s...]. n 】, where s1, s2, ..., s n These are the 1st, 2nd, ..., nth subsequences of the first sequence; n is the total number of time points for all local peaks and local valleys;

[0026] S23. Based on the time differences between adjacent subsequences of the first sequence S, construct the second sequence dS = [d1, d2, ..., d...]. n-1 】, where d1, d2, ..., d n-1 These are the 1st, 2nd, ..., n-1th subsequences of the second sequence, respectively;

[0027] S24. Set the number of loops to K, and let K=1;

[0028] S25. Determine whether the Kth subsequence of the second sequence is greater than the preset time threshold. If the determination result is yes, then the component of the decomposed component C1 at the corresponding time node is used as a long-term fingerprint and stored in the pre-constructed stray current long-term fingerprint library. If the determination result is no, then the component of the decomposed component C1 at the corresponding time node is used as a short-term fingerprint and stored in the pre-constructed stray current short-term fingerprint library.

[0029] S26. Determine if K is less than n-1. If the result is yes, set K=K+1 and return to S25. If the result is no, execute step S3.

[0030] A further improvement of the present invention is that the preset sampling point is at least one of the cables, inverters and combiner boxes of the new energy system, and the new energy system includes wind power system, photovoltaic system and energy storage system.

[0031] A further improvement of this invention lies in acquiring the high-frequency stray current component, low-frequency stray current component, and very low-frequency stray current component at a preset sampling point, specifically including the following steps:

[0032] S11. Determine whether the installation location of the preset sampling point is suitable for an open Hall sensor. If the determination result is yes, proceed to step S12; otherwise, proceed to step S13.

[0033] S12. Based on the open Hall sensor, the original current data of the preset sampling point is obtained, and the high-frequency stray current component A, low-frequency stray current component B and very low-frequency stray current component C are extracted from the original current data using a filter.

[0034] S13. The high-frequency stray current component A and the low-frequency stray current component B of the preset sampling point are directly extracted by the Rogowski coil sensor, and the very low-frequency stray current component C of the preset sampling point is directly extracted by the fluxgate sensor.

[0035] A further improvement of the present invention is that the sampling rate of the open Hall sensor is 16kHz; the sampling rate of the Rogowski coil sensor is 16kHz; and the sampling rate of the fluxgate sensor is 10Hz.

[0036] A further improvement of the present invention is that step S12 can be replaced by the following steps:

[0037] The original current data of the preset sampling point is obtained based on the open Hall sensor. The high-frequency stray current component, low-frequency stray current component and very low-frequency stray current component in the original current data are extracted by the filter. It is determined whether the very low-frequency stray current component is greater than 0.1 times the root mean square of the sum of the high-frequency stray current component and the low-frequency stray current component. If the determination result is yes, the high-frequency stray current component, low-frequency stray current component and very low-frequency stray current component in the original current data are the high-frequency stray current component A, low-frequency stray current component B and very low-frequency stray current component C of the preset sampling point. If the determination result is no, step S13 is executed.

[0038] This invention also provides a source tracing system for stray currents in a new energy system, comprising:

[0039] The first module is used to acquire the high-frequency stray current component A, the low-frequency stray current component B, and the very low-frequency stray current component C at preset sampling points of the new energy system.

[0040] The second module is used to extract the spectral fingerprints of high-frequency stray current component A and low-frequency stray current component B using fast Fourier transform and store them in a pre-built stray current spectral fingerprint library; and to extract the long-time fingerprint and short-time fingerprint of very low-frequency stray current component C using wavelet decomposition and store them in the pre-built stray current long-time fingerprint library and stray current short-time fingerprint library, respectively.

[0041] The third module is used to acquire the high-frequency stray current component a, the low-frequency stray current component b, and the very low-frequency stray current component c of the current data to be traced, and to generate the spectral fingerprint, long-term fingerprint, and short-term fingerprint to be traced. The module then determines the correlation between the spectral fingerprint and the stray current spectral fingerprint database, the long-term fingerprint and the stray current long-term fingerprint database, and the short-term fingerprint and the stray current short-term fingerprint database, respectively. Based on the determination results and combined with external meteorological data, the module completes the source tracing or updates the stray current spectral fingerprint database, the stray current long-term fingerprint database, and the stray current short-term fingerprint database.

[0042] Compared with the prior art, the positive and progressive effects of the present invention are as follows:

[0043] The stray current tracing method for new energy systems provided by this invention generates spectral fingerprints by using Fast Fourier Transform (FFT) on high-frequency stray current component A and low-frequency stray current component B, and extracts long-time and short-time fingerprints by wavelet decomposition on very low-frequency stray current component C. This establishes a stray current spectral fingerprint database, a stray current long-time fingerprint database, and a stray current short-time fingerprint database, forming a multi-dimensional feature database. Based on correlation judgment results and combined with external meteorological data, this method achieves rapid tracing of known interference sources and continuously improves the multi-dimensional feature database through an automatic update mechanism. Furthermore, incorporating external meteorological data into the judgment process distinguishes between stray currents caused by natural factors and equipment failures, thus improving the accuracy of tracing. Attached Figure Description

[0044] The accompanying drawings are provided to further understand the invention and constitute a part of this invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0045] Figure 1 This is a flowchart illustrating a method for tracing the source of stray current in a new energy system according to the present invention.

[0046] Figure 2 This is a schematic diagram of a stray current tracing system for a new energy system according to the present invention;

[0047] Figure 3 This is a waveform diagram of the decomposed component C1 in one embodiment;

[0048] Figure 4 This is a schematic diagram of a long-time fingerprint in the stray current long-time fingerprint database in a specific embodiment.

[0049] Figure 5 This is a schematic diagram of a short-time fingerprint in the stray current short-time fingerprint database in a specific embodiment;

[0050] Figure 6 This is a waveform diagram showing the relationship between the traceable spectral fingerprint and wind speed in a specific embodiment.

[0051] Figure 7 This is a waveform diagram showing the relationship between the traceable spectral fingerprint and solar radiation intensity in a specific embodiment.

[0052] Figure 8 This is a waveform diagram showing the relationship between long-term fingerprints to be traced and wind speed in a specific embodiment;

[0053] Figure 9 This is a waveform diagram showing the relationship between the long-term fingerprint to be traced and the solar radiation intensity in a specific embodiment;

[0054] Figure 10 This is a waveform diagram showing the relationship between short-time fingerprints to be traced and wind speed in a specific embodiment.

