Power distribution line fault location method and system

Abstract

The application provides a power distribution line fault positioning method and system. Transient current traveling wave signals of each monitoring section of a power distribution line and measurement data of a doubly-fed wind turbine are collected when a fault occurs. Equivalent inductance analysis is performed on the doubly-fed wind turbine based on the measurement data to obtain a dominant characteristic frequency band. A plurality of transient signal characteristic quantities are determined according to the dominant characteristic frequency band and the transient current traveling wave signals. The energy distribution of the transient current is determined according to all the transient signal characteristic quantities. The transient energy of each monitoring section of the power distribution line is evaluated according to the energy distribution to obtain the distribution difference of each monitoring section. When the distribution difference of any monitoring section exceeds a threshold value, the monitoring section is determined to be a fault section, and the distance of the fault point in the fault section is determined. The scheme of the application can be used to position the fault of a weak power grid power distribution line under the coupling mechanism of the transient response characteristics of the doubly-fed wind turbine and the meshed loop.

Description

TECHNICAL FIELD

[0001] The present application relates to fault location, more particularly, the present application relates to a power distribution line fault location method and system. BACKGROUND

[0002] The meshed and parallel circuit containing a doubly-fed wind turbine is a typical structure formed after high proportion of wind power is connected to the power distribution network. When the power grid fails, the doubly-fed wind turbine no longer behaves as a traditional single power source, thereby causing the fault current to exhibit a multi-source superposition characteristic, and the transient process is simultaneously affected by multiple factors including the wind turbine control strategy, crowbar protection circuit and power grid operating state.

[0003] In the existing power distribution line fault location method, the external power grid is simplified as a constant voltage source based on the synchronous generator model or ideal voltage source assumption, and the location is achieved by analyzing the transient signal energy or traveling wave propagation time; however, in the weak power grid with high proportion of doubly-fed wind turbines, the fault current output by the wind turbine will significantly affect the system voltage, and the unique rotor transient alternating current component, stator transient direct current component and power grid form electromagnetic coupling, thereby causing the traditional method to produce significant errors due to the neglect of the electromagnetic transient interaction between the power grid impedance and the wind turbine, and the positioning accuracy and adaptability are insufficient; therefore, how to locate the fault of the power distribution line of the weak power grid under the coupling mechanism of the transient response characteristics of the doubly-fed wind turbine and the meshed and parallel circuit has become a difficult problem in the industry. SUMMARY

[0004] The present application provides a power distribution line fault location method and system, which can locate the fault of the power distribution line of the weak power grid under the coupling mechanism of the transient response characteristics of the doubly-fed wind turbine and the meshed and parallel circuit.

[0005] In a first aspect, the present application provides a power distribution line fault location method, wherein the power distribution line is a meshed and parallel circuit containing a doubly-fed wind turbine, and the power distribution line is divided into a plurality of monitoring sections in advance, and the method comprises the following steps:

[0006] Collecting the measurement data of the transient current traveling wave signals generated at the occurrence of the fault at the head and tail ends of each monitoring section of the power distribution line, and the terminal voltage and current of the doubly-fed wind turbine;

[0007] Based on the measurement data, the rotor transient of the doubly-fed wind turbine at the fault of the power distribution line is analyzed by equivalent inductance, the dominant characteristic frequency band generated by the doubly-fed wind turbine at the fault of the power distribution line is obtained, and then a plurality of transient signal characteristic quantities of the power distribution line are determined according to the dominant characteristic frequency band and the transient current traveling wave signals;

[0008] The energy distribution of transient current in the power distribution line during a fault is determined based on all transient signal characteristics. Then, the difference in transient energy between the beginning and end of each monitoring section of the power distribution line is evaluated based on the energy distribution to obtain the distribution difference of each monitoring section.

[0009] All distribution differences are compared with a preset entropy threshold. When the distribution difference of any monitoring segment exceeds the threshold, the monitoring segment is determined to be a fault segment, thereby determining the distance of the fault point within the fault segment.

[0010] In some embodiments, based on the measurement data, the equivalent inductance analysis of the rotor transient of the doubly-fed induction generator (DFIG) during a power line fault is performed to obtain the dominant characteristic frequency band generated by the DFIG during the power line fault, specifically including:

[0011] Based on the measurement data, an equivalent circuit simulation was performed on the mesh parallel circuit containing the doubly fed fan to obtain the transient flux response of each electrical circuit of the doubly fed fan.

[0012] The stator transient DC component generated by stator flux conservation induction is determined based on all transient flux responses;

[0013] The rotor flux conservation and the rotor transient AC component caused by the crowbar input are determined based on all transient flux responses.

[0014] Waveform spectrum analysis of the stator transient DC component and the rotor transient AC component during fault conditions is performed to obtain the dominant characteristic frequency band generated by the doubly fed wind turbine during power distribution line fault conditions.

