CN113326476A - Voltage sag type calculation method based on mixed criterion - Google Patents

Abstract

The invention provides a voltage sag type calculation method based on mixed criteria, which comprises the following steps: establishing a voltage sag type mode library according to the type of the voltage sag and the type conversion characteristic of the voltage sag transmitted by the transformer; constructing a voltage sag type mode library in a six-dimensional vector form; establishing a correlation coefficient matrix of each element in the voltage sag type pattern library in the form of the voltage sag to be calculated and a six-dimensional vector by applying a Pearson correlation coefficient, and measuring and calculating the similarity between the voltage to be matched and each characteristic voltage in the voltage sag type pattern library in the form of the six-dimensional vector; calculating a distance matrix of each vector in a voltage sag type mode library in a voltage sag and six-dimensional vector form to be calculated based on the Chebyshev distance; and constructing a voltage sag type correlation matrix, calculating the maximum correlation, and calculating the voltage sag type according to the maximum correlation. The invention can overcome the defect that the traditional method is sensitive to shallow sag and phase jump.

Description

A Calculation Method of Voltage Sag Types Based on Mixed Criteria

technical field

The invention relates to the technical field of power quality judgment, in particular to a voltage sag type calculation method based on a mixed criterion.

Background technique

With the rapid development of high-tech, the high-end manufacturing industry uses a large number of precision equipment such as programmable controllers and AC contactors, which are very sensitive to voltage sags. Due to voltage sags, precision equipment malfunctions, reports errors, and trips, causing huge economic losses to users, and users complain seriously. Identifying the type of voltage sag at the access point is the key to making grid-side and user-side governance decisions.

At present, various calculation methods of voltage sag types have been proposed by industry and academia. However, when the existing algorithm is applied to sag data with large phase jump or high amplitude, the calculation error rate is high, and the three-phase voltage vector is used as the input value. The six-phase method (SPA) and the symmetric component method (SCA) are the earliest proposed and easy-to-calculate algorithms, but the results are not ideal for shallow sags and large phase-jump sags. The TP-TA method and the TPA method are sensitive to the threshold, and there are many criterion formulas. The F-V method has a simple process, but is sensitive to large phase jumps. The RMS method uses the effective value of the three-phase voltage as the input value to calculate the sag type (A-G type), the SVA method uses the voltage waveform as the input quantity, and the calculation results add the H and I types of swells. The classification is more reasonable, but it has not yet been solved. The problem of large phase jump effects. In order to improve the accuracy of type calculation and improve the adaptability of the classification algorithm, the sensitivity of the calculation method to the sag amplitude and phase jump should be overcome.

Power grid faults are the main cause of voltage sags. During the operation of the power system, the weather, branches overlapping lines and other reasons may cause grid failures. When a fault occurs, the system draws a large current. According to the principle of line voltage division, the voltage near the fault will drop significantly, resulting in a voltage sag; after the fault is cleared, the voltage in the nearby area will return to normal. The three-phase short-circuit fault is the most serious short-circuit fault, which causes the three-phase voltage amplitude to drop basically the same; the inter-phase short-circuit fault will cause the faulted two-phase voltage to drop, and the non-faulty phase voltage amplitude is basically unchanged; the two-phase grounding short-circuit fault causes the fault The phase voltage decreases, the non-fault phase voltage remains unchanged or increases, and the non-fault phase voltage change mainly depends on the system grounding method; the fault phase voltage of a single-phase fault decreases, and the non-fault phase voltage changes are related to system grounding and other factors.

Voltage sag types are defined in terms of three-phase voltage amplitude and phase changes. Assuming that the per-unit value of the power supply electromotive force before the sag is 1p.u., the 9 types of voltage sag/swell are shown in Figure 1, where the dotted line represents the voltage vector before the sag, and the solid line represents the voltage vector during the sag. Swell is a phenomenon that occurs with sag under specific grounding conditions of the system, and is generally described in A-I types.