[0055] Figure 11 This is a waveform diagram showing the relationship between the short-time fingerprint to be traced and the solar radiation intensity in a specific embodiment. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. This is an explanation of the present invention and not a limitation thereof.

[0058] See Figure 1 This invention provides a method for tracing the source of stray current in a new energy system, comprising the following steps:

[0059] S1. Obtain the high-frequency stray current component A, low-frequency stray current component B, and very low-frequency stray current component C at the preset sampling points of the new energy system.

[0060] S2. Use Fast Fourier Transform to extract the spectral fingerprints of high-frequency stray current component A and low-frequency stray current component B, and store them in a pre-built stray current spectral fingerprint library; use wavelet decomposition to extract the long-time fingerprint and short-time fingerprint of very low-frequency stray current component C, and store them in a pre-built stray current long-time fingerprint library and stray current short-time fingerprint library, respectively.

[0061] S3. Obtain the high-frequency stray current component a, low-frequency stray current component b, and very low-frequency stray current component c of the current data to be traced. Generate the spectral fingerprint, long-term fingerprint, and short-term fingerprint to be traced. Determine the correlation between the spectral fingerprint and the stray current spectral fingerprint database, the long-term fingerprint and the stray current long-term fingerprint database, and the short-term fingerprint and the stray current short-term fingerprint database, respectively. Based on the determination results and combined with external meteorological data, complete the source tracing or update of the stray current spectral fingerprint database, the stray current long-term fingerprint database, and the stray current short-term fingerprint database.

[0062] This invention constructs an intelligent source tracing system for stray currents in new energy systems through multi-dimensional feature extraction and a dynamic fingerprint database update mechanism. During the data acquisition phase, three types of stray current components—high-frequency stray current component A, low-frequency stray current component B, and very low-frequency stray current component C—are acquired through preset sampling points, providing a complete data foundation for feature analysis in different frequency bands. For high-frequency stray current component A and low-frequency stray current component B, a fast Fourier transform is used to generate spectral fingerprints, effectively capturing periodic interference characteristics. For very low-frequency stray current component C, wavelet decomposition is performed to extract long-time and short-time fingerprints, overcoming the shortcomings of traditional methods in analyzing non-stationary signals. By establishing stray current spectral fingerprint database, long-time stray current fingerprint database, and short-time stray current fingerprint database, a multi-dimensional feature database is formed. Based on correlation judgment results and combined with external meteorological data, rapid source tracing of known interference sources is achieved, and the multi-dimensional feature database is continuously improved through an automatic update mechanism. Simultaneously, integrating external meteorological data into the judgment process can distinguish between stray current differences caused by natural factors and equipment failures, improving the accuracy of source tracing.

[0063] The preferred method involves determining the correlation between the source-tracing spectral fingerprint and the stray current spectral fingerprint database. Based on the determination result and combined with external meteorological data, the source-tracing or stray current spectral fingerprint database is updated. This process includes the following steps:

[0064] S311. Extract the spectral component with the largest amplitude from the spectral fingerprint to be traced;

[0065] S312. Using frequency as an indicator, determine whether a dominant frequency fingerprint exists in the stray current spectral fingerprint database. The dominant frequency fingerprint is a spectral fingerprint whose frequency difference with the spectral component with the largest amplitude is ≤ ±3%. If the determination result is negative, proceed to step S313. If the determination result is positive, extract the second largest amplitude spectral component from the spectral fingerprint to be traced and the dominant frequency fingerprint. Using frequency as an indicator, determine whether the frequency difference between the second largest amplitude spectral component in the spectral fingerprint to be traced and the second largest amplitude spectral component in the dominant frequency fingerprint is ≤ ±5%. If the determination result is positive, the spectral fingerprint to be traced and the dominant frequency fingerprint in the stray current spectral fingerprint database have the same cause of generation, and the tracing is completed. If the determination result is negative, proceed to step S313.

[0066] S313. Obtain external meteorological data for the same time period as the spectrum fingerprint to be traced, plot the waveform relationship between the spectrum component with the largest amplitude in the spectrum fingerprint to be traced and the external meteorological data. If the waveforms are related, the spectrum fingerprint to be traced is caused by the external meteorological data, and the tracing is completed. Update the spectrum fingerprint to be traced to the stray current spectrum fingerprint database.

[0067] The correlation between the spectral component with the largest amplitude in the traceable spectral fingerprint and the waveform of external meteorological data is determined using the following method:

[0068]

[0069] In the formula, γ1 is the correlation coefficient between the spectral component with the largest amplitude in the spectral fingerprint to be traced and external meteorological data. This represents the spectral component with the largest amplitude in the spectral fingerprint to be traced, along with the amount of external meteorological data for one week. express The Middle The spectral component with the largest amplitude in the traceable spectral fingerprint. express The Middle External meteorological data, This represents the average data volume over one week for the spectral component with the largest amplitude in the spectral fingerprint to be traced. express The average amount of domestic and foreign meteorological data over a week. This indicates the data number of the spectral component with the largest amplitude in the spectral fingerprint to be traced, and the data volume of external meteorological data for one week.

[0070] Calculate the correlation coefficient between the spectral component with the largest amplitude in the spectral fingerprint to be traced and the external meteorological data. If the absolute value of the correlation coefficient is greater than 10%, it is determined that the spectral component with the largest amplitude in the spectral fingerprint to be traced is associated with the waveform of the external meteorological data.

[0071] Determine the correlation between the long-term fingerprint of the source to be traced and the long-term fingerprint database of stray currents. Based on the determination result and combined with external meteorological data, complete the update of the source tracing or stray current long-term fingerprint database. The specific steps include:

[0072] S321. Based on the time period and region of the long-term fingerprint to be traced, determine whether there is a long-term fingerprint in the stray current long-term fingerprint database that is in the same time period and region. If the determination result is yes, the long-term fingerprint to be traced and the long-term fingerprint in the stray current long-term fingerprint database that is in the same time period and region have the same cause of generation, and the tracing is completed; if the determination result is no, proceed to step S322.

[0073] S322. Obtain the external meteorological data for the same time period as the long-term fingerprint to be traced, and draw the waveform relationship between the long-term fingerprint to be traced and the external meteorological data. If the waveforms are related, the long-term fingerprint to be traced is caused by the external meteorological data, and the tracing is completed. Update the long-term fingerprint to be traced to the stray current long-term fingerprint database.

