Power distribution network voltage transient transmission performance recovery method and device, computer equipment and readable storage medium

Abstract

The invention relates to the technical field of power distribution network line relay protection, and provides a power distribution network voltage transient transmission performance recovery method and device, computer equipment and a readable storage medium. The method comprises the following steps: acquiring an initial secondary voltage array of the power distribution network; performing transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array; obtaining a primary predicted voltage array of the power distribution network according to the target secondary voltage array and a pre-constructed voltage mapping model; and determining whether the voltage of the power distribution network is recovered to a normal stable state or not according to the primary prediction voltage array. By adopting the method, the problem of maloperation of the relay protection device caused by distortion of the secondary voltage signal can be solved, few computing resources are needed, and threats to the quick action of the relay protection device can be reduced.

Description

Technical Field

[0001] The present application relates to the technical field of distribution network line relay protection, and in particular to a distribution network voltage transient transmission performance recovery method, device, computer equipment, computer readable storage medium and computer program product. Background Art

[0002] Capacitive Voltage Transformer (CVT) is widely used in medium and high voltage distribution networks due to its low cost and low ferromagnetic resonance. However, compared with electromagnetic voltage transformers, CVTs have poor transient characteristics. When a fault occurs, the secondary voltage of the electromagnetic voltage transformer will experience a serious transient process. The resulting transient error will reduce the signal quality of the secondary voltage and cause the relay protection device to malfunction. Therefore, taking corresponding measures to improve the signal quality of the secondary voltage of the CVT is of great engineering significance for ensuring the correct operation of the secondary equipment.

[0003] The secondary signal recovery problem of capacitor voltage transformer belongs to a complex nonlinear time series prediction problem. The traditional numerical transformation method is difficult to achieve the secondary signal recovery of capacitor voltage transformer under complex working conditions due to the different parameters and variable fault types of capacitor voltage transformer. Neural networks have shown excellent performance in dealing with nonlinear problems. Convolutional neural networks in deep learning technology are widely used in the prediction field of time series problems. Introducing neural networks to recover the secondary signals of capacitor voltage transformers is expected to solve the above difficulties. However, the matrix multiplication operation in the traditional fully connected neural network requires a lot of computing resources, resulting in the problem of long computing time for the online deployment of the model based on the fully connected neural network, which may pose a threat to the speed of relay protection devices. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for restoring the transient transmission performance of distribution network voltage in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for restoring transient voltage transmission performance of a distribution network, comprising:

[0006] Get the initial secondary voltage array of the distribution network;

[0007] Performing transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array;

[0008] Obtaining a predicted primary voltage array of the distribution network according to the target secondary voltage array and a pre-built voltage mapping model;

[0009] According to the primary predicted voltage array, it is determined whether to restore the voltage of the distribution network to a normal stable state.

[0010] In one embodiment, the step of performing ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array includes:

[0011] According to the conversion ratio, the initial secondary voltage array is converted by transformation ratio to obtain a secondary converted voltage array;

[0012] Determining the maximum value of the absolute value of the secondary converted voltage from the secondary converted voltage array;

[0013] Taking the maximum value of the absolute value of the secondary converted voltage as the first reference value;

[0014] A target secondary voltage array is obtained according to the secondary converted voltage array and the first reference value.

[0015] In one of the embodiments, before obtaining the primary predicted voltage array of the distribution network according to the target secondary voltage array and the pre-built voltage mapping model, the method further includes:

[0016] Obtain target primary voltage array samples and target secondary voltage array samples;

[0017] Performing at least one training according to the target primary voltage array samples and the target secondary voltage array samples;

[0018] Among them, each time training, the following steps are performed: through the convolution layer in the parameter to be adjusted model carrying the training parameters of this time, the target secondary voltage array samples are subjected to convolution operation and gradient operation to obtain a secondary voltage feature array; according to the secondary voltage feature array and the target primary voltage array samples, the imbalance of the convolution layer is obtained; through the pooling layer in the parameter to be adjusted model carrying the training parameters of this time and the imbalance of the convolution layer, the secondary voltage feature array is subjected to pooling operation to obtain a first predicted voltage array sample; when the difference between the first predicted voltage array sample and the target primary voltage array sample does not meet the conditions, the training parameters of this time are adjusted to obtain the next training parameters and enter the next training; when the difference between the first predicted voltage array sample and the target primary voltage array sample meets the conditions, the parameter to be adjusted model carrying the training parameters of this time is used as the voltage mapping model.

[0019] In one embodiment, the step of obtaining a target primary voltage array sample and a target secondary voltage array sample includes:

[0020] The transfer characteristics between the primary voltage and the secondary voltage of the capacitor voltage transformer in the distribution network are modeled to obtain the equivalent circuit model of the capacitor voltage transformer;

[0021] Obtaining a transmission line fault simulation model of a distribution network;

[0022] Applying the capacitor voltage transformer equivalent circuit model to the power transmission line simulation model, and obtaining the primary voltage array sample and the secondary voltage array sample of the capacitor voltage transformer when the distribution network line fails by setting different fault conditions and capacitor voltage transformer parameters;

[0023] Performing normalization calculation on the primary voltage array samples to obtain target primary voltage array samples;

[0024] The secondary voltage array samples are subjected to transformation ratio conversion and normalization calculation to obtain target secondary voltage array samples.