Transformers can be connected in many ways, the voltage is propagated from the high voltage side to the low voltage side, and different connection methods may lead to changes in the type of sag. Transformers can be classified into three categories according to the effect of sag type change: Type I (YNyn), Type II (Yy, Dd, Dz) and Type III (Yd, Dy, Yz). The sag type change characteristics caused by the connection mode of the transformer are described by the transfer matrix, as shown in equations (1)-(3).

(1) Class I: This type of transformer does not change the transfer of phase voltage and line voltage, and the transfer matrix is the unit matrix E, such as formula (1).

(2) Class II: This type of transformer does not transmit zero-sequence voltage, and the transfer matrix is as in formula (2).

(3) Class III: It is equivalent to the conversion of phase voltage to line voltage. After conversion, the voltage does not contain zero-sequence components, and phase changes will occur. The transfer matrix is as shown in formula (3).

Different types of voltage sags are propagated through different types of transformers, and the variation rules of the sag types are shown in Figure 2 and Table 1.

Table 1 Propagation law of voltage sag types

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to propose a voltage sag type calculation method based on mixed criteria, which has high accuracy and can overcome the shortcomings of traditional methods that are sensitive to shallow sags and phase jumps.

The present invention adopts the following scheme to realize: a kind of voltage sag type calculation method based on mixed criterion, which specifically includes the following steps:

According to the type of voltage sag and its type transformation characteristics propagating through the transformer, a voltage sag type pattern library is established;

Extract the real part and imaginary part of the three-phase voltage to form a voltage sag type pattern library in the form of a six-dimensional vector;

Use the Pearson correlation coefficient to establish the correlation coefficient matrix of each element in the voltage sag type pattern library in the form of a six-dimensional vector and the voltage sag to be calculated, respectively measure and calculate the voltage to be matched and the voltage sag type pattern library in the form of a six-dimensional vector The similarity between each characteristic voltage;

Calculate the distance matrix of the voltage sag to be calculated and each vector in the voltage sag type pattern library in the form of a six-dimensional vector based on the Chebyshev distance;

Build the voltage sag type correlation matrix, define the correlation index, calculate the maximum correlation, and calculate the voltage sag type according to the maximum correlation.

Further, the Pearson correlation coefficient is used to establish the correlation coefficient matrix of each element in the voltage sag type pattern library in the form of a voltage sag to be calculated and a six-dimensional vector, and respectively measure and calculate the voltage to be matched and the voltage in the form of a six-dimensional vector. The similarity between each characteristic voltage of the sag type pattern library is as follows:

Represent an element in the voltage sag type pattern library in the form of a six-dimensional vector as:

In the formula, V _s1 , V _s3 , and V _s5 represent the real part of the three-phase voltage of any sag data s in the model library, and V _s2 , V _s4 , and V _s6 represent the corresponding imaginary part of the three-phase voltage;

表示为：Convert a measured voltage sag vector to

Expressed as:

In the formula, V _m1 , V _m3 , and V _m5 represent the real part of the three-phase voltage of the measured voltage sag data m, and V _m2 , V _m4 , and V _m6 represent the corresponding imaginary part of the three-phase voltage;

与该某一元素之间的相关系数为：Then the measured voltage sag vector

The correlation coefficient with this element is:

和

分别为两组向量的平均值；In the formula,

and

are the averages of the two sets of vectors, respectively;

Using the above method to calculate the correlation coefficient matrix between the voltage to be measured and the mode library is:

In the formula, the subscript X of the correlation coefficient matrix represents different sag types, and ρ _Xij represents the correlation between the voltage sag to be calculated and the voltage sag of amplitude and phase jump in the X-type model library; the correlation coefficient The larger the value, the stronger the correlation, and the greater the similarity between the corresponding voltage vector to be calculated and the corresponding vector in the pattern library.