[0015] In some embodiments, determining multiple transient signal characteristic quantities of the power distribution line based on the dominant characteristic frequency band and the transient current traveling wave signal specifically includes:

[0016] The transient current traveling wave signal is decomposed into a multi-scale decomposition to obtain a wavelet packet decomposition tree;

[0017] Based on the wavelet packet decomposition tree, multiple energy nodes of the dominant feature frequency band are determined;

[0018] Multiple transient signal characteristics of the power distribution line are extracted from all energy nodes using the dominant characteristic frequency band.

[0019] In some embodiments, determining the energy distribution of the transient current of the distribution line during a fault, based on all transient signal characteristics, specifically includes:

[0020] The energy set at the beginning and the energy set at the end of each monitoring section of the power distribution line are determined based on all transient signal characteristics.

[0021] The probability distribution of multiple head-end energies of the transient current is determined based on all head-end energy sets.

[0022] The probability distribution of multiple end energies of the transient current is determined based on all the end energy sets;

[0023] The set of all the energy probability distributions at the beginning and all the energy probability distributions at the end is taken as the energy distribution of the transient current of the distribution line when the distribution line is faulty.

[0024] In some embodiments, the transient energy differences between the beginning and end of each monitoring section of the power distribution line are evaluated based on the energy distribution, and the distribution differences of each monitoring section are specifically included:

[0025] Obtain the smoothing factor;

[0026] Select a monitoring section as the selected monitoring section;

[0027] Based on the smoothing factor, the relative entropy of transient energy at the beginning and end of the selected monitoring section is determined for the energy distribution;

[0028] The relative entropy is used as the distribution difference when there is a fault at the beginning or end of the selected monitoring section;

[0029] Continue to determine the distribution differences in the remaining monitoring sections.

[0030] In some embodiments, determining the distance to the fault point within the fault zone specifically includes:

[0031] Obtain the absolute timestamps of the arrival of the initial traveling waves at the beginning and end;

[0032] Based on the absolute timestamp, the faulty section is located using a double-ended traveling wave method to obtain the distance to the fault point.

[0033] In some embodiments, a traveling wave acquisition device is used to acquire transient current traveling wave signals generated at the beginning and end of the power distribution line when a fault occurs.

[0034] In some embodiments, the measurement data of the terminal voltage and current of the doubly fed wind turbine are acquired by an electrical quantity measurement unit.

[0035] In some embodiments, all distribution differences are compared with a preset entropy threshold. If the distribution difference of any current monitoring segment does not exceed the threshold, the monitoring segment is continuously monitored.

[0036] Secondly, this application provides a power distribution line fault location system, which includes:

[0037] The data acquisition module is used to acquire transient current traveling wave signals generated at the beginning and end of each monitoring section of the power distribution line when a fault occurs, as well as measurement data of the terminal voltage and current of the doubly fed wind turbine.

[0038] The processing module is used to perform equivalent inductance analysis on the rotor transient of the doubly fed wind turbine when the power distribution line is faulted based on the measurement data, to obtain the dominant characteristic frequency band generated by the doubly fed wind turbine when the power distribution line is faulted, and then to determine multiple transient signal characteristic quantities of the power distribution line based on the dominant characteristic frequency band and the transient current traveling wave signal.

[0039] The processing module is also used to determine the energy distribution of the transient current of the power distribution line when the power distribution line is faulty based on all transient signal characteristics, and then to evaluate the difference in transient energy between the beginning and end of each monitoring section of the power distribution line based on the energy distribution, so as to obtain the distribution difference of each monitoring section.

[0040] The execution module is used to compare all distribution differences with a preset entropy threshold. When the distribution difference of any monitoring segment exceeds the threshold, the monitoring segment is determined to be a fault segment, thereby determining the distance of the fault point within the fault segment.

[0041] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0042] The power distribution line fault location method and system provided in this application first collects transient current traveling wave signals generated at the beginning and end of each monitoring section of the power distribution line when a fault occurs, as well as measurement data of the terminal voltage and current of the doubly-fed induction generator (DFIG). Based on the measurement data, equivalent inductance analysis is performed on the rotor transient of the DFIG during a power distribution line fault to obtain the dominant characteristic frequency band generated by the DFIG during the power distribution line fault. Then, multiple transient signal characteristic quantities of the power distribution line are determined based on the dominant characteristic frequency band and the transient current traveling wave signals. The energy distribution of the transient current of the power distribution line during a fault is determined based on all transient signal characteristic quantities. Then, the difference between the transient energy at the beginning and end of each monitoring section of the power distribution line is evaluated based on the energy distribution to obtain the distribution difference of each monitoring section. All distribution differences are compared with a preset entropy threshold. When the distribution difference of any monitoring section exceeds the threshold, the monitoring section is determined to be a fault section, thereby determining the distance of the fault point within the fault section.