[0025] In one embodiment, the normalizing calculation of the primary voltage array samples to obtain target primary voltage array samples includes:

[0026] Determine the maximum value of the absolute value of the primary voltage from the primary voltage array samples;

[0027] Taking the maximum value of the absolute value of the primary voltage as the second reference value;

[0028] A target primary voltage array sample is obtained according to the primary voltage array sample and the second reference value.

[0029] In one embodiment, determining whether to restore the distribution network voltage to a normal stable state according to the primary predicted voltage array includes:

[0030] When all the primary predicted voltages in the primary predicted voltage array are less than the fault threshold, determining to restore the distribution network voltage to a normal stable state;

[0031] When there is a primary predicted voltage in the primary predicted voltage array that is greater than a fault threshold, it is determined not to restore the distribution network voltage to a normal stable state.

[0032] In a second aspect, the present application also provides a distribution network voltage transient transmission performance recovery device, comprising:

[0033] An initial secondary voltage acquisition module is used to obtain an initial secondary voltage array of a distribution network;

[0034] A target secondary voltage acquisition module is used to perform transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array;

[0035] A primary predicted voltage acquisition module, used to obtain a primary predicted voltage array of the distribution network according to the target secondary voltage array and a pre-built voltage mapping model;

[0036] The judgment recovery module is used to determine whether to restore the voltage of the distribution network to a normal and stable state according to the primary predicted voltage array.

[0037] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the above method.

[0038] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to execute the above method.

[0039] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and the computer program is executed by a processor to execute the above method.

[0040] The above-mentioned distribution network voltage transient transmission performance recovery method, device, computer equipment, computer-readable storage medium and computer program product obtain the initial secondary voltage array of the distribution network; perform transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain the target secondary voltage array; obtain the primary predicted voltage array of the distribution network based on the target secondary voltage array and the pre-constructed voltage mapping model; determine whether to restore the voltage of the distribution network to a normal and stable state based on the primary predicted voltage array. The present application can obtain a relatively accurate primary predicted voltage array based on the target secondary voltage array and the pre-constructed voltage mapping model; determine whether to restore the voltage of the distribution network to a normal and stable state based on the primary predicted voltage array, which solves the problem of false operation of the relay protection device due to distortion of the secondary voltage signal, and requires fewer computing resources, which can reduce the threat to the speed of the relay protection device. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 This is an application environment diagram of a method for recovering transient voltage transmission performance of a distribution network in one embodiment;

[0043] Figure 2 It is a flow chart of a method for recovering transient voltage transmission performance of a distribution network in one embodiment;

[0044] FIG3 (a) is a schematic diagram of the electrical structure of a capacitive voltage transformer in one embodiment;

[0045] FIG3( b ) is a schematic diagram of an equivalent circuit of a capacitive voltage transformer in one embodiment;

[0046] Figure 4 A schematic diagram of a transmission line fault simulation model in one embodiment;

[0047] Figure 5 A schematic diagram of a circuit model in an embodiment;

[0048] Figure 6 It is a flow chart of a method for recovering transient voltage transmission performance of a power distribution network in another embodiment;

[0049] Figure 7 A loss function curve diagram in one embodiment;

[0050] Figure 8 A comparison diagram of a predicted primary voltage array sample and a target primary voltage array sample at a single-phase metallic grounding fault point in an embodiment;

[0051] Fig. 9 A comparison diagram of a predicted primary voltage array sample and a target primary voltage array sample at a two-phase metallic grounding fault point in one embodiment;

[0052] Fig.10 It is a structural block diagram of a device for recovering transient voltage variation performance of a distribution network in one embodiment;

[0053] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] The present application embodiment provides a method for restoring the transient voltage performance of a distribution network. The present application embodiment can be executed by a computer device, such as Figure 1As shown, the computer device can obtain the initial secondary voltage array of the distribution network, and then obtain the primary predicted voltage array, so as to determine whether to restore the voltage of the distribution network to a normal stable state. It is understandable that the computer device can be implemented by a server, or by a terminal, or by an interactive system between a terminal and a server. In this embodiment, the method includes Figure 2 The steps shown are:

[0056] Step S201, obtaining an initial secondary voltage array of the distribution network.

[0057] The voltage data at the distribution network protection installation can be monitored in real time to obtain the initial secondary voltage array U with a window length of 1ms after the distribution network fault.

[0058] Step S202, performing transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array.

[0059] The initial secondary voltage array can be converted and normalized to obtain the target secondary voltage array .

[0060] Step S203, obtaining a predicted primary voltage array of the distribution network according to the target secondary voltage array and the pre-built voltage mapping model.

[0061] The voltage mapping model can predict the primary voltage based on the secondary voltage, and can realize the reconstruction of low-quality secondary voltage signals in the medium-voltage distribution network.