[0074] The correlation between the long-term fingerprint to be traced and the waveform of external meteorological data is determined by the following method:

[0075]

[0076] In the formula, γ2 is the correlation coefficient between the long-term fingerprint to be traced and external meteorological data. This represents the amount of data collected over a week, including long-term fingerprints and external meteorological data, to be traced. express The Middle A long-term fingerprint that needs to be traced. express The Middle External meteorological data, This represents the average amount of long-term fingerprint data over one week that needs to be traced. express The average amount of domestic and foreign meteorological data over a week. This indicates the data number within a week's worth of data, including long-term fingerprints and external meteorological data to be traced.

[0077] Calculate the correlation coefficient between the long-term fingerprint to be traced and the external meteorological data. If the absolute value of the correlation coefficient is greater than 10%, it is determined that the long-term fingerprint to be traced is related to the waveform of the external meteorological data.

[0078] Determine the correlation between the short-time fingerprint of the source to be traced and the stray current short-time fingerprint database. Based on the determination result and combined with external meteorological data, complete the source tracing or stray current short-time fingerprint database update. The specific steps include:

[0079] S331. Based on the time period and region of the short-time fingerprint to be traced, determine whether there is a short-time fingerprint in the stray current short-time fingerprint database that is in the same time period and region. If the determination result is yes, the short-time fingerprint to be traced and the short-time fingerprint in the stray current short-time fingerprint database that is in the same time period and region have the same cause of generation, and the tracing is completed; if the determination result is no, proceed to step S332.

[0080] S332. Obtain external meteorological data for the same time period as the short-time fingerprint to be traced, and plot the waveform relationship between the short-time fingerprint to be traced and the external meteorological data. If the waveforms are correlated, the short-time fingerprint to be traced is caused by the external meteorological data, and the tracing is completed; and update the short-time fingerprint to be traced to the stray current short-time fingerprint database.

[0081] The correlation between the short-time fingerprint to be traced and the waveform of external meteorological data is determined by the following method:

[0082]

[0083] In the formula, γ3 is the correlation coefficient between the short-term fingerprint to be traced and external meteorological data. This represents the amount of short-term fingerprint and external meteorological data collected over one week for tracing purposes. express The Middle A short-term fingerprint to be traced. express The Middle External meteorological data, This represents the average amount of short-term fingerprint data over one week that needs to be traced. express The average amount of domestic and foreign meteorological data over a week. This indicates the data number within a week's worth of short-term fingerprint and external meteorological data to be traced;

[0084] Calculate the correlation coefficient between the short-time fingerprint to be traced and the external meteorological data. If the absolute value of the correlation coefficient is greater than 10%, it is determined that the waveform of the short-time fingerprint to be traced is related to the waveform of the external meteorological data.

[0085] By establishing a hierarchical fingerprint matching mechanism and a dynamic update strategy, accurate identification of stray current sources and continuous optimization of the fingerprint database are achieved. During the matching process between the stray current spectral fingerprint to be traced and the stray current spectral fingerprint database, frequency is used as an indicator to determine whether a dominant frequency fingerprint exists in the stray current spectral fingerprint database. The dominant frequency fingerprint is a spectral fingerprint whose frequency difference with the spectral component with the largest amplitude is ≤±3%. This is combined with a secondary verification that the frequency difference between the second largest amplitude spectral component in the stray current spectral fingerprint and the second largest amplitude spectral component in the dominant frequency fingerprint is ≤±5%, forming a dual frequency matching mechanism. This avoids the risk of misjudgment from single-frequency matching and ensures the correlation verification between different components. For long-term and short-term fingerprints to be traced, matching spatiotemporal characteristics can directly pinpoint interference sources with spatiotemporal regularity.

[0086] When matching fails, the environmental factors are incorporated into the source tracing system by introducing waveform correlation analysis of external meteorological data, effectively distinguishing between the inherent characteristics of the equipment and the sources of environmental interference.

[0087] The dynamic update mechanism enables the fingerprint database to evolve adaptively by adding new, unmatched features, allowing the system to continuously track the impact of dynamic factors such as equipment aging and environmental changes on stray current characteristics. The use of stray current spectral fingerprint databases, long-term stray current fingerprint databases, and short-term stray current fingerprint databases ensures rapid identification of known interference sources. The dynamic update mechanism updates new fingerprints to the corresponding fingerprint databases, enabling adaptive maintenance of these databases. Furthermore, continuously expanding the fingerprint samples in these databases improves the accuracy of source tracing.

[0088] Preferably, the criteria for determining whether there are long-term fingerprints in the same time period and the same region in the stray current long-term fingerprint database are as follows: the time period of the long-term fingerprint in the stray current long-term fingerprint database differs from the time period of the long-term fingerprint to be traced by less than one minute; and the location of the long-term fingerprint and the location of the long-term fingerprint to be traced differs from the location of the long-term fingerprint to be traced by less than 50km.

[0089] The criteria for determining whether short-time fingerprints from the same time period and region exist in the stray current short-time fingerprint database are as follows: the time period of the short-time fingerprint in the stray current short-time fingerprint database differs from the time period of the short-time fingerprint to be traced by less than one second; and the region of the short-time fingerprint differs from the region of the short-time fingerprint to be traced by less than 10 km.

[0090] Precise matching is achieved by setting differentiated time periods and regional locations. For long-term fingerprints, a dual constraint of a one-minute time period difference and a 50km regional location difference is adopted to consider the gradual change characteristics of long-term meteorological factors (such as diurnal temperature difference and regional climate) while avoiding false associations with cross-regional interference sources. For short-term fingerprints, a strict constraint of a one-second time period difference and a 10km regional location difference is adopted to capture the local features of transient events (such as lightning strikes and equipment start-up and shutdown), prevent false matching of short-term high-frequency interference, and thus improve the positioning accuracy of stray current sources.

[0091] Preferably, the external meteorological data includes at least solar radiation intensity and wind speed.

[0092] By explicitly including solar radiation intensity and wind speed as two key parameters in external meteorological data, the problem of insufficient correlation between meteorological factors in stray current tracing is resolved. Solar radiation intensity is a core influencing factor on the operating status of new energy equipment such as photovoltaic systems. Its changes directly lead to changes in the current characteristics of the equipment, thereby generating stray currents with specific spectral characteristics. Wind speed is an important environmental parameter for the operation of wind power generation systems. Changes in wind speed cause mechanical vibration and changes in the electromagnetic characteristics of the generator set, thus generating stray current characteristics related to wind speed. Using these two parameters as necessary external meteorological data allows for the establishment of a direct mapping relationship between external meteorological parameters and current characteristics, providing a crucial basis for determining whether stray currents are caused by external meteorological factors.