[0043] Therefore, in the fault location method for power distribution lines, this application first collects transient current traveling wave signals generated at the beginning and end of each monitoring section of the power distribution line when a fault occurs, as well as measurement data of the terminal voltage and current of the doubly-fed induction generator (DFIG). Based on the measurement data, equivalent inductance analysis is performed on the rotor transient of the DFIG during a power distribution line fault to obtain the dominant characteristic frequency band generated by the DFIG during the power distribution line fault. Then, based on the dominant characteristic frequency band and the transient current traveling wave signal, multiple transient signal characteristic quantities of the power distribution line are determined. The transient signal characteristic quantity refers to the characteristic value of the energy of the transient current traveling wave within the dominant characteristic frequency band determined by the fault mechanism of the DFIG, which facilitates the analysis of the transient current traveling wave signal. When a doubly fed wind turbine (DFF) fails, a specific transient frequency component determined by rotor motion and crowbar engagement is used as a physical criterion to address the feature extraction inaccuracies caused by neglecting the unique fault characteristics of the wind turbine in weak grid scenarios. Secondly, based on all transient signal characteristics, the energy distribution of the transient current in the distribution line during a fault is determined. Then, based on this energy distribution, the differences in transient energy between the beginning and end of each monitoring section of the distribution line are assessed to obtain the distribution differences of each monitoring section. All distribution differences are compared with a preset entropy threshold. When the distribution difference of any monitoring section exceeds the threshold, the monitoring section is determined to be a fault section, thereby determining the distance to the fault point within the fault section. This scheme can locate faults in distribution lines of weak grids using the coupling mechanism of the DFF transient response characteristics and the mesh parallel loop. Attached Figure Description

[0044] Figure 1 This is an exemplary flowchart of a power distribution line fault location method according to some embodiments of this application;

[0045] Figure 2 This is an exemplary flowchart illustrating the determination of the dominant feature frequency band according to some embodiments of this application;

[0046] Figure 3 This is a schematic diagram of the structure of a wavelet packet decomposition tree according to some embodiments of this application;

[0047] Figure 4 This is a schematic diagram of the structure of a power distribution line fault location system according to some embodiments of this application;

[0048] Figure 5 This is a schematic diagram of the structure of a computer device for implementing a power distribution line fault location method according to some embodiments of this application. Detailed Implementation

[0049] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0050] refer to Figure 1 The figure is an exemplary flowchart of a power distribution line fault location method according to some embodiments of this application. The power distribution line fault location method mainly includes the following steps:

[0051] In step 101, transient current traveling wave signals generated at the beginning and end of each monitoring section of the power distribution line when a fault occurs, as well as measurement data of the terminal voltage and current of the doubly fed wind turbine, are collected.

[0052] In practice, the transient current traveling wave signals generated at the beginning and end of the power distribution line during a fault can be acquired using the following method: Traveling wave acquisition devices (such as feeder terminal units with traveling wave acquisition functions) with high-frequency sampling capabilities are installed at the beginning (e.g., substation outlet) and end (e.g., wind farm grid connection point) of each monitoring area of ​​the power distribution line. Simultaneously, a dedicated electrical quantity measurement unit (such as a PMU) is configured at the wind farm grid connection point or wind turbine outlet. All acquisition devices are equipped with a high-precision synchronous clock source (such as a GPS or BeiDou clock) to ensure that the data acquisition time at the beginning and end is strictly synchronized. The system continuously monitors the changes in zero-sequence voltage and phase voltage of the power distribution line. When the voltage surge amplitude exceeds the set threshold, it is immediately determined that a fault has occurred, and the acquisition devices of all monitoring points are triggered simultaneously to record the transient current traveling wave signal and the voltage and current measurement data of the wind turbine terminal voltage and current within a specific time window before and after the fault occurs. The specific time window is usually set to 1-5 power frequency cycles (i.e., 20-100 milliseconds) before and after the fault occurs. In this application, it is set to 50 milliseconds, and the sampling frequency is 1MHz. Other methods can be used in other embodiments, which are not limited here.

[0053] It should be noted that the terms "starting point" and "ending point" as used in this application refer to each individual monitored line segment. In complex distribution lines containing multiple segments, each segment has its corresponding starting point and ending point measurement points, and the system will collect data and perform subsequent analysis on all segments in parallel. A measurement point may simultaneously be the ending point of the previous segment and the starting point of the next segment.

[0054] In step 102, based on the measurement data, the equivalent inductance analysis of the rotor transient of the doubly fed wind turbine during a power distribution line fault is performed to obtain the dominant characteristic frequency band generated by the doubly fed wind turbine during a power distribution line fault. Then, based on the dominant characteristic frequency band and the transient current traveling wave signal, multiple transient signal characteristic quantities of the power distribution line are determined.