[0062] The target secondary voltage array can be input into a pre-built voltage mapping model to obtain the primary predicted voltage array of the distribution network, thereby realizing the recovery of the distorted secondary voltage of the distribution network.

[0063] Step S204: determining whether to restore the voltage of the distribution network to a normal stable state according to the primary predicted voltage array.

[0064] The predicted voltage array can be The relevant relay protection devices are input for fault identification, and according to the fault identification results, it is determined whether the voltage of the distribution network should be restored to a normal and stable state.

[0065] In the above-mentioned distribution network voltage transient transmission performance recovery method, a relatively accurate primary predicted voltage array can be obtained according to the target secondary voltage array and the pre-built voltage mapping model; according to the primary predicted voltage array, it can be determined whether to restore the voltage of the distribution network to a normal and stable state, which solves the problem of false operation of the relay protection device due to the distortion of the secondary voltage signal, and requires less computing resources, which can reduce the threat to the speed of the relay protection device.

[0066] In one of the embodiments, the initial secondary voltage array is subjected to ratio conversion and normalization calculation to obtain a target secondary voltage array, and the specific steps are as follows: according to the conversion ratio, the initial secondary voltage array is subjected to ratio conversion to obtain a secondary converted voltage array; from the secondary converted voltage array, the maximum value of the absolute value of the secondary converted voltage is determined; the maximum value of the absolute value of the secondary converted voltage is used as the first reference value; and according to the secondary converted voltage array and the first reference value, the target secondary voltage array is obtained.

[0067] The conversion ratio can be obtained according to the number of turns of the primary winding and the number of turns of the secondary winding of the capacitor voltage transformer. According to the conversion ratio, the initial secondary voltage array is converted by ratio to obtain the secondary converted voltage array, as shown in formula (1).

[0068] (1)

[0069] in, is the secondary converted voltage array; , are the primary and secondary winding turns of the capacitive voltage transformer respectively.

[0070] The maximum value of the absolute value of the secondary converted voltage can be used as the first reference value; according to the secondary converted voltage array and the first reference value, a target secondary voltage array is obtained, as shown in formula (2).

[0071] (2)

[0072] in, is the target secondary voltage array, is the first reference value.

[0073] In this embodiment, the initial secondary voltage array is converted according to the conversion ratio to obtain a secondary converted voltage array; a first reference value is determined from the secondary converted voltage array; and a target secondary voltage array is obtained according to the secondary converted voltage array and the first reference value, which can improve the convergence degree of the target secondary voltage array, thereby obtaining a more accurate primary predicted voltage array.

[0074] In one of the embodiments, before obtaining a primary predicted voltage array of the distribution network according to the target secondary voltage array and the pre-built voltage mapping model, the method provided by the present application also includes: obtaining target primary voltage array samples and target secondary voltage array samples; performing at least one training according to the target primary voltage array samples and the target secondary voltage array samples; wherein, during each training, the following steps are performed: performing convolution operation and gradient operation on the target secondary voltage array samples through the convolution layer in the parameter to be adjusted model carrying the training parameters of this time, to obtain a secondary voltage feature array; obtaining the imbalance of the convolution layer according to the secondary voltage feature array and the target primary voltage array samples; performing pooling operation on the secondary voltage feature array through the pooling layer and the imbalance of the convolution layer in the parameter to be adjusted model carrying the training parameters of this time, to obtain a primary predicted voltage array sample; when the difference between the primary predicted voltage array sample and the target primary voltage array sample does not meet the conditions, adjusting the current training parameters to obtain the next training parameters and entering the next training; when the difference between the primary predicted voltage array sample and the target primary voltage array sample meets the conditions, the parameter to be adjusted model carrying the training parameters of this time is used as the voltage mapping model.

[0075] The target primary voltage array samples and the target secondary voltage array samples may be input into the parameter to be adjusted model and trained at least once; wherein, during each training, the following steps are performed:

[0076] By carrying the convolution layer in the parameter to be adjusted model with the training parameters, convolution operation and gradient operation are performed on the target quadratic voltage array samples to obtain the quadratic voltage feature array, as shown in formula (3).

[0077] (3)

[0078] in, is the secondary voltage characteristic array, is the target secondary voltage array sample, represents the convolution operation, represents the gradient operation, w is the number of convolution kernels in a single convolution layer filter, b is the size of the convolution kernel, and the initial values ​​of w and b are both set to 20.

[0079] According to the secondary voltage feature array and the target primary voltage array samples, the imbalance of the convolution layer is obtained, as shown in formula (4).

[0080] (4)

[0081] in, is the imbalance of the convolutional layer, An array of target primary voltage samples.

[0082] By carrying the imbalance of the pooling layer and the convolution layer in the parameter adjustment model of the training parameters, the secondary voltage feature array is pooled, and the extracted features in the convolution layer are selected to obtain a predicted voltage array sample, as shown in formula (5).

