Multi-branch high-voltage hybrid line fault voltage traveling wave head inversion method and system

Abstract

The invention discloses a multi-branch high-voltage hybrid line fault voltage traveling wave head inversion method and system, and belongs to the technical field of power grid fault positioning. The method comprises the steps of building an overhead-cable mixed line simulation model, an overhead line voltage traveling wave sensor model and a cable line voltage traveling wave sensor model; the method comprises the following steps: setting a fault, measuring a secondary waveform and a primary traveling waveform, constructing primary and secondary voltage traveling wave signal data sets, performing phase-mode conversion on primary and secondary voltage traveling wave signals to convert the primary and secondary voltage traveling wave signals into a line mode voltage and a zero mode voltage, respectively constructing voltage traveling wave feature sets according to different scenes, and inputting the voltage traveling wave feature sets into an LSTM model for training to obtain a trained waveform inversion model; and inputting an actually measured fault voltage traveling wave waveform, and outputting an inverted primary traveling wave waveform. According to the method, the LSTM model is used, a sensor transfer function does not need to be analyzed, two different scenes of fault distance measurement and fault section identification are considered, characteristic quantities of different frequency band components are extracted, and waveform inversion can be more accurately realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid fault location, and in particular relates to a method and system for inverting a traveling wave head of a fault voltage of a multi-branch high-voltage hybrid line. Background Art

[0002] Accurately obtaining the arrival time of the traveling wave of the power grid fault is of great significance to improving the accuracy of the traveling wave positioning technology. Due to the limitations of the existing measurement technology level, it is impossible to directly obtain the primary traveling wave signal of the power grid side fault. The traveling wave positioning can only be performed by measuring the secondary signal after the transmission through sensors and other devices. The line fault current traveling wave detection device currently used in the field uses the traveling wave head information on the secondary side of the voltage sensor or current sensor to locate the fault. However, due to the insufficient physical properties of the sensor itself and the different transmission characteristics of the traveling wave signals in each characteristic frequency band of the signal acquisition module, the waveform of the primary traveling wave signal of the power grid will be distorted to a certain extent after the transmission through the sensor, detection module, etc., resulting in the inconsistency between the secondary waveform and the primary waveform characteristics. In addition, the refraction and reflection of the multi-branch high-voltage hybrid line is more complicated, making it difficult to accurately identify the wave head, which may lead to large positioning errors.

[0003] At present, there are two main solutions to the problem of waveform distortion after the traveling wave signal is transmitted through the sensor: 1) Study the internal structure of the sensor to reduce the degree of distortion. For example, there is a waveform distortion correction method based on Rogowski coil measurement in the prior art;

[0004] 2) Use the secondary side traveling wave signal to reversely deduce the primary side signal to obtain the real primary side signal. At present, the second method is mostly used to obtain the actual fault transient signal. This reverse derivation process is a typical inversion problem solving process. Existing technologies for signal inversion problems include:

[0005] The prior art has a voltage traveling wave detection method combined with Tikhonov regularization. On the premise of modeling the traveling wave sensor, a sensor inverse model is established based on the transfer function of the traveling wave sensor model, and the Tikhonov regularization method is used to solve the ill-posed problems that occur during the inversion problem solving process. However, the method based on regularization inversion is greatly affected by the regularization parameter.

[0006] There is a voltage traveling wave detection method based on L1 regularization inversion in the prior art, but the method based on regularization inversion is greatly affected by the regularization parameter. There is a voltage traveling wave detection method based on EMD decomposition combined with Wiener filtering in the prior art, which uses empirical mode decomposition and Wiener filtering to invert the traveling wave signal, but there is a mode aliasing effect in the signal decomposition process, which affects the inversion accuracy;

[0007] The prior art includes the analysis and optimization design of the main insulation electric field of the oil-immersed inverted current transformer, and the improved method of current transformer saturation identification based on instantaneous current characteristics. Both of them aim at the distortion problem of the secondary signal after the current sensor is saturated, and use the neural network algorithm to correct the distorted waveform. However, when correcting the distortion problem of the saturated signal, the selection of the sensor model has a great influence on the compensation effect, and the compensation of the secondary signal cannot reproduce the primary signal well.

[0008] The existing technologies include phase difference analysis and correction of Rogowski coil current transformers, and Rogowski coil low-frequency distortion correction technology for lightning current measurement of transmission lines. Both of them are based on waveform distortion correction methods and require the calculation of the sensor's propagation function, but this parameter is usually greatly affected by the sensor's internal structure and external environment.

[0009] The prior art has a fault waveform inversion method for transient attenuation feature reconstruction, which extracts fault features within a selected time window of the field measured fault signal and randomly divides them into training samples and test samples, constructs a multi-dimensional fault feature matrix using the fault features extracted from the training samples, obtains a fault waveform inversion model through convolutional neural network training, and then obtains a new fault waveform through adaptive adjustment of the fault feature set parameters, and then uses the test sample to test the accuracy of the newly generated waveform, thereby conveniently and comprehensively conducting functional tests on the primary and secondary fusion intelligent switch for fault handling. However, this method uses limited field measured distribution network fault signals as samples, and its main purpose is to expand the samples of the waveform inversion test system, and does not focus on the transmission characteristics of the primary and secondary traveling waves. Summary of the invention

[0010] In order to solve the deficiencies in the prior art, the present invention provides a method and system for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line to solve the problem of difficulty in identifying the traveling wave head.

[0011] The present invention adopts the following technical solution.