[0055] In some embodiments, reference Figure 2As shown, this figure is an exemplary flowchart for determining the dominant characteristic frequency band in some embodiments of this application. In this embodiment, the equivalent inductance analysis of the rotor transient of the doubly-fed wind turbine during a power distribution line fault is performed based on the measurement data. The dominant characteristic frequency band generated by the doubly-fed wind turbine during a power distribution line fault can be obtained by the following steps:

[0056] First, in step 1021, an equivalent circuit simulation is performed on the mesh parallel circuit containing the doubly fed fan based on the measurement data to obtain the transient flux response of each electrical circuit of the doubly fed fan.

[0057] Secondly, in step 1022, the stator transient DC component generated by the stator flux conservation induction is determined based on all transient flux responses;

[0058] Furthermore, in step 1023, the rotor flux conservation and the rotor transient AC component caused by the crowbar input are determined based on all transient flux responses.

[0059] Finally, in step 1024, waveform spectrum analysis is performed on the stator transient DC component and the rotor transient AC component during a fault to obtain the dominant characteristic frequency band generated by the doubly fed wind turbine when the power distribution line is faulty.

[0060] In specific implementation, the equivalent circuit simulation of the mesh parallel circuit containing the doubly-fed induction generator (DFIG) is performed based on the measurement data. The transient flux response of each electrical circuit of the DFIG can be obtained in the following way: An equivalent circuit model of the DFIG considering grid impedance, crowbar circuit, and mesh parallel structure is obtained. The measurement data and transient current traveling wave signal are input as boundary conditions into the equivalent circuit model. The model, measurement data, and transient current traveling wave signal are loaded using an electromagnetic transient simulation tool (such as Simulink). The voltage-flux differential equations of the DFIG are then numerically solved to obtain the dynamic change data of the stator and rotor flux linkages of the DFIG during a power line fault. The obtained dynamic change data of the stator and rotor flux linkages are used as the transient flux response of the corresponding electrical circuits. Other methods can also be used in other embodiments, which are not limited here.

[0061] It should be noted that the transient flux response in this application refers to the complete response of the flux state quantities inside the stator and rotor circuits of the doubly-fed induction generator (DFIG) as a function of time during the transient process following a power line fault. The equivalent circuit model, by introducing grid impedance and circuit parameters, accurately describes the interaction between the DFIG and the grid under fault conditions. According to the law of electromagnetic induction, the external port characteristics (voltage, current) of the electrical circuit are determined by its internal flux state. Therefore, by solving this equivalent circuit model under the boundary conditions of measurement data and transient current traveling wave signals, the dynamic changes in the flux of the DFIG can be obtained, i.e., the inherent changes caused by the fault characteristics. Furthermore, the dynamic change data of the stator flux and rotor flux are both time-varying rotor magnetic field state data, used to describe the distribution, exchange, and attenuation process of magnetic field energy during the fault.

[0062] In specific implementation, determining the stator transient DC component induced by stator flux conservation based on all transient flux responses can be achieved in the following way: according to the stator flux conservation principle, the stator transient DC component induced during the fault due to the inability of flux to change abruptly is separated from the flux response of the stator circuit by high-pass filtering or feature extraction algorithms (such as empirical mode decomposition), and decays according to a certain time constant. Determining the rotor transient AC component caused by rotor flux conservation and crowbar activation based on all transient flux responses can be achieved in the following way: according to the rotor flux conservation principle and considering the influence of crowbar circuit activation, the rotor transient AC component induced by rotor flux conservation and crowbar action, whose frequency is directly related to the rotor slip frequency during the wind turbine fault, is extracted from the flux response of the rotor circuit using band-pass filtering or frequency analysis techniques. Other methods can also be used in other embodiments, which are not limited here.

[0063] It should be noted that the stator transient DC component in this application refers to the stator current component that exhibits a decaying DC quantity in the synchronous rotating coordinate system (i.e., the dp coordinate system), and the rotor transient AC component refers to the rotor current component that exhibits a decaying AC quantity in the synchronous rotating coordinate system. According to Lenz's law, at the instant of a fault, in order to maintain its flux linkage, the stator winding will induce a magnetic flux linkage that is stationary in space, with a direction opposite to the power frequency flux linkage before the fault. That is, in the coordinate system where the stator is stationary, it exhibits a DC magnetic field with a decaying amplitude. When this DC magnetic field is transformed to a synchronous rotating coordinate system with the power grid for observation... During measurement, it manifests as a DC quantity that decays according to the stator circuit time constant. Secondly, after the crowbar circuit is engaged, the rotor-side converter is short-circuited, and the rotor winding loses external voltage excitation. In order to maintain the conservation of rotor flux at the moment of fault, a transient DC flux that is stationary relative to the rotor itself will be induced in the rotor circuit. This flux manifests as a DC quantity in the rotor coordinate system that rotates with the rotor at the slip frequency. When transformed to a coordinate system that rotates synchronously with the power grid, due to the relative motion between the coordinate systems (i.e., the slip frequency), this DC component manifests as a decaying AC quantity with a frequency equal to the slip frequency.