[0083] (5)

[0084] in, is a sample of the predicted voltage array, is the secondary voltage characteristic array, is the imbalance of the convolutional layer, is the number of filters in the pooling layer, is the range of a single pooling layer filter, , The initial value of is set to 1.

[0085] The root mean square error (RMSE) can be used as the loss function RMSE , measures the difference between the predicted voltage array sample and the target voltage array sample, as shown in formula (6).

[0086] (6)

[0087] Among them, K is the length of the data to be evaluated, which can be taken as 100; The target voltage array sample is a single voltage array sample; It is a sample of the predicted voltage array.

[0088] The mean absolute percentage error (MAPE) can be used as the precision evaluation loss function e MAPE , which measures the prediction accuracy of the model whose parameters are to be adjusted, as shown in formula (7).

[0089] (7)

[0090] Among them, K is the length of the data to be evaluated, which can be taken as 100; The target voltage array sample is a single voltage array sample; It is a sample of the predicted voltage array.

[0091] The training cutoff condition of the model whose parameters are to be adjusted can be set according to the actual situation, as shown in formula (8).

[0092] (8)

[0093] in, is the loss function eRMSE and the mean absolute percentage error e MAPE The loss function threshold.

[0094] When the loss function e RMSE Or the mean absolute percentage error e MAPE When it is greater than the loss function threshold, the difference between the predicted voltage array sample and the target voltage array sample does not meet the condition. At this time, adjust the training parameters w, b, and The adjustment method is to increase the parameter by a corresponding step size, obtain the next training parameter and enter the next training, where the adjustment step size of w and b is set to 4. and The adjustment step size is set to 1.

[0095] When the loss function e RMSE and the mean absolute percentage error e MAPE When they are all less than or equal to the loss function threshold, the difference between the predicted voltage array sample and the target voltage array sample meets the condition, and the parameter adjustment model carrying the training parameters is used as the voltage mapping model.

[0096] After determining the training parameters of the voltage mapping model, the imbalance of the convolutional layer of the voltage mapping model can be determined. , in the application stage of the voltage mapping model, the imbalance of the convolutional layer It participates in the calculation as a fixed intermediate variable value in the voltage mapping model.

[0097] In this embodiment, at least one training is performed based on the target primary voltage array samples and the target secondary voltage array samples. When the difference between the predicted voltage array samples and the target primary voltage array samples meets the conditions, the parameter adjustment model carrying the training parameters is used as the voltage mapping model. Through the above training method, a voltage mapping model with good prediction performance can be obtained.

[0098] In one of the embodiments, target primary voltage array samples and target secondary voltage array samples are obtained, and the specific steps are as follows: modeling the transfer characteristics between the primary voltage and the secondary voltage of the capacitor voltage transformer in the distribution network to obtain an equivalent circuit model of the capacitor voltage transformer; obtaining a transmission line fault simulation model of the distribution network; applying the capacitor voltage transformer equivalent circuit model to the transmission line simulation model, and obtaining the primary voltage array samples and secondary voltage array samples of the capacitor voltage transformer when the distribution network line fails by setting different fault conditions and capacitor voltage transformer parameters; normalizing the primary voltage array samples to obtain target primary voltage array samples; performing transformation ratio conversion and normalization calculation on the secondary voltage array samples to obtain target secondary voltage array samples.

[0099] The transfer characteristics between the primary voltage and the secondary voltage of the capacitor voltage transformer in the distribution network can be modeled according to the electromagnetic transient simulation software (Power Systems Computer Aided Design, PSCAD), and the equivalent circuit model of the capacitor voltage transformer can be obtained.

[0100] The electrical structure and equivalent circuit of the capacitor voltage transformer are shown in FIG3 , wherein FIG3 (a) is the electrical structure diagram of the capacitor voltage transformer, and FIG3 (b) is the equivalent circuit diagram of the capacitor voltage transformer.

[0101] In the electrical structure of Figure 3 (a), C1 is a high-voltage capacitor, C2 is a low-voltage capacitor; L is a compensating reactor; G is a protective gap; an is a secondary measurement winding; and T is an intermediate transformer.

[0102] In the equivalent circuit of Figure 3(b), C e is the equivalent capacitance, which is the sum of the high-voltage capacitance C1 and the low-voltage capacitance C2. The intermediate transformer adopts a T-type equivalent circuit for equivalent circuit, L m and R m are the equivalent impedance of the excitation branch of the intermediate transformer, L m =1.35H, R m =2000Ω, L k , R k is the sum of the leakage impedance of the primary winding of the intermediate transformer and the equivalent impedance of the compensator, L k =90.41μH, R k =4.2Ω, R2 and L2 are the sum of the resistance and inductance converted to the primary load and the leakage impedance of the secondary winding of the intermediate transformer, R2=10Ω, L2=231.8mH; R f , L f , C f 、r f They are the resistance, inductance, capacitance and small resistance of the damper respectively, where L f =18.60μH; U pri0 is the primary voltage, U sec0 is the secondary voltage, is the converted secondary voltage. Among them, there are three main circuit parameters that affect the transmission characteristics of the capacitive voltage transformer, namely, the equivalent capacitance C e , Damping resistance R f , damping capacitor C f .