Further, the distance matrix of each vector in the voltage sag type pattern library in the form of the voltage sag to be calculated and the voltage sag type pattern library in the form of a six-dimensional vector is specifically calculated based on the Chebyshev distance:

和实测电压

的切比雪夫距离

的定义如下：A vector in the voltage sag type pattern library in the form of a six-dimensional vector

and measured voltage

Chebyshev distance

is defined as follows:

In the formula, k represents 1,2,3...,∞;

For the sag type X, the Chebyshev distance matrix of the voltage sag to be calculated and each vector of the pattern library is:

的电压暂降的切比雪夫距离。In the formula, m _Xij represents the voltage sag to be calculated and the amplitude of the X-type pattern library is |V _i |, and the phase is

The Chebyshev distance of the voltage sag.

Further, constructing a voltage sag type correlation matrix, defining a correlation index, and calculating a maximum correlation, and calculating the voltage sag type according to the maximum correlation is specifically:

For the sag type X, define the sag type correlation matrix K _X , each element in K _X is equal to the value obtained by dividing each element in the correlation coefficient matrix P _X and the corresponding element in the distance matrix M _X , as follows:

in,

的电压暂降的相关度；相关系数ρ越大，切比雪夫距离m越小，相关度越大，表示待计算暂降与某向量的匹配度越高。In the formula, k _Xij represents the amplitude of the sag to be calculated and the X type pattern library is |V _i |, and the phase jump becomes

The correlation degree of the voltage sag; the larger the correlation coefficient ρ, the smaller the Chebyshev distance m, and the greater the correlation degree, indicating that the sag to be calculated has a higher matching degree with a certain vector.

Find the maximum correlation between the sag to be calculated and each element in the K _X matrix:

k _X-max =max(K _X );

The maximum correlation degree is obtained for each dip type, the maximum value km _max among the maximum correlation degrees of all types is selected, and the related dip types are matched according to km _max .

The present invention also provides a voltage sag type computing system based on a hybrid criterion, comprising a memory, a processor, and a computer program instruction stored in the memory and capable of being executed by the processor. When the processor executes the computer program instruction, The method steps as described above can be implemented.

The present invention also provides a computer-readable storage medium on which computer program instructions that can be executed by a processor are stored, and when the processor executes the computer program instructions, the method steps described above can be implemented.

Compared with the prior art, the present invention has the following beneficial effects: the present invention is verified by the simulation data and the measured data, for the measured data, the algorithm can accurately calculate the sag type, which provides a basis for analyzing the influence of the sag event on the sensitive equipment. At the same time, based on the constructed type pattern library, the method of the present invention can realize millisecond-level fast calculation, is suitable for application in voltage sag monitoring systems or terminals, and calculates real-time information of sag types, and has great application value.

Description of drawings

FIG. 1 shows nine types of voltage sags/swells according to an embodiment of the present invention.

FIG. 2 is a transformation law of voltage sag types through a transformer according to an embodiment of the present invention.

FIG. 3 is a schematic flowchart of a method according to an embodiment of the present invention.

FIG. 4 is an IEEE14 node testing system according to an embodiment of the present invention.

FIG. 5 is a simulated voltage sag RMS according to an embodiment of the present invention. Among them, (a) is the curve of the RMS value of dip 1 over time, (b) is the curve of the RMS value of dip 2 over time, and (c) is the curve of the RMS value of dip 3 over time.

FIG. 6 is a schematic diagram of a simulated data voltage vector according to an embodiment of the present invention.

FIG. 7 is a schematic diagram of an actual measured data voltage vector according to an embodiment of the present invention.

Detailed ways

The present invention will be further described below with reference to the accompanying drawings and embodiments.

It should be noted that the following detailed description is exemplary and intended to provide further explanation of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

It should be noted that the terminology used herein is for the purpose of describing specific embodiments only, and is not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural as well, furthermore, it is to be understood that when the terms "comprising" and/or "including" are used in this specification, it indicates that There are features, steps, operations, devices, components and/or combinations thereof.