[0064] In specific implementation, waveform spectrum analysis of the stator transient DC component and the rotor transient AC component during a fault can be performed to obtain the dominant characteristic frequency band generated by the doubly-fed induction generator (DFIG) during a power distribution line fault. This can be achieved in the following way: obtain the rotor slip frequency of the DFIG during the fault; perform Fast Fourier Transform on the time-domain waveforms of the stator transient DC component and the rotor transient AC component respectively to obtain the stator transient spectrum and the rotor transient spectrum; calculate the fundamental frequency of the rotor current using the rotor slip frequency and the power frequency of the grid; and then use the fundamental frequency as the center. The main harmonics (such as the second and third harmonics) at the rotor slip frequency and the main sidebands generated by the power frequency modulation are covered to obtain the characteristic frequency band. The low-frequency region with the most concentrated energy in the stator transient spectrum is overlapped with the above characteristic frequency band, and the two frequency bands are merged. The merged frequency band is taken as the dominant characteristic frequency band generated by the doubly fed fan when the power distribution line is faulty. Conversely, the total energy of the two frequency bands in their respective spectra is calculated, and the frequency band with the highest total energy is selected as the dominant characteristic frequency band. Other methods can be used in other embodiments, which are not limited here.

[0065] It should be noted that the dominant characteristic frequency band in this application is a specific frequency range with concentrated energy, determined by the transient physical mechanism of doubly fed wind turbine faults. This frequency band integrates the transient response characteristics of the stator circuit and the rotor circuit, and is used to accurately select the nodes corresponding to this frequency band from the wavelet packet decomposition tree. This facilitates the extraction of transient signal features that characterize the essence of wind turbine faults and have strong anti-interference capabilities from complex transient traveling wave signals, thereby providing effective input features for the subsequent accurate location of fault sections.

[0066] In some embodiments, determining multiple transient signal characteristic quantities of the power distribution line based on the dominant characteristic frequency band and the transient current traveling wave signal can be achieved by the following steps:

[0067] The transient current traveling wave signal is decomposed into a multi-scale decomposition to obtain a wavelet packet decomposition tree;

[0068] Based on the wavelet packet decomposition tree, multiple energy nodes of the dominant feature frequency band are determined;

[0069] Multiple transient signal characteristics of the power distribution line are extracted from all energy nodes using the dominant characteristic frequency band.

[0070] In specific implementation, multi-scale decomposition of the transient current traveling wave signal to obtain a wavelet packet decomposition tree can be achieved in the following way: multi-scale decomposition of the transient current traveling wave signal is performed through wavelet packet transform to obtain a wavelet packet decomposition tree covering the entire frequency band. (Refer to...) Figure 3 As shown in the figure, this is a schematic diagram of the wavelet packet decomposition tree structure in some embodiments of this application. The determination of multiple energy nodes in the dominant characteristic frequency band based on the wavelet packet decomposition tree can be achieved in the following way: all wavelet packet nodes whose frequency range falls within the dominant characteristic frequency band are located and selected in the wavelet packet decomposition tree, and all obtained wavelet packet nodes are used as multiple energy nodes in the dominant characteristic frequency band. The extraction of multiple transient signal features of the power distribution line from all energy nodes through the dominant characteristic frequency band can be achieved in the following way: the signal energy of each energy node is calculated using the energy formula, and the set of all obtained signal energies is used as multiple transient signal features of the power distribution line. Other methods can also be used in other embodiments, which are not limited here.

[0071] It should be noted that the transient signal characteristic quantity in this application refers to the characteristic value of the energy of the transient current traveling wave within the dominant characteristic frequency band determined by the fault mechanism of the doubly fed wind turbine. It is used to associate the physical mechanism represented by the dominant characteristic frequency band with the signal processing represented by the wavelet packet energy, thereby constructing a digital feature that can accurately characterize the fault state. The specific transient frequency component (i.e., the dominant characteristic frequency band) determined by the rotor motion and the crowbar insertion during the fault of the doubly fed wind turbine is used as the physical criterion. This overcomes the problem of inaccurate feature extraction caused by ignoring the unique fault characteristics of the wind turbine in the weak grid scenario of traditional methods, thus facilitating the subsequent location of the fault area.

[0072] In step 103, the energy distribution of the transient current of the power distribution line during a fault is determined based on all transient signal characteristics. Then, the difference in transient energy between the beginning and end of each monitoring section of the power distribution line is evaluated based on the energy distribution to obtain the distribution difference of each monitoring section.

[0073] In some embodiments, determining the energy distribution of the transient current in a power distribution line during a fault, based on all transient signal characteristics, can be achieved using the following steps:

[0074] The energy set at the beginning and the energy set at the end of each monitoring section of the power distribution line are determined based on all transient signal characteristics.