[0103] The transmission line fault simulation model can be built based on the electromagnetic transient simulation software. Figure 4As shown, it is an AC distribution system at a voltage level of 10kV, in which the capacitive voltage transformer is connected to the bus M; k1 and k2 are different fault points, k1 is a single-phase metallic grounding fault point with a distance of 10km from the bus M, and k2 is a two-phase grounding fault point with a distance of 100km from the bus M.

[0104] The equivalent circuit model of capacitor voltage transformer is applied to the transmission line simulation model. By setting different fault conditions and capacitor voltage transformer parameters, the primary voltage array samples and secondary voltage array samples of the capacitor voltage transformer when the distribution network line fails are obtained.

[0105] Capacitor voltage transformer parameters, fault type, fault location, fault phase angle, transition resistance, load capacity and load power factor have an impact on the transfer characteristics of capacitor voltage transformer. N =10kV, rated current I N =100A, rated frequency f N =50Hz, equivalent internal impedance Z s =14.48ohms, minimum load capacity 2MVA, maximum load capacity 5MVA, overhead line adopts Type equivalent circuit, the total length of the protected line is 50km, the line model is as follows Figure 5 As shown in the figure, the transformers on both sides of the transmission line are Y0-△ connected, with a transformation ratio of 35kV / 10kV. The specific settings are as follows:

[0106] ① Fault type: single-phase grounding (Ag), two-phase grounding fault (ABg), three-phase symmetrical fault;

[0107] ② Fault initial phase angle: traverse from 0 to 90°, and set the interval step to 5°;

[0108] ③ Transition resistance: 0~300Ω, with 30Ω as the step length;

[0109] ④ Fault location: traverse from 0 to 50 km on the protected line with a step length of 1.25 km;

[0110] ⑤ Capacitive voltage transformer equivalent capacitance C e : Traverse from 39.285 to 157.123 nF, with a step size of 7.857 nF;

[0111] ⑥ Capacitive voltage transformer damping resistor R f : traverse from 65.9 to 263.6kohms, with a step size of 1.32kohms;

[0112] ⑦ Capacitive voltage transformer damping capacitor C f: Traverse from 7.929 to 31.716nF, with a step size of 1.86nF;

[0113] ⑧ Load capacity: traverse from 2 to 5 MVA, with a step size of 0.2 MVA;

[0114] ⑨Load power factor : Traverse from 0.64 to 0.96, with a step size of 0.04.

[0115] In the electromagnetic transient simulation software, the sampling rate f s The frequency is set to 100kHz, the fault duration is 1s, and multiple operations of different fault conditions are set on the line according to the above settings. The primary fault voltage data and the secondary fault voltage data collected by the capacitive voltage transformer are collected to form a fault voltage array, including the primary voltage array sample. and secondary voltage array samples Each of the above arrays contains 7200 sets of data, each of which stores 100 points of data, that is, data with a time length of 1ms.

[0116] The primary voltage array samples are normalized and calculated to obtain target primary voltage array samples; the secondary voltage array samples are ratio-converted and normalized and calculated to obtain target secondary voltage array samples.

[0117] In this embodiment, the equivalent circuit model of the capacitor voltage transformer is applied to the transmission line simulation model. By setting different fault conditions and capacitor voltage transformer parameters, the primary voltage array samples and secondary voltage array samples of the capacitor voltage transformer when the distribution network line fails are obtained, and rich training data are generated. The primary voltage array samples and the secondary voltage array samples are processed to obtain target primary voltage array samples and target secondary voltage array samples, so as to prepare data for the training of the voltage mapping model.

[0118] In one of the embodiments, a normalization calculation is performed on the primary voltage array samples to obtain a target primary voltage array sample, and the specific steps are as follows: determining the maximum value of the absolute value of the primary voltage from the primary voltage array samples; using the maximum value of the absolute value of the primary voltage as a second reference value; and obtaining the target primary voltage array sample based on the primary voltage array samples and the second reference value.

[0119] From the primary voltage array samples, determine the maximum value of the primary voltage absolute value; use the maximum value of the primary voltage absolute value as the second reference value; and obtain the target primary voltage array samples according to the primary voltage array samples and the second reference value, as shown in formula (9).

[0120] (9)

[0121] in, is the target primary voltage array sample, is a voltage array sample, is the second reference value.

[0122] In this embodiment, target primary voltage array samples are obtained according to the primary voltage array samples and the second reference value, so as to prepare data for the training of the voltage mapping model.

[0123] In one of the embodiments, whether to restore the distribution network voltage to a normal and stable state is determined based on a primary predicted voltage array, and the specific steps are as follows: when all the primary predicted voltages in the primary predicted voltage array are less than the fault threshold, it is determined that the distribution network voltage is restored to a normal and stable state; when there is a primary predicted voltage in the primary predicted voltage array that is greater than the fault threshold, it is determined not to restore the distribution network voltage to a normal and stable state.

[0124] The fault threshold can be set according to the actual situation.