[0093] Preferred, see Figure 3 The extraction of long-time and short-time fingerprints from the very low frequency stray current component C using wavelet decomposition specifically includes the following steps:

[0094] S21. The very low frequency stray current component C is decomposed into 8 components by using the db4 wavelet in a 7-segment multi-resolution decomposition.

[0095] S22. Let C1 be the decomposition component with the lowest resolution. Extract the time nodes of all local peaks and local valleys from the waveform of decomposition component C1. Sort the time nodes in ascending order according to their chronological order. Based on the ascending order, construct the first sequence S = [s1, s2, ..., s...]. n 】, where s1, s2, ..., s n These are the 1st, 2nd, ..., nth subsequences of the first sequence; n is the total number of time points for all local peaks and local valleys;

[0096] S23. Based on the time differences between adjacent subsequences of the first sequence S, construct the second sequence dS = [d1, d2, ..., d...]. n-1 】, where d1, d2, ..., d n-1 These are the 1st, 2nd, ..., n-1th subsequences of the second sequence, respectively;

[0097] S24. Set the number of loops to K, and let K=1;

[0098] S25. Determine whether the Kth subsequence of the second sequence is greater than the preset time threshold. If the determination result is yes, then the component of the decomposed component C1 at the corresponding time node is used as a long-term fingerprint and stored in the pre-constructed stray current long-term fingerprint library. If the determination result is no, then the component of the decomposed component C1 at the corresponding time node is used as a short-term fingerprint and stored in the pre-constructed stray current short-term fingerprint library.

[0099] S26. Determine if K is less than n-1. If the result is yes, set K=K+1 and return to S25. If the result is no, execute step S3.

[0100] By combining a specific wavelet decomposition method with time series analysis, refined feature extraction of the very low frequency (VLF) stray current component C was achieved. A 7-segment multi-resolution decomposition using the db4 wavelet was employed, which can adapt to the non-stationary characteristics of VLF signals. The resulting eight decomposed components cover different frequency band characteristics, with the lowest resolution component C1 reflecting the low-frequency evolution trend. By extracting the time nodes of local peaks and valleys of the decomposed component C1 and arranging them in ascending order, a first sequence S is constructed to capture key changes in the signal in the time domain. A second sequence dS is then constructed based on the time difference between adjacent sub-sequence time nodes, quantifying the time interval characteristics of the fluctuations. By setting the number of iterations K and introducing a preset time threshold, the system dynamically determines whether the component of decomposed component C1 at the corresponding time node belongs to a long-term fingerprint or a short-term fingerprint, thus separating and storing slowly changing long-term features (such as the influence of natural meteorological factors) and short-term abrupt features (such as equipment failure) in decomposed component C1. By iterating through all sub-sequences of the second sequence dS, full-time coverage of decomposed component C1 is ensured, ultimately forming clearly categorized long-term and short-term fingerprints, providing a reliable basis for subsequent source tracing in conjunction with meteorological data.

[0101] In a specific embodiment of the present invention, the extraction of long-time and short-time fingerprints from the very low frequency stray current component C using wavelet decomposition can be implemented using MATLAB code.

[0102] In a specific embodiment of the present invention, the preset time threshold is 15 seconds.

[0103] Preferably, the preset sampling point is at least one of the cables, inverters, and combiner boxes of the new energy system, and the new energy system includes wind power system, photovoltaic system, and energy storage system.

[0104] By defining the specific locations of preset sampling points and the types of new energy systems, a detection system covering the core current path was constructed. Cables, inverters, and combiner boxes were chosen as sampling points because cables carrying the main current loop are prone to conducted interference; inverters, as core power conversion equipment, exhibit high-frequency switching characteristics; and combiner boxes, as nodes where multiple branches converge, are prone to superimposed current effects, representing key links in the generation and propagation of stray currents. Wind power systems, photovoltaic systems, and energy storage systems were limited to the detection targets because the pitch control system and generator of wind power systems, the DC-AC conversion links of photovoltaic systems, and the charging and discharging circuits of energy storage systems have different stray current generation mechanisms. Targeted fingerprint databases need to be established through classified detection, ensuring full coverage of key current nodes and achieving differentiated capture of stray current characteristics from different new energy systems. This lays a data foundation for the accurate extraction of subsequent spectral fingerprints, long-term fingerprints, and short-term fingerprints.

[0105] Preferably, acquiring the high-frequency stray current component, low-frequency stray current component, and very low-frequency stray current component at preset sampling points specifically includes the following steps:

[0106] S11. Determine whether the installation location of the preset sampling point is suitable for an open Hall sensor. If the determination result is yes, proceed to step S12; otherwise, proceed to step S13.

[0107] S12. Based on the open Hall sensor, the original current data of the preset sampling point is obtained, and the high-frequency stray current component A, low-frequency stray current component B and very low-frequency stray current component C are extracted from the original current data using a filter.

[0108] S13. The high-frequency stray current component A and the low-frequency stray current component B of the preset sampling point are directly extracted by the Rogowski coil sensor, and the very low-frequency stray current component C of the preset sampling point is directly extracted by the fluxgate sensor.

[0109] Existing stray current measurement technologies primarily originate from polarization current detection methods used in the metal corrosion detection industry. There are no dedicated testing methods for stray currents in power systems specifically designed for new energy systems. Currently, the main current detection methods used in power systems include open-aperture Hall effect sensors, Rogowski coil sensors, and current transformers. Current transformers are primarily used for power frequency current measurement in metering scenarios and are not suitable for measuring stray currents in new energy systems. Open-aperture Hall effect sensors are mainly used to measure stray currents in small conductors (diameter less than 5cm). When the installation location is unsuitable for open-aperture Hall effect sensors, a combination of a Rogowski coil sensor and a large fluxgate sensor can be used for stray current measurement.

[0110] By dynamically selecting sensor combinations and signal processing strategies, the adaptability problem of stray current acquisition in complex scenarios of new energy systems is solved. For applicable scenarios, a single open-aperture Hall sensor combined with filtering technology is used to achieve full-band component extraction; for unapplicable scenarios, a combination of Rogowski coil and fluxgate sensor is adopted to directly separate high-frequency stray current component A, low-frequency stray current component B, and very low-frequency stray current component C through the physical characteristics of different sensors. The open-aperture Hall sensor is selected based on its non-contact measurement advantage, which is suitable for space-constrained installation locations; while the high-frequency response characteristics of the Rogowski coil and the low-frequency sensitivity characteristics of the fluxgate sensor complement each other, ensuring independent acquisition of each frequency band component in complex electromagnetic environments. The scene adaptability of the detection device is achieved through a conditional judgment mechanism.