As shown in FIG. 3 , this embodiment provides a method for calculating a voltage sag type based on a mixed criterion, which specifically includes the following steps:

According to the type of voltage sag and its type transformation characteristics propagating through the transformer, a voltage sag type pattern library is established; the voltage sag type is determined by the characteristic voltage V and the phase jump value of each phase. The drop database covers 130,000 sets of voltage characteristics.

Extract the real part and imaginary part of the three-phase voltage to form a voltage sag type pattern library in the form of a six-dimensional vector;

Use the Pearson correlation coefficient to establish the correlation coefficient matrix of each element in the voltage sag type pattern library in the form of a six-dimensional vector and the voltage sag to be calculated, respectively measure and calculate the voltage to be matched and the voltage sag type pattern library in the form of a six-dimensional vector The similarity between each characteristic voltage;

Calculate the distance matrix of the voltage sag to be calculated and each vector in the voltage sag type pattern library in the form of a six-dimensional vector based on the Chebyshev distance;

Build the voltage sag type correlation matrix, define the correlation index, calculate the maximum correlation, and calculate the voltage sag type according to the maximum correlation.

Preferably, in order to accurately identify the type of voltage sag, its three-phase amplitude and phase jump value are calculated respectively to form a 6-dimensional vector. Among them, types F and G are the distortion forms of types C and D; most users use delta connection to connect to the power grid, so type B and E sags generally do not propagate to the user side. Therefore, A, C, D, H, I types are the targets for type calculation.

(1) Characteristic voltage V

The characteristic voltage V is a single-shot characteristic of voltage sag proposed by IEEE 1564-2014 to describe the degree and type of voltage sag. Different from the definition of residual voltage, the characteristic voltage V is represented by the lowest value of the six-phase voltage. The six-phase voltage includes three-phase drop voltage and three-phase line voltage. The curve of V ₀ (t) zero-sequence voltage changing with time can be obtained from equation (4).

Among them, V _a (t), V _b (t), and V _c are the curves of the three-phase phase voltages changing with time, respectively. Defining the drop voltage as the phase voltage minus the zero-sequence voltage, the three-phase drop voltage can be obtained as:

In order to keep the magnitude of the three-phase line voltage consistent with the three-phase drop voltage, let the three-phase line voltage be:

In (5) and (6), there are six-phase voltages in total, of which the lowest phase voltage is the characteristic voltage V(t), which represents the characteristic voltage V of the event in the form of a vector. Defining positive and negative sequence factors F, V and F can characterize the type of voltage sag. Taking the C _a type sag in Figure 1 as an example, its voltage expression is:

Under various short-circuit conditions, the positive and negative sequence factors F are almost always close to 1; the sag caused by motor startup may cause F to be as low as 0.7. Consider the sag type calculation problem caused by different short-circuit faults, therefore, set the value of F in all type expressions to be 1.

To fully describe all possible voltage sags, the characteristic voltage V is constructed. The dip amplitude ranges from 0.01 to 1, and 100 different amplitudes are set at equal intervals. The phase jump range does not have to consider the full range of -180° to +180°. When the system X/R impedance ratio is 10 and the feeder impedance ratio X/R is 0.5, the impedance angle has a maximum negative value, which is about -60°. If the X/R of the system and the feeder are almost equal, the impedance angle has a minimum positive value, which is about 10°. Therefore, in most cases, the actual impedance angle is between -60° and +10°. Based on the above reasons, set 100 different phases, consider a phase margin of 20°, range from -80° to +30°, step size interval is 110°/100, and combine the amplitude and phase one by one to get 100*100 characteristic voltages of different amplitudes and phases, the composition model of V is shown below.

Among them, i and j are the numbers of each group, and equations (9) and (10) represent the amplitude and phase of the characteristic voltage.