[0012] The first aspect of the present invention provides a method for inverting a traveling wave head of a multi-branch high-voltage hybrid line fault voltage, comprising:

[0013] Step 1: Based on the multi-branch high-voltage hybrid line power grid topology, build and obtain the overhead-cable hybrid line simulation model, the overhead line voltage traveling wave sensor model, and the cable line voltage traveling wave sensor model;

[0014] Step 2: Set a fault on the overhead-cable hybrid line simulation model obtained in step 1, measure the secondary voltage traveling wave waveforms output by the overhead line voltage traveling wave sensor model and the cable voltage traveling wave sensor model obtained in step 1 and the primary voltage traveling wave waveforms at their installation locations, construct a primary voltage traveling wave signal data set based on the measured primary voltage traveling wave waveform, and construct a secondary voltage traveling wave signal data set based on the measured secondary voltage traveling wave waveform;

[0015] Step 3: Perform phase mode transformation on the primary voltage traveling wave signal and the secondary voltage traveling wave signal in the primary voltage traveling wave signal data set and the secondary voltage traveling wave signal data set constructed in step 2, and construct a voltage traveling wave feature set according to different scenarios and the data after phase mode transformation;

[0016] Step 4: Use the voltage traveling wave feature set constructed in step 3 to train the LSTM model to obtain a trained waveform inversion model;

[0017] Step 5: Input the measured fault voltage traveling wave waveform into the trained waveform inversion model obtained in step 4 to output an inverted primary traveling wave waveform, thereby realizing the inversion of the multi-branch high-voltage hybrid line fault voltage traveling wave head.

[0018] Preferably, in step 1, building an overhead line voltage traveling wave sensor model and a cable line voltage traveling wave sensor model includes:

[0019] An electric field induction voltage transformer is installed on the overhead line tower. According to the spatial position of the electric field induction voltage transformer relative to the conductor and the physical relationship between the electric field strength and the voltage, a traveling wave sensor model of the overhead line voltage is constructed.

[0020] A capacitive voltage divider transformer is installed at the end of the cable, and a cable line voltage traveling wave sensor model is built based on the capacitive voltage divider transformer.

[0021] Preferably, in step 2, setting faults on the overhead-cable hybrid line simulation model obtained in step 1 comprises: randomly setting multiple groups of faults at the junctions of overhead lines, cables and overhead cables of the multi-branch high-voltage overhead-cable hybrid line simulation model, each group of faults comprising a fault location, a fault resistance and a fault initial phase angle;

[0022] The fault resistance is randomly selected from 0Ω to 3kΩ, and the fault initial phase angle is randomly selected from 0° to 90°.

[0023] Preferably, in step 3, the collected primary voltage traveling wave signal and secondary voltage traveling wave signal are subjected to phase mode transformation, including: obtaining a primary line mode voltage traveling wave and a primary zero mode voltage traveling wave after phase mode transformation of the primary voltage traveling wave signal, and obtaining a secondary line mode voltage traveling wave and a secondary zero mode voltage traveling wave after phase mode transformation of the secondary voltage traveling wave signal.

[0024] Preferably, a voltage traveling wave feature set is constructed according to different scenarios and data after phase mode transformation, including different scenarios including fault location scenario and fault section identification scenario; the amplitude and phase information of the primary line mode voltage traveling wave and the amplitude and phase information of the secondary line mode voltage traveling wave are extracted to construct a voltage traveling wave feature set in the fault location scenario;

[0025] The amplitude and phase information of the primary zero-mode voltage traveling wave and the amplitude and phase information of the secondary zero-mode voltage traveling wave are extracted to construct a voltage traveling wave feature set for the fault section identification scenario.

[0026] Preferably, the amplitude and phase information in the voltage traveling wave feature set in the fault location scenario and the voltage traveling wave feature set in the fault section identification scenario include phase, high-mid frequency amplitude ratio, high-low frequency amplitude ratio, high-mid frequency phase difference and high-low frequency phase difference.

[0027] Preferably, the high-intermediate frequency amplitude ratio and the high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault location scenario are respectively the amplitude ratio and the phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and the component of 10kHz to 100kHz;

[0028] The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault location scenario are the amplitude ratio and phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively;

[0029] Among them, 100kHz~200kHz is used as the high-frequency component during fault distance measurement, 10kHz~100kHz is used as the medium-frequency component during fault distance measurement, and 20Hz~10kHz is used as the low-frequency component during fault distance measurement.

[0030] Preferably, the high-intermediate frequency amplitude ratio and the high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are respectively the amplitude ratio and the phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and the component of 10kHz to 100kHz;

[0031] The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are the amplitude ratio and phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively;

[0032] Among them, 100kHz~200kHz is used as the high-frequency component when identifying the fault section, 10kHz~100kHz is used as the medium-frequency component when identifying the fault section, and 20Hz~10kHz is used as the low-frequency component when identifying the fault section.

[0033] Preferably, in step 4, the amplitude and phase information of the secondary line mode voltage traveling wave feature in the voltage traveling wave feature set in the fault location scenario is used as the input set, and the amplitude and phase information of the primary line mode voltage traveling wave in the voltage traveling wave feature set during fault location is used as the label set, and the voltage traveling wave feature set in the fault location scenario is divided into a training set, a verification set and a test set;

[0034] The amplitude and phase information of the secondary zero-mode voltage traveling wave in the voltage traveling wave feature set in the fault section identification scenario are set as the input set, and the amplitude and phase information of the primary zero-mode voltage traveling wave feature in the voltage traveling wave feature set during fault section identification are set as the label set. The voltage traveling wave feature set in the fault section identification scenario is divided into a training set, a validation set and a test set.