[0075] The probability distribution of multiple head-end energies of the transient current is determined based on all head-end energy sets.

[0076] The probability distribution of multiple end energies of the transient current is determined based on all the end energy sets;

[0077] The set of all the energy probability distributions at the beginning and all the energy probability distributions at the end is taken as the energy distribution of the transient current of the distribution line when the distribution line is faulty.

[0078] In specific implementation, the energy set at the beginning and the energy set at the end of each monitoring section of the power distribution line can be determined based on all transient signal characteristics in the following manner: the set of signal energies of all energy nodes corresponding to the measurement points at the beginning of each monitoring section of the power distribution line in the transient signal characteristics is taken as the energy set at the beginning of the corresponding monitoring section, and the set of signal energies of all energy nodes corresponding to the measurement points at the end of each monitoring section of the power distribution line in the transient signal characteristics is taken as the energy set at the end of the corresponding monitoring section. Other methods can also be used in other embodiments, which are not limited here.

[0079] In specific implementation, determining multiple head-end energy probability distributions of transient current based on all head-end energy sets can be achieved in the following way: calculate the sum of signal energies of all energy nodes in each head-end energy set; then, divide each signal energy in each head-end energy set by the corresponding sum of head-end energies to obtain the head-end energy probability distribution of transient current in each monitoring section. Similarly, determining multiple end-end energy probability distributions of transient current based on all end-end energy sets can be achieved in the following way: calculate the sum of signal energies of all energy nodes in each end-end energy set; then, divide each signal energy in each end-end energy set by the corresponding sum of end-end energies to obtain the end-end energy probability distribution of transient current in each monitoring section. Here, the head-end energy probability distribution describes the distribution characteristics of fault transient energy in the head-end signal frequency domain structure when a power distribution line fault occurs, and the end-end energy probability distribution describes the distribution characteristics of fault transient energy in the end-end signal frequency domain structure when a power distribution line fault occurs. Other methods can also be used in other embodiments, which are not limited here.

[0080] It should be noted that for each monitoring segment, the first-end energy probability distribution is a probability vector, where each component represents the proportion of the energy of a specific wavelet packet node at the first-end measurement point of a single monitoring segment within the dominant characteristic frequency band to the total energy of that measurement point. The last-end energy probability distribution is another probability vector, where each component represents the proportion of the energy of a specific wavelet packet node at the last-end measurement point of a single monitoring segment within the dominant characteristic frequency band to the total energy of that measurement point. These components are used to quantitatively describe the distribution of fault transients at the first and last ends, facilitating subsequent determination of distribution differences.

[0081] In some embodiments, the difference in transient energy between the beginning and end of each monitoring section of the power distribution line is evaluated based on the energy distribution, and the distribution difference of each monitoring section can be obtained by the following steps:

[0082] Obtain the smoothing factor;

[0083] Select a monitoring section as the selected monitoring section;

[0084] Based on the smoothing factor, the relative entropy of transient energy at the beginning and end of the selected monitoring section is determined for the energy distribution;

[0085] The relative entropy is used as the distribution difference when there is a fault at the beginning or end of the selected monitoring section;

[0086] Continue to determine the distribution differences in the remaining monitoring sections.

[0087] In specific implementation, the relative entropy of the transient energy at the beginning and end of the selected monitoring segment based on the smoothing factor can be achieved in the following way: the probability distribution of the energy at the end and the probability distribution of the energy at the beginning of the selected monitoring segment in the energy distribution is normalized by the smoothing factor to ensure the stability of the numerical calculation. Secondly, the relative entropy of the probability distribution of the energy at the end and the probability distribution of the energy at the beginning of the selected monitoring segment is calculated by the formula of Kullback-Leibler divergence to obtain the relative entropy of the transient energy at the beginning and the end of the selected monitoring segment in the mesh parallel loop. Other methods can also be used in other embodiments, which are not limited here.

[0088] It should be noted that the distribution difference in this application refers to the degree of dissimilarity between the transient energy probability distributions at the beginning and end of each monitoring section when a power distribution line fault occurs. The relative entropy of transient energy refers to the information theory index that quantifies the difference between the energy probability distribution at the end and the energy probability distribution at the beginning. When a power distribution line fault occurs, the transient current traveling wave signals upstream and downstream of the fault point will exhibit different frequency domain energy distribution characteristics. By converting the transient signal energy at the beginning and end into probability distributions and calculating their relative entropy, this distribution difference can be accurately captured from the perspective of information theory. That is, the larger the relative entropy value, the lower the similarity of the energy distributions of the two measuring points, thus providing a reliable quantitative basis for fault section determination. This facilitates the limitation of traditional fault location methods relying on signal characteristics that are easily interfered with in scenarios with weak power grids and high proportions of new energy sources (such as wind farms).