[0125] When all the primary prediction voltages in the primary prediction voltage array are less than the fault threshold, it indicates that the secondary voltage signal is distorted and causes the relay protection device to malfunction. At this time, the distribution network voltage can be restored to a normal and stable state. When there is a primary prediction voltage in the primary prediction voltage array that is greater than the fault threshold, it indicates that the distribution network is in a fault state. At this time, the distribution network voltage will not be restored to a normal and stable state.

[0126] In order to better understand the above method, an application example of the method for recovering transient voltage transmission performance of the distribution network of the present application is described in detail below. Figure 6 shown.

[0127] Step S601, building a transmission line fault simulation model according to electromagnetic transient simulation software.

[0128] Step S602, applying the capacitor voltage transformer equivalent circuit model to the transmission line simulation model, and obtaining target primary voltage array samples and target secondary voltage array samples by setting different fault conditions and capacitor voltage transformer parameters.

[0129] Specifically, the transfer characteristics between the primary voltage and the secondary voltage of the capacitor voltage transformer in the distribution network are modeled to obtain an equivalent circuit model of the capacitor voltage transformer; a transmission line fault simulation model of the distribution network is obtained; the capacitor voltage transformer equivalent circuit model is applied to the transmission line simulation model, and by setting different fault conditions and capacitor voltage transformer parameters, the primary voltage array samples and the secondary voltage array samples of the capacitor voltage transformer when the distribution network line fails are obtained; the primary voltage array samples are normalized to obtain target primary voltage array samples; the secondary voltage array samples are ratio-converted and normalized to obtain target secondary voltage array samples.

[0130] Step S603 , performing at least one training according to the target primary voltage array samples and the target secondary voltage array samples to obtain a voltage mapping model.

[0131] Specifically, the python TensorFlow library is used to build the convolution layer in the model whose parameters are to be adjusted. The number of kernels w and the initial value of the convolution kernel size b are both set to 20. The python TensorFlow library is used to build the pooling layer in TGN. The number of filters in the pooling layer , the range of a single pooling layer filter The initial value is set to 1. The target primary voltage array samples and the target secondary voltage array samples are input into the parameter adjustment model shown in formula (10), and training is performed at least once to obtain a voltage mapping model.

[0132] (10)

[0133] The input of the model whose parameters are to be adjusted is the target primary voltage array sample. and target secondary voltage array samples , the output is a predicted voltage array sample , Represents the calculation process of the convolution layer and the pooling layer; Θ is w, b, , A collection of four parameters.

[0134] Each time training is performed, the following steps are performed: through the convolution layer in the parameter to be adjusted model carrying the training parameters, convolution operation and gradient operation are performed on the target quadratic voltage array samples to obtain the absolute value of the quadratic voltage feature array =1.21 =0.74 ;According to the secondary voltage feature array and the target primary voltage array samples, the imbalance of the convolution layer is obtained =0.361; By carrying the imbalance of the pooling layer and the convolution layer in the parameter to be adjusted in the training model, the secondary voltage feature array is pooled to obtain the absolute value of the predicted voltage array sample. =2.63 =0.97 .

[0135] The root mean square error (RMSE) can be used as the loss function RMSE , we can also measure the difference between the predicted voltage array sample and the target voltage array sample, and we can also use the mean absolute percentage error (MAPE) as the precision evaluation loss function e MAPE , which measures the accuracy of the model prediction whose parameters are to be adjusted. The loss function (e RMSE 、e MAPE ) Curve such as Figure 7 As shown. According to the actual situation, the training cutoff condition of the model whose parameters are to be adjusted can be set, as shown in formula (8). According to the actual situation, the loss function threshold e can be set. set It is set to 0.0006, as shown in formula (11).

[0136] (11)

[0137] The error of the model with the parameters to be adjusted under the initial parameters =0.0316, =0.0188, both are greater than the loss function threshold, then the difference between the predicted voltage array sample and the target voltage array sample does not meet the condition. At this time, adjust the training parameters and set the adjustment interval of parameters w and b to 4. , The adjustment interval is set to 1, w, b, , The adjustment method is to increase the parameters by a corresponding interval each time, obtain the next training parameters and enter the next training.

[0138] Adaptive Moment Estimation (Adam) can be used as an adaptive optimizer to automatically adjust parameters until the loss function e RMSE and mean absolute percentage error are all less than or equal to the loss function threshold, indicating that the difference between the predicted voltage array sample and the target voltage array sample meets the condition, then the parameter adjustment is stopped, and the parameter adjustment model carrying the training parameters is used as the voltage mapping model, where the number of convolution kernels in the convolution layer in the training parameters is w. i =64, convolution kernel size b i =32, the number of filters in the pooling layer =3, the filter range of a single pooling layer =2.

[0139] Step S604, deploying the voltage mapping model to the relevant relay protection device, and obtaining the primary predicted voltage array of the distribution network according to the initial secondary voltage array of the distribution network and the voltage mapping model to determine whether to restore the voltage of the distribution network to a normal and stable state.