[0035] The second aspect of the present invention provides a multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system, which is used to run the multi-branch high-voltage hybrid line fault voltage traveling wave head inversion method described in the first aspect, comprising:

[0036] Hybrid line construction module: used to build and obtain the overhead-cable hybrid line simulation model, overhead line voltage traveling wave sensor model and cable line voltage traveling wave sensor model based on the multi-branch high-voltage hybrid line power grid topology;

[0037] Data acquisition module: used to set faults on the overhead-cable hybrid line simulation model, measure the secondary voltage traveling wave waveforms output by the overhead line voltage traveling wave sensor model and the cable voltage traveling wave sensor model and the primary voltage traveling wave waveforms at their installation locations, construct a primary voltage traveling wave signal data set based on the measured primary voltage traveling wave waveforms, and construct a secondary voltage traveling wave signal data set based on the measured secondary voltage traveling wave waveforms;

[0038] Feature extraction module: used to perform phase mode transformation on the primary voltage traveling wave signal and the secondary voltage traveling wave signal in the constructed primary voltage traveling wave signal data set and the secondary voltage traveling wave signal data set, and construct a voltage traveling wave feature set according to different scenarios and the data after phase mode transformation;

[0039] Model training module: used to train the LSTM model using the constructed voltage traveling wave feature set to obtain a trained waveform inversion model;

[0040] Wave head inversion module: It is used to input the measured fault voltage traveling wave waveform into the trained waveform inversion model to output the inverted primary traveling wave waveform, so as to realize the traveling wave head inversion of the fault voltage of multi-branch high-voltage hybrid line.

[0041] Preferably, building an overhead line voltage traveling wave sensor model and a cable line voltage traveling wave sensor model includes:

[0042] An electric field induction voltage transformer is installed on the overhead line tower. According to the spatial position of the electric field induction voltage transformer relative to the conductor and the physical relationship between the electric field strength and the voltage, a traveling wave sensor model of the overhead line voltage is constructed.

[0043] A capacitive voltage divider transformer is installed at the end of the cable, and a cable line voltage traveling wave sensor model is built based on the capacitive voltage divider transformer.

[0044] Preferably, setting faults on the obtained overhead-cable hybrid line simulation model comprises: randomly setting multiple groups of faults at the junctions of overhead lines, cables and overhead cables of the multi-branch high-voltage overhead-cable hybrid line simulation model, each group of faults comprising a fault location, a fault resistance and a fault initial phase angle;

[0045] The fault resistance is randomly selected from 0Ω to 3kΩ, and the fault initial phase angle is randomly selected from 0° to 90°.

[0046] Preferably, the collected primary voltage traveling wave signal and secondary voltage traveling wave signal are subjected to phase mode transformation, including: obtaining a primary line mode voltage traveling wave and a primary zero mode voltage traveling wave after phase mode transformation of the primary voltage traveling wave signal, and obtaining a secondary line mode voltage traveling wave and a secondary zero mode voltage traveling wave after phase mode transformation of the secondary voltage traveling wave signal.

[0047] Preferably, a voltage traveling wave feature set is constructed according to different scenarios and data after phase mode transformation, including different scenarios including fault location scenario and fault section identification scenario; the amplitude and phase information of the primary line mode voltage traveling wave and the amplitude and phase information of the secondary line mode voltage traveling wave are extracted to construct a voltage traveling wave feature set in the fault location scenario;

[0048] The amplitude and phase information of the primary zero-mode voltage traveling wave and the amplitude and phase information of the secondary zero-mode voltage traveling wave are extracted to construct a voltage traveling wave feature set for the fault section identification scenario.

[0049] Preferably, the amplitude and phase information in the voltage traveling wave feature set in the fault location scenario and the voltage traveling wave feature set in the fault section identification scenario include phase, high-mid frequency amplitude ratio, high-low frequency amplitude ratio, high-mid frequency phase difference and high-low frequency phase difference.

[0050] Preferably, the high-intermediate frequency amplitude ratio and the high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault location scenario are respectively the amplitude ratio and the phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and the component of 10kHz to 100kHz;

[0051] The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault location scenario are the amplitude ratio and phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively;

[0052] Among them, 100kHz~200kHz is used as the high-frequency component during fault distance measurement, 10kHz~100kHz is used as the medium-frequency component during fault distance measurement, and 20Hz~10kHz is used as the low-frequency component during fault distance measurement.

[0053] Preferably, the high-intermediate frequency amplitude ratio and the high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are respectively the amplitude ratio and the phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and the component of 10kHz to 100kHz;

[0054] The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are the amplitude ratio and phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively;

[0055] Among them, 100kHz~200kHz is used as the high-frequency component when identifying the fault section, 10kHz~100kHz is used as the medium-frequency component when identifying the fault section, and 20Hz~10kHz is used as the low-frequency component when identifying the fault section.

[0056] Preferably, the amplitude and phase information of the secondary line mode voltage traveling wave characteristics in the voltage traveling wave characteristic set in the fault location scenario are used as the input set, and the amplitude and phase information of the primary line mode voltage traveling wave in the voltage traveling wave characteristic set during fault location are used as the label set, and the voltage traveling wave characteristic set in the fault location scenario is divided into a training set, a verification set and a test set;

[0057] The amplitude and phase information of the secondary zero-mode voltage traveling wave in the voltage traveling wave feature set in the fault section identification scenario are set as the input set, and the amplitude and phase information of the primary zero-mode voltage traveling wave feature in the voltage traveling wave feature set during fault section identification are set as the label set. The voltage traveling wave feature set in the fault section identification scenario is divided into a training set, a validation set and a test set.