[0111] Preferably, the sampling rate of the open Hall sensor is 16 kHz; the sampling rate of the Rogowski coil sensor is 16 kHz; and the sampling rate of the fluxgate sensor is 10 Hz.

[0112] By setting specific sampling rate parameters, adaptive data acquisition is performed for the physical characteristics of stray currents in different frequency bands. Both the open-aperture Hall sensor and the Rogowski coil sensor adopt a high sampling rate of 16kHz, which can accurately capture the fast transient characteristics of high-frequency stray current component A and low-frequency stray current component B, avoiding aliasing distortion during sampling. The fluxgate sensor adopts a sampling rate of 10Hz, which ensures the integrity of the very low frequency stray current component C while taking into account the high sensitivity of the fluxgate sensor to extremely low frequency signals. This effectively solves the technical difficulty of accurately extracting the very low frequency stray current component C due to its weak amplitude and slow change, providing a raw data foundation that conforms to the physical characteristics of different frequency bands for the subsequent construction of the fingerprint database.

[0113] Preferably, step S12 can be replaced by the following steps:

[0114] The original current data of the preset sampling point is obtained based on the open Hall sensor. The high-frequency stray current component, low-frequency stray current component and very low-frequency stray current component in the original current data are extracted by the filter. It is determined whether the very low-frequency stray current component is greater than 0.1 times the root mean square of the sum of the high-frequency stray current component and the low-frequency stray current component. If the determination result is yes, the high-frequency stray current component, low-frequency stray current component and very low-frequency stray current component in the original current data are the high-frequency stray current component A, low-frequency stray current component B and very low-frequency stray current component C of the preset sampling point. If the determination result is no, step S13 is executed.

[0115] By establishing a dynamic sensor data validity verification mechanism, the acquisition process of high-frequency stray current component A, low-frequency stray current component B, and very low-frequency stray current component C at preset sampling points was optimized. First, after acquiring the raw current data based on the open-aperture Hall sensor, a filter is used for multi-band separation. A quantitative verification standard is set for the very low frequency stray current component: it is determined whether the very low frequency stray current component is greater than 0.1 times the root mean square of the sum of the high frequency stray current component and the low frequency stray current component. If the judgment result is yes, the detection result of the open-aperture Hall sensor is valid, and the high frequency stray current component, low frequency stray current component, and very low frequency stray current component in the raw current data can be directly identified as the high frequency stray current component A, low frequency stray current component B, and very low frequency stray current component C of the preset sampling point. If the judgment result is no, the backup detection scheme is triggered, switching to the combined detection method of Rogowski coil sensor and fluxgate sensor. This dual verification mechanism not only ensures the data reliability in the applicable scenarios of the open-aperture Hall sensor, but also automatically switches to a more accurate detection method when the detection accuracy of the open-aperture Hall sensor is insufficient, avoiding the detection blind zone problem that may exist in complex field environments.

[0116] Based on the same inventive concept, this invention also provides a source tracing system for stray currents in a new energy system, comprising:

[0117] The first module is used to acquire the high-frequency stray current component A, the low-frequency stray current component B, and the very low-frequency stray current component C at preset sampling points of the new energy system.

[0118] The second module is used to extract the spectral fingerprints of high-frequency stray current component A and low-frequency stray current component B using fast Fourier transform and store them in a pre-built stray current spectral fingerprint library; and to extract the long-time fingerprint and short-time fingerprint of very low-frequency stray current component C using wavelet decomposition and store them in the pre-built stray current long-time fingerprint library and stray current short-time fingerprint library, respectively.

[0119] The third module is used to acquire the high-frequency stray current component a, the low-frequency stray current component b, and the very low-frequency stray current component c of the current data to be traced, and to generate the spectral fingerprint, long-term fingerprint, and short-term fingerprint to be traced. The module then determines the correlation between the spectral fingerprint and the stray current spectral fingerprint database, the long-term fingerprint and the stray current long-term fingerprint database, and the short-term fingerprint and the stray current short-term fingerprint database, respectively. Based on the determination results and combined with external meteorological data, the module completes the source tracing or updates the stray current spectral fingerprint database, the stray current long-term fingerprint database, and the stray current short-term fingerprint database.

[0120] This system can guide the detection of stray currents and the identification of their sources in the field.

[0121] See Figure 2 In a specific embodiment of the present invention, a stray current tracing system for a new energy system is provided. When satellite signals are available, a satellite signal antenna receives a satellite clock signal, and a microsecond-level time is obtained through a satellite clock signal circuit. This microsecond-level time is then stored in a timing circuit. The timing circuit sends the stored microsecond-level time to a data storage and analysis module. Simultaneously, an oscillation timer circuit provides a 160MHz crystal oscillator signal. When no satellite signal is available, the timing circuit sends internal clock data to the data storage and analysis module. In scenarios where an open-aperture Hall sensor is applicable, the output of the open-aperture Hall sensor is sampled at a high speed using a sampling rate of 16kHz, and the sampled data is stored in a data processing and storage module. In scenarios where an open-aperture Hall sensor is not applicable, the output of a Rogowski coil sensor is sampled at a high speed using a sampling rate of 16kHz, and the sampled data is stored in the data processing and storage module. Simultaneously, the output of a large-size fluxgate sensor is sampled using a sampling rate of 10Hz, and the sampled data is stored in the data processing and storage module. The data processing and storage module then processes the data to obtain spectral fingerprints, long-time fingerprints, and short-time fingerprints.

[0122] In a specific embodiment of the present invention, wavelet decomposition is used to extract long-time and short-time fingerprints from the very low frequency stray current component. See [link to relevant documentation]. Figure 4 For the extracted long-term fingerprint, see Figure 5 The extracted short-time fingerprint is obtained. Through wavelet decomposition, the lowest resolution decomposition component C1 in the original waveform of the irregularly changing very low frequency stray current component C is extracted. C1 is taken as the principal component of the very low frequency stray current component, and the time nodes of all local peaks and local valleys are extracted from the waveform of the decomposition component C1 as the time features of the long-time fingerprint and the short-time fingerprint.

[0123] The high-frequency and low-frequency stray current components essentially originate from the harmonic currents of the rectifier and DC boost converter in the renewable energy power generation system, as well as the harmonic currents of the grid-connected inverter, and are related to the real-time renewable energy power generation. The real-time renewable energy power generation is closely related to real-time solar radiation intensity and wind speed (renewable energy power generation exhibits a correlated nonlinear saturation relationship with solar radiation intensity and wind speed). The harmonic currents of the rectifier and DC boost converter in the renewable energy power generation system, and the grid-connected inverter, are related to the manufacturer's control methods and processes. Therefore, the high-frequency and low-frequency stray currents inherently carry clear manufacturer characteristics, allowing for traceability based on waveform analysis.