[0089] In step 104, all distribution differences are compared with a preset entropy threshold. When the distribution difference of any monitoring segment exceeds the threshold, the monitoring segment is determined to be a fault segment, thereby determining the distance of the fault point within the fault segment.

[0090] It should be noted that all distribution differences are compared with a preset entropy threshold. If the distribution difference of any current monitoring segment does not exceed the threshold, the monitoring segment will continue to be monitored.

[0091] In some embodiments, determining the distance to the fault point within the fault zone can be achieved using the following steps:

[0092] Obtain the absolute timestamps of the arrival of the initial traveling waves at the beginning and end;

[0093] Based on the absolute timestamp, the faulty section is located using a double-ended traveling wave method to obtain the distance to the fault point.

[0094] In specific implementation, the distance to the fault point can be obtained by performing double-ended traveling wave positioning on the fault section based on the absolute timestamp, which can be achieved in the following way: obtain the absolute timestamps of the initial fault current traveling wave arriving at the beginning and end of the monitored section line recorded by a high-precision synchronous clock (such as GPS or Beidou) from the system, and calculate the fault distance of the fault area by combining the core formula of the double-ended traveling wave method with the absolute timestamps; other methods can also be used in other embodiments, which are not limited here.

[0095] It should be noted that the distribution difference (i.e., relative entropy) calculated for each monitoring section is compared in parallel with a preset entropy threshold. This threshold can be determined through statistical optimization after simulation and historical data analysis of a large number of power distribution networks with doubly fed wind turbines covering different fault locations, transition resistances, and noise levels. When the distribution difference value of any monitoring section exceeds this threshold, it confirms from an information theory perspective that there is a statistically significant difference in the transient energy distribution at the beginning and end of the monitoring section, thus determining that the section is a fault section. This transforms the complex fault location problem into a reliable pattern recognition problem, which helps to reduce the difficulty of misjudgment caused by interference from traditional single electrical quantity criteria in the context of complex fault current characteristics.

[0096] In another aspect, in some embodiments, this application provides a power distribution line fault location system, with reference to... Figure 4 The figure is a schematic diagram of the structure of a power distribution line fault location system according to some embodiments of this application. The power distribution line fault location system includes: a data acquisition module 401, a processing module 402, and an execution module 403, which are described below:

[0097] The acquisition module 401 in this application is mainly used to acquire transient current traveling wave signals generated at the beginning and end of each monitoring section of the power distribution line when a fault occurs, as well as the measurement data of the terminal voltage and current of the doubly fed wind turbine.

[0098] Processing module 402, in this application, is used to perform equivalent inductance analysis on the rotor transient of the doubly fed wind turbine when the power distribution line is faulted based on the measurement data, to obtain the dominant characteristic frequency band generated by the doubly fed wind turbine when the power distribution line is faulted, and then to determine multiple transient signal characteristic quantities of the power distribution line based on the dominant characteristic frequency band and the transient current traveling wave signal.

[0099] It should be noted that the processing module 402 in this application is also used to determine the energy distribution of the transient current of the power distribution line when the power distribution line is faulty based on all transient signal characteristics, and then to evaluate the difference in transient energy between the beginning and end of each monitoring section of the power distribution line based on the energy distribution, so as to obtain the distribution difference of each monitoring section.

[0100] The execution module 403 in this application is mainly used to compare all distribution differences with a preset entropy threshold. When the distribution difference of any monitoring segment exceeds the threshold, the monitoring segment is determined to be a fault segment, thereby determining the distance of the fault point within the fault segment.

[0101] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described power distribution line fault location method.

[0102] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing a power distribution line fault location method according to some embodiments of this application. The power distribution line fault location method in the above embodiments can... Figure 5 The computer device shown is used to implement this, and the computer device includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0103] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0104] The communication bus 502 can be used to transmit information between the aforementioned components.

[0105] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0106] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0107] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0108] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0109] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0110] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for locating faults in power distribution lines.