[0058] The third aspect of the present invention proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when loaded into the processor, implements a method for inverting the traveling wave head of a multi-branch high-voltage hybrid line fault voltage as described in the first aspect.

[0059] The fourth aspect of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method for inverting the traveling wave head of a multi-branch high-voltage hybrid line fault voltage as described in the first aspect.

[0060] The beneficial effects of the present invention are as follows:

[0061] (1) The present invention does not need to analyze the sensor transfer function, which can avoid the influence of the inaccurate transfer function on the inversion result.

[0062] (2) A simulation model is built based on the actual multi-branch high-voltage hybrid line power grid topology and line parameters, and different fault locations, different fault resistances and different fault initial phase angles are randomly set, which is closer to the actual situation and makes it easier to obtain sample data.

[0063] (3) The overhead lines and cables of multi-branch high-voltage hybrid lines are considered separately. The electric field induction voltage transformer is used to measure the overhead line voltage traveling wave, and the capacitor voltage transformer is used to measure the cable voltage traveling wave. Two voltage traveling wave measurement methods are considered, which are more suitable for the characteristics of multi-branch high-voltage hybrid lines.

[0064] (4) By training multiple groups of data, the present invention makes the LSTM model more reliable and the output inversion of a single traveling wave more accurate.

[0065] (5) The present invention takes into account the fault distance measurement and fault section identification scenarios, and uses line mode and zero mode primary and secondary voltages for waveform inversion, respectively, which can provide more comprehensive data support for the fault distance measurement method based on line mode and zero mode voltage traveling waves.

[0066] (6) The present invention extracts four characteristic quantities of amplitude, phase, amplitude ratio and phase difference of line mode and zero mode voltage components in different frequency bands, which can realize waveform inversion more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a flow chart of the method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line in the present invention. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0069] Embodiment 1 of the present invention provides a method for inverting the traveling wave head of a multi-branch high-voltage hybrid line fault voltage. Figure 1 As shown, the following steps are included:

[0070] Step 1: Based on the hybrid line power grid topology, use PSCAD to build and obtain the overhead-cable hybrid line simulation model, overhead line voltage traveling wave sensor model, and cable line voltage traveling wave sensor model;

[0071] The hybrid line grid topology includes actual hybrid line grid topology and line parameters. According to the actual hybrid line grid topology and line parameters, an overhead-cable hybrid line simulation model is built using PSCAD.

[0072] Line parameters include line model, length, and spacing.

[0073] The overhead lines and cables of the hybrid lines are considered separately, and an overhead line voltage sensor model based on the electric field sensing principle is built. An electric field induction voltage transformer is installed on the overhead line tower. According to the spatial position of the electric field induction voltage transformer relative to the conductor, the physical relationship between the electric field strength and the voltage is used to build an overhead line voltage sensor model to measure the overhead line voltage traveling wave.

[0074] The installation of the electric field induction voltage transformer does not require direct contact with the measuring wire. Choosing an electric field induction voltage transformer to measure voltage in overhead bare wires with high voltage and high current can avoid the safety risks of contact with high voltage lines.

[0075] A capacitive voltage-dividing cable line voltage sensor model was built, a capacitive voltage-dividing voltage transformer was installed at the cable end, and the cable voltage traveling wave was measured based on the capacitive coupling principle.

[0076] The capacitive voltage divider voltage transformer measures voltage based on the capacitive voltage divider principle. Its ring structure is convenient for clipping onto the line with insulation layer, and is more suitable for use in cable lines.

[0077] Step 2: Set a fault on the overhead-cable hybrid line simulation model obtained in step 1, measure the secondary voltage traveling wave waveforms output by the overhead line voltage traveling wave sensor model and the cable voltage traveling wave sensor model obtained in step 1, and the primary voltage traveling wave waveforms at their installation locations, construct a primary voltage traveling wave signal dataset based on the measured primary voltage traveling wave waveform, and construct a secondary voltage traveling wave signal dataset based on the measured secondary voltage traveling wave waveform.

[0078] Voltage traveling wave measurement points are arranged at the beginning and end of each branch line, and 1000 groups of faults with different fault locations, different fault resistances and different fault initial phase angles are randomly set. Each group of faults includes the fault location, fault resistance and fault initial phase angle. The primary voltage traveling wave waveform generated after the operation of the overhead-cable hybrid line simulation model is obtained, and the secondary voltage traveling wave waveform output by the voltage sensor at the installation location of the overhead line and cable voltage traveling wave sensor is measured to construct the primary voltage traveling wave data set and the secondary voltage traveling wave data set.

[0079] Faults are randomly set at the junction of overhead lines, cables and overhead cables in the simulation model. The fault resistance is randomly selected from 0Ω to 3kΩ, the initial fault phase angle is randomly selected from 0° to 90°, and the fault distance starts from the beginning of the line, with a point taken every 10 meters until the end of the line.