[0124] In a specific embodiment of the present invention, the spectral component with the largest amplitude in the traceable spectral fingerprint is the second harmonic, and there is no dominant frequency fingerprint in the stray current spectral fingerprint database. External meteorological data for the same time period as the traceable spectral fingerprint is obtained. Using the spectral component with the largest amplitude in the traceable spectral fingerprint, combined with the external meteorological data, a waveform relationship diagram is plotted between the effective value of the second harmonic and the wind speed and solar radiation intensity within the same time period. (See [reference]). Figure 6 and Figure 7 The correlation coefficient between the spectral component with the largest amplitude in the traceable spectral fingerprint and wind speed was calculated to be -32.94%; the correlation coefficient between the spectral component with the largest amplitude in the traceable spectral fingerprint and solar radiation intensity was calculated to be -14.94%. The absolute values ​​of both correlation coefficients are greater than 10%, indicating that the spectral component with the largest amplitude in the traceable spectral fingerprint is correlated with both wind speed and solar radiation intensity. Furthermore, the absolute value of the correlation coefficient between the spectral component with the largest amplitude in the traceable spectral fingerprint and wind speed is greater than the absolute value of the correlation coefficient between the spectral component with the largest amplitude in the traceable spectral fingerprint and solar radiation intensity. In practical applications, the impact caused by solar radiation intensity can be prioritized for mitigation.

[0125] Because wind speed fluctuates considerably, the short-term fluctuations of the second harmonic (fluctuations with periods on the order of minutes) are inferred to be caused by changes in wind speed. Since solar radiation intensity changes slowly, the long-term fluctuations of the second harmonic (fluctuations with periods on the order of days) are caused by changes in solar radiation intensity. If the spectral component with the largest amplitude is a stationary waveform with a period on the order of months, then the spectral component with the largest amplitude is unrelated to wind speed and solar radiation intensity.

[0126] Very low frequency (VLF) stray current components are generally caused by DC leakage current due to contamination of the support structures and cable insulation defects in new energy systems such as photovoltaics and energy storage. In the ideal state at the beginning of operation, when there is no contamination of the support structures and cable insulation defects, the stray current corresponding to each MW of new energy power generation is approximately no more than 0.01 mA. After operation, as the degree of contamination of the support structures and the increase in cable insulation defects, the leakage current of the support structures and cables intensifies, and VLF stray current components also appear. Simultaneously, some wind power projects are also equipped with energy storage facilities, which can lead to the presence of VLF stray current components. In a specific embodiment of this invention, see [reference needed]. Figure 8 and Figure 9 The correlation coefficient between the long-term fingerprint to be traced and wind speed was calculated to be -32.31%, and the correlation coefficient between the long-term fingerprint to be traced and solar radiation intensity was calculated to be 13.01%. The absolute values ​​of both correlation coefficients are greater than 10%, indicating that the long-term fingerprint to be traced is correlated with both wind speed and solar radiation intensity. In a specific embodiment of the present invention, see... Figure 10 and Figure 11 The correlation coefficient between the short-term fingerprint to be traced and wind speed was calculated to be -57.06%; the correlation coefficient between the short-term fingerprint to be traced and solar radiation intensity was calculated to be 24.97%. The absolute values ​​of both correlation coefficients are greater than 10%, indicating that the long-term fingerprint to be traced is correlated with both wind speed and solar radiation intensity. Finally, it should be noted that the embodiments listed above are merely one or more specific manifestations of the technical solution of this invention. Their purpose is to clearly illustrate the concept, principle, and application of this invention through specific examples, and is by no means intended to limit the scope of protection of this invention to these specific embodiments. In fact, the true value of this invention lies in its proposed technical ideas and innovations, rather than its manifestations or implementation methods.

[0127] For those skilled in the art, after thoroughly reading and understanding the technical solution of this invention, they are fully capable of making various changes, modifications, or equivalent substitutions to the specific implementation of the invention based on their own professional knowledge and skills. These changes may include, but are not limited to: adjusting the range of technical parameters, optimizing the algorithm flow to improve efficiency, and replacing some technical components to achieve better compatibility or reduce costs. As long as these modified technical solutions substantially retain the technical features claimed by the original invention, that is, they can still achieve the core functions and effects of this invention, then these changes should be considered to fall within the scope of protection of the pending claims of this invention.

[0128] Furthermore, with the continuous progress and development of technology, new technical means and methods are constantly emerging, which provides ample space for further improvement and perfection of this invention. Therefore, the scope of protection of this invention should also include reasonable and foresightful improvements and extensions based on existing technology. As long as these improvements and extensions do not depart from the basic principles and core concepts of this invention, they should be considered equivalents of this invention and are equally protected by patent rights.