[0111] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0112] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method of locating a fault on a power distribution line, wherein, The power distribution line is a meshed and parallel circuit containing a doubly-fed wind turbine, and the power distribution line is divided into multiple monitoring sections in advance, and the method comprises the following steps: Collecting transient current traveling wave signals generated at the first end and the last end of each monitoring section of the power distribution line when a fault occurs, and collecting measurement data of the terminal voltage and current of the doubly-fed wind turbine; Based on the measurement data, the rotor transient of the doubly-fed wind turbine when the power distribution line fails is analyzed by equivalent inductance, the dominant characteristic frequency band generated by the doubly-fed wind turbine when the power distribution line fails is obtained, and then multiple transient signal characteristic quantities of the power distribution line are determined according to the dominant characteristic frequency band and the transient current traveling wave signal; According to all the transient signal characteristic quantities, the energy distribution of the transient current of the power distribution line when the power distribution line fails is determined, and then the transient energy at the first end and the last end of each monitoring section of the power distribution line is evaluated according to the energy distribution, and the distribution difference of each monitoring section is obtained; All the distribution differences are compared with a preset entropy threshold value, and when the distribution difference of any monitoring section exceeds the threshold value, the monitoring section is determined as a fault section, and the distance of the fault point in the fault section is determined; Wherein, based on the measurement data, the rotor transient of the doubly-fed wind turbine when the power distribution line fails is analyzed by equivalent inductance, and the dominant characteristic frequency band generated by the doubly-fed wind turbine when the power distribution line fails is obtained, and the method comprises the following steps: According to the measurement data, the meshed and parallel circuit containing the doubly-fed wind turbine is simulated by equivalent circuit, and the transient flux linkage response of each electrical circuit of the doubly-fed wind turbine is obtained; Based on all the transient flux linkage responses, the stator transient direct current component generated by the stator flux linkage conservation induction is determined; According to all the transient flux linkage responses, the rotor transient alternating current component caused by the rotor flux linkage conservation and crowbar injection is determined; The stator transient direct current component and the rotor transient alternating current component are analyzed by waveform spectrum when the fault occurs, and the dominant characteristic frequency band generated by the doubly-fed wind turbine when the power distribution line fails is obtained; Wherein, according to the dominant characteristic frequency band and the transient current traveling wave signal, multiple transient signal characteristic quantities of the power distribution line are determined, and the method comprises the following steps: The transient current traveling wave signal is decomposed by multi-scale, and a wavelet packet decomposition tree is obtained; Based on the wavelet packet decomposition tree, multiple energy nodes of the dominant characteristic frequency band are determined; The multiple transient signal characteristic quantities of the power distribution line are extracted from all the energy nodes through the dominant characteristic frequency band; Wherein, according to all the transient signal characteristic quantities, the energy distribution of the transient current of the power distribution line when the power distribution line fails is determined, and the method comprises the following steps: According to all the transient signal characteristic quantities, the first end energy set and the last end energy set of each monitoring section of the power distribution line are determined; According to all the first end energy sets, multiple first end energy probability distributions of the transient current are determined; According to all the last end energy sets, multiple last end energy probability distributions of the transient current are determined; The set composed of all the first end energy probability distributions and all the last end energy probability distributions is taken as the energy distribution of the transient current of the power distribution line when the power distribution line fails. The distribution difference of each monitoring section is obtained by differentiating transient energy of the start end and the end of each monitoring section of the power distribution line according to the energy distribution, and the distribution difference of each monitoring section specifically includes: obtaining a smoothing factor; selecting a monitoring section as a selected monitoring section; determining the relative entropy of the transient energy of the start end and the end of the selected monitoring section based on the energy distribution and the smoothing factor; taking the relative entropy as the distribution difference when the start end and the end of the selected monitoring section fail; continuing to determine the distribution difference of the remaining monitoring sections; wherein the distance of the fault point in the fault section specifically includes: obtaining absolute time stamps of the start end and the end initial traveling wave; performing double-end traveling wave positioning on the fault section based on the absolute time stamps to obtain the distance of the fault point.

2. The method of claim 1, wherein, The transient current traveling wave signals generated by the start end and the end of the power distribution line when the fault occurs are collected by the traveling wave collection device.

3. The method of claim 1, wherein, The measurement data of the generator terminal voltage and current of the doubly-fed wind turbine are collected by the electrical quantity measurement unit.

4. The method of claim 1, wherein, All the distribution differences are compared with a preset entropy threshold value, and if the distribution difference of any monitoring section does not exceed the threshold value, the monitoring section is continuously monitored.

5. A power distribution line fault location system employing the method of any one of claims 1 to 4 for power distribution line fault location, characterized by, The power distribution line fault positioning system includes: a collection module for collecting transient current traveling wave signals generated by the start end and the end of each monitoring section of the power distribution line when the fault occurs and measurement data of the generator terminal voltage and current of the doubly-fed wind turbine; a processing module for performing equivalent inductance analysis on the rotor transient of the doubly-fed wind turbine when the power distribution line fails based on the measurement data to obtain a dominant characteristic frequency band generated by the doubly-fed wind turbine when the power distribution line fails, and then determining a plurality of transient signal characteristic quantities of the power distribution line according to the dominant characteristic frequency band and the transient current traveling wave signals; the processing module is further configured to determine the energy distribution of the transient current of the power distribution line when the power distribution line fails according to all the transient signal characteristic quantities, and then differentiate the transient energy of the start end and the end of each monitoring section of the power distribution line according to the energy distribution to obtain the distribution difference of each monitoring section; an execution module for comparing all the distribution differences with a preset entropy threshold value, and determining that the monitoring section is a fault section when the distribution difference of any monitoring section exceeds the threshold value, so as to determine the distance of the fault point in the fault section.

Images