[0140] Specifically, an initial secondary voltage array of the distribution network is obtained; the initial secondary voltage array is ratio-converted and normalized to obtain a target secondary voltage array; a primary predicted voltage array of the distribution network is obtained based on the target secondary voltage array and a pre-built voltage mapping model; and based on the primary predicted voltage array, it is determined whether to restore the voltage of the distribution network to a normal and stable state.

[0141] At the single-phase metallic grounding fault point k1, the predicted voltage array sample is obtained. Array samples of target primary voltage The comparison is as follows Figure 8 As shown. Among them, the root mean square error e RMSE Only , mean absolute percentage error Only The prediction accuracy of the primary voltage array is high, the accuracy of judging whether the voltage of the distribution network is restored to a normal and stable state is high, and the transient transmission performance of the distribution network voltage is restored with good effect.

[0142] At the two-phase metallic grounding fault point k2, a predicted voltage array sample Target primary voltage array sample The comparison is as follows Fig. 9 As shown. Among them, the root mean square error e RMSE Only , mean absolute percentage error Only The prediction accuracy of the primary voltage array is high, the accuracy of judging whether the voltage of the distribution network is restored to a normal and stable state is high, and the transient transmission performance of the distribution network voltage is restored with good effect.

[0143] In this embodiment, a relatively accurate primary predicted voltage array can be obtained based on the target secondary voltage array and the pre-constructed voltage mapping model; based on the primary predicted voltage array, it can be determined whether to restore the voltage of the distribution network to a normal and stable state, thereby solving the problem of false operation of the relay protection device due to distortion of the secondary voltage signal, and requiring fewer computing resources, which can reduce the threat to the speed of the relay protection device.

[0144] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0145] Based on the same inventive concept, the embodiment of the present application also provides a distribution network voltage transient transmission performance recovery device for implementing the distribution network voltage transient transmission performance recovery method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more distribution network voltage transient transmission performance recovery device embodiments provided below can refer to the limitations of the distribution network voltage transient transmission performance recovery method above, and will not be repeated here.

[0146] In an exemplary embodiment, Fig.10 As shown, a distribution network voltage transient transmission performance recovery device is provided, wherein:

[0147] An initial secondary voltage acquisition module 1001 is used to acquire an initial secondary voltage array of a distribution network;

[0148] The target secondary voltage acquisition module 1002 is used to perform transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array;

[0149] A primary predicted voltage acquisition module 1003 is used to obtain a primary predicted voltage array of the distribution network according to the target secondary voltage array and a pre-built voltage mapping model;

[0150] The judgment recovery module 1004 is used to determine whether to restore the voltage of the distribution network to a normal and stable state according to the primary predicted voltage array.

[0151] In one of the embodiments, the target secondary voltage acquisition module 1002 is further used to: perform ratio conversion on the initial secondary voltage array according to the conversion ratio to obtain a secondary converted voltage array; determine the maximum value of the absolute value of the secondary converted voltage from the secondary converted voltage array; use the maximum value of the absolute value of the secondary converted voltage as a first reference value; and obtain a target secondary voltage array according to the secondary converted voltage array and the first reference value.

[0152] In one embodiment, the device further includes a model training module, which is used to: obtain target primary voltage array samples and target secondary voltage array samples; perform at least one training according to the target primary voltage array samples and the target secondary voltage array samples; wherein, during each training, the following steps are performed: through the convolution layer in the parameter to be adjusted model carrying the current training parameters, the target secondary voltage array samples are subjected to convolution operation and gradient operation to obtain a secondary voltage feature array; according to the secondary voltage feature array and the target primary voltage array samples, the imbalance of the convolution layer is obtained; through the pooling layer in the parameter to be adjusted model carrying the current training parameters and the imbalance of the convolution layer, the secondary voltage feature array is subjected to pooling operation to obtain a predicted voltage array sample; when the difference between the predicted voltage array sample and the target primary voltage array sample does not meet the condition, the current training parameter is adjusted to obtain the next training parameter and enter the next training; when the difference between the predicted voltage array sample and the target primary voltage array sample meets the condition, the parameter to be adjusted model carrying the current training parameters is used as the voltage mapping model.

[0153] In one of the embodiments, the model training module is also used to: model the transfer characteristics between the primary voltage and the secondary voltage of the capacitor voltage transformer in the distribution network to obtain an equivalent circuit model of the capacitor voltage transformer; obtain a transmission line fault simulation model of the distribution network; apply the capacitor voltage transformer equivalent circuit model to the transmission line simulation model, and obtain the primary voltage array samples and secondary voltage array samples of the capacitor voltage transformer when the distribution network line fails by setting different fault conditions and capacitor voltage transformer parameters; perform normalization calculation on the primary voltage array samples to obtain target primary voltage array samples; perform transformation ratio conversion and normalization calculation on the secondary voltage array samples to obtain target secondary voltage array samples.

[0154] In one of the embodiments, the model training module is also used to: determine the maximum value of the absolute value of the primary voltage from the primary voltage array samples; use the maximum value of the absolute value of the primary voltage as a second reference value; and obtain the target primary voltage array samples based on the primary voltage array samples and the second reference value.