Claims

1. A method for tracing the source of stray current in a new energy system, characterized in that, The method comprises the following steps: S1, obtaining high-frequency stray current component A, low-frequency stray current component B and very low-frequency stray current component C of a preset sampling point of a new energy system; S2, extracting the frequency spectrum fingerprints of the high-frequency stray current component A and the low-frequency stray current component B by using fast Fourier transform and storing them in a pre-constructed stray current frequency spectrum fingerprint library; extracting the long-time fingerprints and short-time fingerprints of the very low-frequency stray current component C by using wavelet decomposition and storing them in a pre-constructed stray current long-time fingerprint library and a pre-constructed stray current short-time fingerprint library, respectively; S3, obtaining high-frequency stray current component a, low-frequency stray current component b and very low-frequency stray current component c of current data to be traced, generating a to-be-traced frequency spectrum fingerprint, a to-be-traced long-time fingerprint and a to-be-traced short-time fingerprint, respectively judging the correlation of the to-be-traced frequency spectrum fingerprint with the stray current frequency spectrum fingerprint library, the correlation of the to-be-traced long-time fingerprint with the stray current long-time fingerprint library and the correlation of the to-be-traced short-time fingerprint with the stray current short-time fingerprint library, and completing tracing or updating the stray current frequency spectrum fingerprint library, the stray current long-time fingerprint library and the stray current short-time fingerprint library according to the judgment result and in combination with external meteorological data; judging the correlation of the to-be-traced frequency spectrum fingerprint with the stray current frequency spectrum fingerprint library, completing tracing or updating the stray current frequency spectrum fingerprint library according to the judgment result and in combination with external meteorological data, and specifically comprising the following steps: S311, extracting the frequency spectrum component with the largest amplitude in the to-be-traced frequency spectrum fingerprint; S312, taking frequency as an index to judge whether a main frequency fingerprint exists in the stray current frequency spectrum fingerprint library, the main frequency fingerprint being a frequency spectrum fingerprint with a frequency difference of ≤±3% from the frequency spectrum component with the largest amplitude; if the judgment result is no, step S313 is performed; if the judgment result is yes, the frequency spectrum component with the second largest amplitude in the to-be-traced frequency spectrum fingerprint and the main frequency fingerprint is extracted, and the frequency difference of the frequency spectrum component with the second largest amplitude in the to-be-traced frequency spectrum fingerprint from the frequency spectrum component with the second largest amplitude in the main frequency fingerprint is judged; if the judgment result is yes, the to-be-traced frequency spectrum fingerprint and the main frequency fingerprint in the stray current frequency spectrum fingerprint library have the same cause, and tracing is completed; if the judgment result is no, step S313 is performed; S313, obtaining external meteorological data of the to-be-traced frequency spectrum fingerprint in the same time period, drawing a waveform relationship diagram of the frequency spectrum component with the largest amplitude in the to-be-traced frequency spectrum fingerprint and the external meteorological data; if the waveforms are associated, the to-be-traced frequency spectrum fingerprint is caused by the external meteorological data, and tracing is completed; and the to-be-traced frequency spectrum fingerprint is updated to the stray current frequency spectrum fingerprint library; judging the correlation of the to-be-traced long-time fingerprint with the stray current long-time fingerprint library, completing tracing or updating the stray current long-time fingerprint library according to the judgment result and in combination with external meteorological data, and specifically comprising the following steps: S321, judging whether a long-time fingerprint in the same time period and the same region exists in the stray current long-time fingerprint library based on the time period and the region of the to-be-traced long-time fingerprint; if the judgment result is yes, the to-be-traced long-time fingerprint and the long-time fingerprint in the same time period and the same region in the stray current long-time fingerprint library have the same cause, and tracing is completed; if the judgment result is no, step S322 is performed; S322, external meteorological data of the same time period as the long-time fingerprint to be traced is obtained, a waveform relationship diagram of the long-time fingerprint to be traced and the external meteorological data is drawn, if the waveforms are associated, the long-time fingerprint to be traced is caused by the external meteorological data, the tracing is completed, and the long-time fingerprint to be traced is updated to the stray current long-time fingerprint library; The correlation of the short-time fingerprint to be traced and the stray current short-time fingerprint library is judged, and according to the judgment result, the external meteorological data is combined to complete the tracing or update the stray current short-time fingerprint library, and the method specifically comprises the following steps: S331, whether there is a short-time fingerprint of the same time period and the same region in the stray current short-time fingerprint library is judged based on the time period and the region of the short-time fingerprint to be traced, if the judgment result is yes, the short-time fingerprint to be traced and the short-time fingerprint of the same time period and the same region in the stray current short-time fingerprint library have the same cause, the tracing is completed, if the judgment result is no, step S332 is executed; S332, external meteorological data of the same time period as the short-time fingerprint to be traced is obtained, a waveform relationship diagram of the short-time fingerprint to be traced and the external meteorological data is drawn, if the waveforms are associated, the short-time fingerprint to be traced is caused by the external meteorological data, the tracing is completed, and the short-time fingerprint to be traced is updated to the stray current short-time fingerprint library.

2. The method of claim 1, wherein, The judgment basis that the long-time fingerprint of the same time period and the same region exists in the stray current long-time fingerprint library is specifically that the time period of the long-time fingerprint in the stray current long-time fingerprint library is within one minute different from the time period of the long-time fingerprint to be traced, and the region of the long-time fingerprint is within 50 km different from the region position of the long-time fingerprint to be traced; The judgment basis that the short-time fingerprint of the same time period and the same region exists in the stray current short-time fingerprint library is specifically that the time period of the short-time fingerprint in the stray current short-time fingerprint library is within one second different from the time period of the short-time fingerprint to be traced, and the region of the short-time fingerprint is within 10 km different from the region position of the short-time fingerprint to be traced.

3. The method of tracing the source of stray current in a new energy system according to claim 1, wherein, The external meteorological data at least includes solar radiation intensity and wind speed.

4. The method of tracing the source of stray current in a new energy system according to claim 1, wherein, The long-time fingerprint and the short-time fingerprint in the very low frequency stray current component C are extracted by using wavelet decomposition, and the method specifically comprises the following steps: S21, db4 wavelet is used for 7-section multi-resolution decomposition of the very low frequency stray current component C, and 8 decomposition components are obtained; S22, set the decomposition component with the lowest resolution as C1, extract all local peak and local valley time nodes from the waveform of the decomposition component C1, arrange the time nodes in ascending order according to the chronological order, and based on the ascending order arrangement result, construct a first sequence S=【s1, s2, …, sn】. n 】,wherein s1, s2, …, sn are the 1st, 2nd, …, nth subsequences of the first sequence respectively; n is the total number of all local peak and local valley time nodes. n ​ S23、based on the time difference of adjacent sub-sequences of the first sequence S, a second sequence dS=【d1, d2, …, dn-1】 is constructed, wherein d1, d2, …, dn-1 are the 1st, 2nd, …, n-1st sub-sequences of the second sequence respectively. n-1 】. n-1 d1, d2, …, dn-1 are the 1st, 2nd, …, n-1st sub-sequences of the second sequence respectively. S24, the cycle number is set as K, and K=1; S25, whether the Kth subsequence of the second sequence is greater than a preset time threshold is judged, if the judgment result is yes, the component of the decomposition component C1 at the corresponding time node is taken as the long-time fingerprint, and is stored in the pre-constructed stray current long-time fingerprint library, if the judgment result is no, the component of the decomposition component C1 at the corresponding time node is taken as the short-time fingerprint, and is stored in the pre-constructed stray current short-time fingerprint library; S26, whether K is less than n-1 is judged, if the judgment result is yes, K=K+1 is set, and S25 is returned, if the judgment result is no, step S3 is executed.

5. The method of tracing the source of stray current in a new energy system according to claim 1, wherein, The preset sampling point is at least one of a cable, an inverter and a busbar of a new energy system, and the new energy system includes a wind power system, a photovoltaic system and an energy storage system.