[0080] Step 3: Perform phase mode transformation on the primary voltage traveling wave signal and the secondary voltage traveling wave signal in the primary voltage traveling wave signal data set and the secondary voltage traveling wave signal data set constructed in step 2, and construct a voltage traveling wave feature set according to different scenarios based on the data after phase mode transformation;

[0081] The primary voltage traveling wave signal and the secondary voltage traveling wave signal of each measuring point after the fault are measured, and the collected primary voltage traveling wave and secondary voltage traveling wave are subjected to phase mode transformation. The primary voltage traveling wave signal is converted into the primary line mode voltage and the primary zero mode voltage, and the secondary voltage traveling wave signal is converted into the secondary line mode voltage and the secondary zero mode voltage traveling wave. Different scenarios include fault location and fault section identification. The amplitude and phase information of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave are extracted to construct the voltage traveling wave feature set under the fault location scenario.

[0082] The frequency component range of the line mode voltage traveling wave is selected to be 20Hz to 200kHz. The amplitude ratio and phase difference of the components of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and 10kHz to 100kHz are calculated to obtain the high-frequency amplitude ratio and high-frequency phase difference of the primary line mode voltage traveling wave and the high-frequency amplitude ratio and high-frequency phase difference of the secondary line mode voltage traveling wave;

[0083] Calculate the amplitude ratio and phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave at 100kHz~200kHz and 20Hz~10kHz, and obtain the high-low frequency amplitude ratio and high-low frequency phase difference of the primary line mode voltage traveling wave and the high-low frequency amplitude ratio and high-low frequency phase difference of the secondary line mode voltage traveling wave;

[0084] 100kHz~200kHz is used as the high frequency component during fault distance measurement, 10kHz~100kHz is used as the medium frequency component during fault distance measurement, and 20Hz~10kHz is used as the low frequency component during fault distance measurement;

[0085] A primary line-mode voltage traveling wave feature set including the amplitude, phase, high-intermediate frequency amplitude ratio, high-to-low frequency amplitude ratio, high-intermediate frequency phase difference and high-to-low frequency phase difference of the primary line-mode voltage traveling wave and a secondary line-mode voltage traveling wave feature set including the amplitude, phase, high-intermediate frequency amplitude ratio, high-to-low frequency amplitude ratio, high-intermediate frequency phase difference and high-to-low frequency phase difference of the secondary line-mode voltage traveling wave are constructed. The primary line-mode voltage traveling wave feature set and the secondary line-mode voltage traveling wave feature set are combined to obtain the voltage traveling wave feature set in the fault location scenario.

[0086] The secondary line mode voltage traveling wave feature set is used as the input set, and the primary line mode voltage traveling wave feature set is used as the label set. The voltage traveling wave feature set in the fault location scenario is divided into a training set, a validation set, and a test set in a ratio of 6:3:1. The amplitude and phase information of the primary zero mode voltage traveling wave and the amplitude and phase information of the secondary zero mode voltage traveling wave are extracted to construct the voltage traveling wave feature set in the fault section identification scenario.

[0087] Select the frequency component range of the zero-mode voltage traveling wave to be 20Hz to 200kHz. Calculate the amplitude ratio and phase difference of the components of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and 10kHz to 100kHz, the high-frequency amplitude ratio and high-frequency phase difference of the primary zero-mode voltage traveling wave, and the high-frequency amplitude ratio and high-frequency phase difference of the secondary zero-mode voltage traveling wave;

[0088] Calculate the amplitude ratio and phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave at 100kHz-200kHz and 20Hz-10kHz, and obtain the high-low frequency amplitude ratio and high-low frequency phase difference of the primary zero-mode voltage traveling wave and the high-low frequency amplitude ratio and high-low frequency phase difference of the secondary zero-mode voltage traveling wave;

[0089] 100kHz~200kHz is used as the high frequency component when identifying the fault section, 10kHz~100kHz is used as the medium frequency component when identifying the fault section, and 20Hz~10kHz is used as the low frequency component when identifying the fault section;

[0090] A primary zero-mode voltage traveling wave feature set including the amplitude, phase, high-intermediate frequency amplitude ratio, high-to-low frequency amplitude ratio, high-intermediate frequency phase difference and high-to-low frequency phase difference of the primary zero-mode voltage traveling wave and a secondary zero-mode voltage traveling wave feature set including the amplitude, phase, high-intermediate frequency amplitude ratio, high-to-low frequency amplitude ratio, high-intermediate frequency phase difference and high-to-low frequency phase difference of the secondary zero-mode voltage traveling wave are constructed. The primary zero-mode voltage traveling wave feature set and the secondary zero-mode voltage traveling wave feature set are combined to obtain the voltage traveling wave feature set in the fault section identification scenario.

[0091] The secondary zero-mode voltage traveling wave feature set is used as the input set, and the primary zero-mode voltage traveling wave feature set is used as the label set. The voltage traveling wave feature set in the fault section identification scenario is divided into a training set, a validation set, and a test set with a ratio of 6:3:1.

[0092] Step 4: Use the voltage traveling wave feature set constructed in step 3 to train the LSTM (Long Short-Term Memory) model to obtain a trained waveform inversion model;

[0093] When performing fault location measurement, the secondary line mode voltage traveling wave feature set in the voltage traveling wave feature set in the fault location measurement scenario is used as the input set, and the primary line mode voltage traveling wave feature set in the voltage traveling wave feature set during fault location measurement is used as the label set. The voltage traveling wave feature set in the fault location measurement scenario is divided into a training set, a validation set, and a test set.

[0094] The secondary zero-mode voltage traveling wave feature set in the voltage traveling wave feature set in the fault section identification scenario is taken as the input set, and the primary zero-mode voltage traveling wave feature set in the voltage traveling wave feature set during fault section identification is taken as the label set. The voltage traveling wave feature set in the fault section identification scenario is divided into training set, verification set and test set.