[0155] In one of the embodiments, the judgment recovery module 1004 is also used to: when all the primary predicted voltages in the primary predicted voltage array are less than the fault threshold, determine to restore the distribution network voltage to a normal and stable state; when there is a primary predicted voltage in the primary predicted voltage array that is greater than the fault threshold, determine not to restore the distribution network voltage to a normal and stable state.

[0156] Each module in the above distribution network voltage transient transmission performance recovery device can be implemented in whole or in part by software, hardware and their combination. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0157] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.11 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data of a distribution network voltage transient transmission performance recovery method. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a distribution network voltage transient transmission performance recovery method is implemented.

[0158] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0159] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0160] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0161] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0162] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0163] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0164] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0165] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for restoring transient voltage performance of a distribution network, characterized in that: The method comprises: Get the initial secondary voltage array of the distribution network; Performing transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array; Obtaining a predicted primary voltage array of the distribution network according to the target secondary voltage array and a pre-built voltage mapping model; According to the primary predicted voltage array, it is determined whether to restore the voltage of the distribution network to a normal stable state.

2. The method according to claim 1, characterized in that: The step of performing transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array includes: According to the conversion ratio, the initial secondary voltage array is converted by transformation ratio to obtain a secondary converted voltage array; Determining the maximum value of the absolute value of the secondary converted voltage from the secondary converted voltage array; Taking the maximum value of the absolute value of the secondary converted voltage as the first reference value; A target secondary voltage array is obtained according to the secondary converted voltage array and the first reference value.

3. The method according to claim 1, characterized in that Before obtaining a predicted primary voltage array of the distribution network according to the target secondary voltage array and the pre-built voltage mapping model, the method further includes: Obtain target primary voltage array samples and target secondary voltage array samples; Performing at least one training according to the target primary voltage array samples and the target secondary voltage array samples; Among them, each time training, the following steps are performed: through the convolution layer in the parameter to be adjusted model carrying the training parameters of this time, the target secondary voltage array samples are subjected to convolution operation and gradient operation to obtain a secondary voltage feature array; according to the secondary voltage feature array and the target primary voltage array samples, the imbalance of the convolution layer is obtained; through the pooling layer in the parameter to be adjusted model carrying the training parameters of this time and the imbalance of the convolution layer, the secondary voltage feature array is subjected to pooling operation to obtain a first predicted voltage array sample; when the difference between the first predicted voltage array sample and the target primary voltage array sample does not meet the conditions, the training parameters of this time are adjusted to obtain the next training parameters and enter the next training; when the difference between the first predicted voltage array sample and the target primary voltage array sample meets the conditions, the parameter to be adjusted model carrying the training parameters of this time is used as the voltage mapping model.

4. The method according to claim 3, characterized in that The obtaining of target primary voltage array samples and target secondary voltage array samples comprises: The transfer characteristics between the primary voltage and the secondary voltage of the capacitor voltage transformer in the distribution network are modeled to obtain the equivalent circuit model of the capacitor voltage transformer; Obtaining a transmission line fault simulation model of a distribution network; Applying the capacitor voltage transformer equivalent circuit model to the power transmission line simulation model, and obtaining the primary voltage array sample and the secondary voltage array sample of the capacitor voltage transformer when the distribution network line fails by setting different fault conditions and capacitor voltage transformer parameters; Performing normalization calculation on the primary voltage array samples to obtain target primary voltage array samples; The secondary voltage array samples are subjected to transformation ratio conversion and normalization calculation to obtain target secondary voltage array samples.

5. The method according to claim 4, characterized in that The normalizing calculation of the primary voltage array samples to obtain target primary voltage array samples includes: Determine the maximum value of the absolute value of the primary voltage from the primary voltage array samples; Taking the maximum value of the absolute value of the primary voltage as the second reference value; A target primary voltage array sample is obtained according to the primary voltage array sample and the second reference value.

6. The method according to claim 1, characterized in that The step of determining whether to restore the distribution network voltage to a normal stable state according to the primary predicted voltage array includes: When all the primary predicted voltages in the primary predicted voltage array are less than the fault threshold, determining to restore the distribution network voltage to a normal stable state; When there is a primary predicted voltage in the primary predicted voltage array that is greater than a fault threshold, it is determined not to restore the distribution network voltage to a normal stable state.

7. A distribution network voltage transient transmission performance recovery device, characterized in that: The device comprises: An initial secondary voltage acquisition module is used to obtain an initial secondary voltage array of a distribution network; A target secondary voltage acquisition module is used to perform transformation ratio conversion and normalization calculation on the initial secondary voltage array to obtain a target secondary voltage array; A primary predicted voltage acquisition module, used to obtain a primary predicted voltage array of the distribution network according to the target secondary voltage array and a pre-built voltage mapping model; The judgment recovery module is used to determine whether to restore the voltage of the distribution network to a normal and stable state according to the primary predicted voltage array.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images