6. The method of tracing the source of stray current in a new energy system according to claim 1, wherein, The high-frequency stray current component, the low-frequency stray current component and the very low-frequency stray current component of the preset sampling point are acquired, and the method specifically comprises the following steps: S11, judging whether the installation position of the preset sampling point is suitable for the open Hall sensor, if the result of the judgment is yes, executing step S12, if the result of the judgment is no, executing step S13; S12, acquiring the original current data of the preset sampling point based on the open Hall sensor, and extracting the high-frequency stray current component A, the low-frequency stray current component B and the very low-frequency stray current component C in the original current data by using a filter; S13, directly extracting the high-frequency stray current component A and the low-frequency stray current component B of the preset sampling point by using the Rogowski coil sensor, and directly extracting the very low-frequency stray current component C of the preset sampling point by using the fluxgate sensor.

7. The method of tracing the source of stray current in a new energy system according to claim 6, characterized in that, The sampling rate of the open Hall sensor is 16 kHz; the sampling rate of the Rogowski coil sensor is 16 kHz; and the sampling rate of the fluxgate sensor is 10 Hz.

8. The method of tracing stray current of a new energy system according to claim 6, characterized in that, The step S12 can be replaced by the following steps: Acquiring the original current data of the preset sampling point based on the open Hall sensor, extracting the high-frequency stray current component, the low-frequency stray current component and the very low-frequency stray current component in the original current data by using a filter, judging whether the very low-frequency stray current component is greater than 0.1 times the root mean square of the sum of the high-frequency stray current component and the low-frequency stray current component, if the result of the judgment is yes, the high-frequency stray current component A, the low-frequency stray current component B and the very low-frequency stray current component C of the preset sampling point are the high-frequency stray current component, the low-frequency stray current component and the very low-frequency stray current component in the original current data, if the result of the judgment is no, executing step S13.

9. A system for tracing a stray current of a new energy system, characterized in that, Comprise: A first module for acquiring the high-frequency stray current component A, the low-frequency stray current component B and the very low-frequency stray current component C of the preset sampling point of the new energy system; A second module for extracting the frequency spectrum fingerprint of the high-frequency stray current component A and the low-frequency stray current component B by using fast Fourier transform and storing in a pre-constructed stray current frequency spectrum fingerprint library; Extracting the long-time fingerprint and the short-time fingerprint in the very low-frequency stray current component C by using wavelet decomposition and storing in a pre-constructed stray current long-time fingerprint library and a stray current short-time fingerprint library respectively; A third module for acquiring the high-frequency stray current component a, the low-frequency stray current component b and the very low-frequency stray current component c of the current data to be traced, generating the frequency spectrum fingerprint to be traced, the long-time fingerprint to be traced and the short-time fingerprint to be traced, judging the correlation of the frequency spectrum fingerprint to be traced and the stray current frequency spectrum fingerprint library, the correlation of the long-time fingerprint to be traced and the stray current long-time fingerprint library and the correlation of the short-time fingerprint to be traced and the stray current short-time fingerprint library respectively, and completing the tracing or the update of the stray current frequency spectrum fingerprint library, the stray current long-time fingerprint library and the stray current short-time fingerprint library according to the judgment result and in combination with external meteorological data; Judging the correlation of the frequency spectrum fingerprint to be traced and the stray current frequency spectrum fingerprint library, and completing the update of the stray current frequency spectrum fingerprint library according to the judgment result and in combination with external meteorological data, and the method specifically comprises the following steps: S311, extracting the frequency spectrum component with the largest amplitude in the frequency spectrum fingerprint to be traced; S312, judging whether the main frequency fingerprint exists in the stray current spectrum fingerprint library with frequency as an index, the main frequency fingerprint being a spectrum fingerprint with a frequency difference of ≤±3% from the spectrum component with the largest amplitude; if the judgment result is no, step S313 is executed; if the judgment result is yes, the spectrum component with the second largest amplitude in the stray current spectrum fingerprint and the main frequency fingerprint is extracted, and whether the frequency difference between the spectrum component with the second largest amplitude in the stray current spectrum fingerprint and the spectrum component with the second largest amplitude in the main frequency fingerprint is ≤±5% is judged with frequency as an index; if the judgment result is yes, the stray current spectrum fingerprint and the main frequency fingerprint in the stray current spectrum fingerprint library have the same cause, and the tracing is completed; if the judgment result is no, step S313 is executed; S313, obtaining the external meteorological data of the same time period as the stray current spectrum fingerprint, drawing a waveform relationship diagram of the spectrum component with the largest amplitude in the stray current spectrum fingerprint and the external meteorological data, if the waveforms are associated, the stray current spectrum fingerprint is caused by the external meteorological data, the tracing is completed, and the stray current spectrum fingerprint is updated to the stray current spectrum fingerprint library; judging the correlation between the long-time fingerprint to be traced and the stray current long-time fingerprint library, according to the judgment result, combining the external meteorological data, the tracing or the update of the stray current long-time fingerprint library is completed, specifically including the following steps: S321, judging whether the long-time fingerprint of the same time period and the same region exists in the stray current long-time fingerprint library based on the time period and the region of the long-time fingerprint to be traced; if the judgment result is yes, the long-time fingerprint to be traced and the long-time fingerprint of the same time period and the same region in the stray current long-time fingerprint library have the same cause, and the tracing is completed; if the judgment result is no, step S322 is executed; S322, obtaining the external meteorological data of the same time period as the long-time fingerprint to be traced, drawing a waveform relationship diagram of the long-time fingerprint to be traced and the external meteorological data, if the waveforms are associated, the long-time fingerprint to be traced is caused by the external meteorological data, the tracing is completed, and the long-time fingerprint to be traced is updated to the stray current long-time fingerprint library; judging the correlation between the short-time fingerprint to be traced and the stray current short-time fingerprint library, according to the judgment result, combining the external meteorological data, the tracing or the update of the stray current short-time fingerprint library is completed, specifically including the following steps: S331, judging whether the short-time fingerprint of the same time period and the same region exists in the stray current short-time fingerprint library based on the time period and the region of the short-time fingerprint to be traced; if the judgment result is yes, the short-time fingerprint to be traced and the short-time fingerprint of the same time period and the same region in the stray current short-time fingerprint library have the same cause, and the tracing is completed; if the judgment result is no, step S332 is executed; S332, obtaining the external meteorological data of the same time period as the short-time fingerprint to be traced, drawing a waveform relationship diagram of the short-time fingerprint to be traced and the external meteorological data, if the waveforms are associated, the short-time fingerprint to be traced is caused by the external meteorological data, the tracing is completed, and the short-time fingerprint to be traced is updated to the stray current short-time fingerprint library.

Images