[0095] The LSTM model includes an input layer, an LSTM layer, a Dropout layer, and an output layer. The shape of the input data of the input layer depends on the data set; the LSTM layer uses two layers of LSTM, each layer is set to 64 units, the activation function is "tanh", and the gated activation function is "sigmoid"; in the Dropout layer, a Dropout of 0.2 is added after each LSTM layer; the Dense layer is used as the output layer, and the activation function is softmax. The loss function selected includes the mean square error as the loss function, and the Adam optimizer with adaptive learning rate characteristics is used. After the LSTM model is built, the model is trained using the prepared training set data to identify the amplitude-frequency and phase-frequency transmission characteristics of the voltage sensor, and a trained voltage traveling wave head inversion model is obtained.

[0096] Step 5: Input the measured fault voltage traveling wave waveform into the trained waveform inversion model obtained in step 4 to output an inverted primary traveling wave waveform, thereby realizing the inversion of the multi-branch high-voltage hybrid line fault voltage traveling wave head.

[0097] The beneficial effects of the present invention are as follows:

[0098] (1) The present invention does not need to analyze the sensor transfer function, which can avoid the influence of the inaccurate transfer function on the inversion result.

[0099] (2) A simulation model is built based on the actual multi-branch high-voltage hybrid line power grid topology and line parameters, and different fault locations, different fault resistances and different fault initial phase angles are randomly set, which is closer to the actual situation and makes it easier to obtain sample data.

[0100] (3) The overhead lines and cables of multi-branch high-voltage hybrid lines are considered separately. The electric field induction voltage transformer is used to measure the overhead line voltage traveling wave, and the capacitor voltage transformer is used to measure the cable voltage traveling wave. Two voltage traveling wave measurement methods are considered, which are more suitable for the characteristics of multi-branch high-voltage hybrid lines.

[0101] (4) By training multiple groups of data, the present invention makes the LSTM model more reliable and the output inversion of a single traveling wave more accurate.

[0102] (5) The present invention takes into account the fault distance measurement and fault section identification scenarios, and uses line mode and zero mode primary and secondary voltages for waveform inversion, respectively, which can provide more comprehensive data support for the fault distance measurement method based on line mode and zero mode voltage traveling waves.

[0103] (6) The present invention extracts four characteristic quantities of amplitude, phase, amplitude ratio and phase difference of line mode and zero mode voltage components in different frequency bands, which can realize waveform inversion more accurately.

[0104] Embodiment 2 of the present invention provides a multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system, which is used to run a multi-branch high-voltage hybrid line fault voltage traveling wave head inversion method in embodiment 1, comprising:

[0105] Hybrid line construction module: used to build and obtain the overhead-cable hybrid line simulation model, overhead line voltage traveling wave sensor model and cable line voltage traveling wave sensor model based on the multi-branch high-voltage hybrid line power grid topology;

[0106] Data acquisition module: used to set faults on the overhead-cable hybrid line simulation model, measure the secondary voltage traveling wave waveforms output by the overhead line voltage traveling wave sensor model and the cable voltage traveling wave sensor model and the primary voltage traveling wave waveforms at their installation locations, construct a primary voltage traveling wave signal data set based on the measured primary voltage traveling wave waveforms, and construct a secondary voltage traveling wave signal data set based on the measured secondary voltage traveling wave waveforms;

[0107] Feature extraction module: used to perform phase mode transformation on the primary voltage traveling wave signal and the secondary voltage traveling wave signal in the constructed primary voltage traveling wave signal data set and the secondary voltage traveling wave signal data set, and construct a voltage traveling wave feature set according to different scenarios based on the data after the phase mode transformation;

[0108] Model training module: used to train the LSTM model using the constructed voltage traveling wave feature set to obtain a trained waveform inversion model;

[0109] Wave head inversion module: It is used to input the measured fault voltage traveling wave waveform into the trained waveform inversion model to output the inverted primary traveling wave waveform, so as to realize the traveling wave head inversion of the fault voltage of multi-branch high-voltage hybrid line.

[0110] Embodiment 3 of the present invention proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is loaded into the processor, the method for inverting the traveling wave head of a multi-branch high-voltage hybrid line fault voltage described in Embodiment 1 is implemented.

[0111] Embodiment 4 of the present invention proposes a computer-readable storage medium, which stores a computer program, and is characterized in that when the computer program is executed by a processor, it implements the method for inverting the traveling wave head of a multi-branch high-voltage hybrid line fault voltage described in Embodiment 1.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line, characterized in that: include: Step 1: Based on the multi-branch high-voltage hybrid line power grid topology, build and obtain the overhead-cable hybrid line simulation model, the overhead line voltage traveling wave sensor model, and the cable line voltage traveling wave sensor model; Step 2: Set a fault on the overhead-cable hybrid line simulation model obtained in step 1, measure the secondary voltage traveling wave waveforms output by the overhead line voltage traveling wave sensor model and the cable voltage traveling wave sensor model obtained in step 1 and the primary voltage traveling wave waveforms at their installation locations, construct a primary voltage traveling wave signal data set based on the measured primary voltage traveling wave waveform, and construct a secondary voltage traveling wave signal data set based on the measured secondary voltage traveling wave waveform; Step 3: Perform phase mode transformation on the primary voltage traveling wave signal and the secondary voltage traveling wave signal in the primary voltage traveling wave signal data set and the secondary voltage traveling wave signal data set constructed in step 2, and construct a voltage traveling wave feature set according to different scenarios and the data after phase mode transformation; Step 4: Use the voltage traveling wave feature set constructed in step 3 to train the LSTM model to obtain a trained waveform inversion model; Step 5: Input the measured fault voltage traveling wave waveform into the trained waveform inversion model obtained in step 4 to output an inverted primary traveling wave waveform, thereby realizing the inversion of the multi-branch high-voltage hybrid line fault voltage traveling wave head.

2. A method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 1, characterized in that: In step 1, an overhead line voltage traveling wave sensor model and a cable line voltage traveling wave sensor model are constructed, including: An electric field induction voltage transformer is installed on the overhead line tower. According to the spatial position of the electric field induction voltage transformer relative to the conductor and the physical relationship between the electric field strength and the voltage, a traveling wave sensor model of the overhead line voltage is constructed. A capacitive voltage divider transformer is installed at the end of the cable, and a cable line voltage traveling wave sensor model is built based on the capacitive voltage divider transformer.

3. The method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 1 is characterized by: In step 2, setting faults on the overhead-cable hybrid line simulation model obtained in step 1 includes: randomly setting multiple groups of faults at the junctions of overhead lines, cables and overhead cables of the multi-branch high-voltage overhead-cable hybrid line simulation model, each group of faults including a fault location, a fault resistance and a fault initial phase angle; The fault resistance is randomly selected from 0Ω to 3kΩ, and the fault initial phase angle is randomly selected from 0° to 90°.

4. The method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 1 is characterized by: In step 3, the collected primary voltage traveling wave signal and secondary voltage traveling wave signal are subjected to phase mode transformation, including: obtaining a primary line mode voltage traveling wave and a primary zero mode voltage traveling wave after phase mode transformation of the primary voltage traveling wave signal, and obtaining a secondary line mode voltage traveling wave and a secondary zero mode voltage traveling wave after phase mode transformation of the secondary voltage traveling wave signal.

5. The method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 4 is characterized by: According to different scenarios and data after phase mode transformation, a voltage traveling wave feature set is constructed, including different scenarios including fault location scenario and fault section identification scenario; Extract the amplitude and phase information of the primary line mode voltage traveling wave and the amplitude and phase information of the secondary line mode voltage traveling wave, and construct a voltage traveling wave feature set in the fault location scenario; The amplitude and phase information of the primary zero-mode voltage traveling wave and the amplitude and phase information of the secondary zero-mode voltage traveling wave are extracted to construct a voltage traveling wave feature set for the fault section identification scenario.

6. A method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 5, characterized in that: The amplitude and phase information in the voltage traveling wave feature set in the fault location scenario and the voltage traveling wave feature set in the fault section identification scenario include the high-mid frequency amplitude ratio, the high-low frequency amplitude ratio, the high-mid frequency phase difference and the high-low frequency phase difference.

7. A method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 6, characterized in that: The high-intermediate frequency amplitude ratio and high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault location scenario are the amplitude ratio and phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and 10kHz to 100kHz respectively; The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault location scenario are the amplitude ratio and phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively; Among them, 100kHz~200kHz is used as the high-frequency component during fault distance measurement, 10kHz~100kHz is used as the medium-frequency component during fault distance measurement, and 20Hz~10kHz is used as the low-frequency component during fault distance measurement.

8. The method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 6 is characterized by: The high-intermediate frequency amplitude ratio and high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are the amplitude ratio and phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and 10kHz to 100kHz respectively; The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are the amplitude ratio and phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively; Among them, 100kHz~200kHz is used as the high-frequency component when identifying the fault section, 10kHz~100kHz is used as the medium-frequency component when identifying the fault section, and 20Hz~10kHz is used as the low-frequency component when identifying the fault section.

9. The method for inverting the traveling wave head of fault voltage of a multi-branch high-voltage hybrid line according to claim 1 is characterized by: In step 4, the amplitude and phase information of the secondary line mode voltage traveling wave feature in the voltage traveling wave feature set in the fault location scenario is used as the input set, and the amplitude and phase information of the primary line mode voltage traveling wave in the voltage traveling wave feature set during fault location is used as the label set, and the voltage traveling wave feature set in the fault location scenario is divided into a training set, a validation set, and a test set; The amplitude and phase information of the secondary zero-mode voltage traveling wave in the voltage traveling wave feature set in the fault section identification scenario are set as the input set, and the amplitude and phase information of the primary zero-mode voltage traveling wave feature in the voltage traveling wave feature set during fault section identification are set as the label set. The voltage traveling wave feature set in the fault section identification scenario is divided into a training set, a validation set and a test set.

10. A multi-branch high-voltage hybrid line fault voltage traveling wave front inversion system according to claim 1, used to run a multi-branch high-voltage hybrid line fault voltage traveling wave front inversion method according to any one of claims 1-9, characterized in that: Hybrid line construction module: used to build and obtain the overhead-cable hybrid line simulation model, overhead line voltage traveling wave sensor model and cable line voltage traveling wave sensor model based on the multi-branch high-voltage hybrid line power grid topology; Data acquisition module: used to set faults on the overhead-cable hybrid line simulation model, measure the secondary voltage traveling wave waveforms output by the overhead line voltage traveling wave sensor model and the cable voltage traveling wave sensor model and the primary voltage traveling wave waveforms at their installation locations, construct a primary voltage traveling wave signal data set based on the measured primary voltage traveling wave waveforms, and construct a secondary voltage traveling wave signal data set based on the measured secondary voltage traveling wave waveforms; Feature extraction module: used to perform phase mode transformation on the primary voltage traveling wave signal and the secondary voltage traveling wave signal in the constructed primary voltage traveling wave signal data set and the secondary voltage traveling wave signal data set, and construct a voltage traveling wave feature set according to different scenarios and the data after phase mode transformation; Model training module: used to train the LSTM model using the constructed voltage traveling wave feature set to obtain a trained waveform inversion model; Wave head inversion module: It is used to input the measured fault voltage traveling wave waveform into the trained waveform inversion model to output the inverted primary traveling wave waveform, so as to realize the traveling wave head inversion of the fault voltage of multi-branch high-voltage hybrid line.

11. A multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 10, characterized in that: Build the overhead line voltage traveling wave sensor model and the cable line voltage traveling wave sensor model, including: An electric field induction voltage transformer is installed on the overhead line tower. According to the spatial position of the electric field induction voltage transformer relative to the conductor and the physical relationship between the electric field strength and the voltage, a traveling wave sensor model of the overhead line voltage is constructed. A capacitive voltage divider transformer is installed at the end of the cable, and a cable line voltage traveling wave sensor model is built based on the capacitive voltage divider transformer.

12. The multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 10, characterized in that: Setting faults on the obtained overhead-cable hybrid line simulation model includes: randomly setting multiple groups of faults at the junctions of overhead lines, cables and overhead cables of the multi-branch high-voltage overhead-cable hybrid line simulation model, each group of faults including a fault location, a fault resistance and a fault initial phase angle; The fault resistance is randomly selected from 0Ω to 3kΩ, and the fault initial phase angle is randomly selected from 0° to 90°.

13. The multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 10, characterized in that: The collected primary voltage traveling wave signal and secondary voltage traveling wave signal are subjected to phase mode transformation, including: obtaining a primary line mode voltage traveling wave and a primary zero mode voltage traveling wave after phase mode transformation of the primary voltage traveling wave signal, and obtaining a secondary line mode voltage traveling wave and a secondary zero mode voltage traveling wave after phase mode transformation of the secondary voltage traveling wave signal.

14. A multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 13, characterized in that: According to different scenarios and data after phase mode transformation, a voltage traveling wave feature set is constructed, including different scenarios including fault location scenario and fault section identification scenario; Extract the amplitude and phase information of the primary line mode voltage traveling wave and the amplitude and phase information of the secondary line mode voltage traveling wave, and construct a voltage traveling wave feature set in the fault location scenario; The amplitude and phase information of the primary zero-mode voltage traveling wave and the amplitude and phase information of the secondary zero-mode voltage traveling wave are extracted to construct a voltage traveling wave feature set for the fault section identification scenario.

15. A multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 14, characterized in that: The amplitude and phase information in the voltage traveling wave feature set in the fault location scenario and the voltage traveling wave feature set in the fault section identification scenario include the high-mid frequency amplitude ratio, the high-low frequency amplitude ratio, the high-mid frequency phase difference and the high-low frequency phase difference.

16. A multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 15, characterized in that: The high-intermediate frequency amplitude ratio and high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault location scenario are the amplitude ratio and phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and 10kHz to 100kHz respectively; The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault location scenario are the amplitude ratio and phase difference of the primary line mode voltage traveling wave and the secondary line mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively; Among them, 100kHz~200kHz is used as the high-frequency component during fault distance measurement, 10kHz~100kHz is used as the medium-frequency component during fault distance measurement, and 20Hz~10kHz is used as the low-frequency component during fault distance measurement.

17. A multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 15, characterized in that: The high-intermediate frequency amplitude ratio and high-intermediate frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are the amplitude ratio and phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave in the frequency component range of 100kHz to 200kHz and 10kHz to 100kHz respectively; The high-to-low frequency amplitude ratio and high-to-low frequency phase difference in the voltage traveling wave feature set in the fault section identification scenario are the amplitude ratio and phase difference of the primary zero-mode voltage traveling wave and the secondary zero-mode voltage traveling wave at 100kHz to 200kHz and 20Hz to 10kHz respectively; Among them, 100kHz~200kHz is used as the high-frequency component when identifying the fault section, 10kHz~100kHz is used as the medium-frequency component when identifying the fault section, and 20Hz~10kHz is used as the low-frequency component when identifying the fault section.

18. The multi-branch high-voltage hybrid line fault voltage traveling wave head inversion system according to claim 10, characterized in that: The amplitude and phase information of the secondary line mode voltage traveling wave characteristics in the voltage traveling wave feature set in the fault location scenario are used as the input set, and the amplitude and phase information of the primary line mode voltage traveling wave in the voltage traveling wave feature set during fault location is used as the label set. The voltage traveling wave feature set in the fault location scenario is divided into a training set, a validation set and a test set. The amplitude and phase information of the secondary zero-mode voltage traveling wave in the voltage traveling wave feature set in the fault section identification scenario are set as the input set, and the amplitude and phase information of the primary zero-mode voltage traveling wave feature in the voltage traveling wave feature set during fault section identification are set as the label set. The voltage traveling wave feature set in the fault section identification scenario is divided into a training set, a validation set and a test set.

19. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into a processor, a method for inverting a traveling wave head of a multi-branch high-voltage hybrid line fault voltage is implemented according to any one of claims 1 to 9.

20. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, a method for inverting a traveling wave head of a multi-branch high-voltage hybrid line fault voltage is implemented according to any one of claims 1 to 9.

Images