(2) Type Pattern Library

Generate A, C, D, H, I type voltage sag data to build type pattern library. Among them, type A sags are three-phase sags; the last four types are asymmetric sags, and each type of sags can be characterized by phases a, b, and c to generate data. Based on the characteristic voltage constructed above, the characteristic voltage is substituted into the A, C, D, H, and I types of voltage sag expressions in Figure 1 to generate voltage sag data and build a type pattern library. Among them, type A is a type of sag, and C, D, H, and I can construct 12 types of sags with three phases a, b, and c as the characteristic phases, forming 13*10,000 sets of voltage sag data. As in formula (11), the real and imaginary parts of each sag data are extracted to form a set of 6-dimensional vectors.

In the formula, V _s1 , V _s3 , and V _s5 represent real parts, and V _s2 , V _s4 , and V _s6 represent imaginary parts. For each type of dip, 100*100 sets of 6-dimensional vectors can be formed. Taking type D _a as an example, each element in equation (12) represents a 6-dimensional vector of sag data, such as V _Da-1-2 , which is the characteristic voltage formed when the amplitude is 0.01pu and the phase is -78.89° V is brought into the D _a class sag formula in Figure 1, and the obtained 6-dimensional vector of the real part and imaginary part of each phase voltage of the three-phase voltage is obtained.

13 types of voltage sag data, forming a total of 130,000 groups of 6-dimensional vectors, constitute a voltage sag type pattern library.

In this embodiment, to identify the type of a certain piece of voltage sag data, its 6-dimensional feature can be extracted to form a 6-dimensional vector such as formula (11), and the similarity of the 6-dimensional vector of the piece of data with each vector in the pattern library can be compared. , to determine its type. The invention proposes a dual criterion based on the Pearson correlation coefficient and the Chebyshev distance, measures the similarity between the voltage sag data to be identified and each sag data in the pattern library, defines the maximum correlation index, and calculates the voltage sag. type.

Among them, the Pearson correlation coefficient is used to measure the degree of correlation between two variables. The larger the correlation coefficient, the stronger the correlation between the two variables, and vice versa. The Pearson correlation coefficient is used to measure the degree of correlation between two variables. The larger the correlation coefficient, the stronger the correlation between the two variables, and vice versa.

和

分别为变量X、Y的均值。in,

and

are the mean values of the variables X and Y, respectively.

In this embodiment, the Pearson correlation coefficient is used to establish the correlation coefficient matrix of the voltage sag to be calculated and each element in the voltage sag type pattern library in the form of a six-dimensional vector, and measure and calculate the voltage to be matched and the six-dimensional vector respectively. The similarity between each characteristic voltage of the voltage sag type pattern library of the form is as follows:

Represent an element in the voltage sag type pattern library in the form of a six-dimensional vector as:

In the formula, V _s1 , V _s3 , and V _s5 represent the real part of the three-phase voltage of any sag data s in the model library, and V _s2 , V _s4 , and V _s6 represent the corresponding imaginary part of the three-phase voltage;

表示为：Convert a measured voltage sag vector to

Expressed as:

In the formula, V _m1 , V _m3 , and V _m5 represent the real part of the three-phase voltage of the measured voltage sag data m, and V _m2 , V _m4 , and V _m6 represent the corresponding imaginary part of the three-phase voltage;

与该某一元素之间的相关系数为：Then the measured voltage sag vector

The correlation coefficient with this element is:

和

分别为两组向量的平均值；In the formula,

and

are the averages of the two sets of vectors, respectively;

Using the above method to calculate the correlation coefficient matrix between the voltage to be measured and the mode library is:

的电压暂降的相关性；相关系数越大，表明对应的待计算电压向量与模式库中对应向量的相似度越大。In the formula, the subscript X of the correlation coefficient matrix represents different sag types, namely A, C _a , C _b , C _c , D _a , D _b , D _c , H _a , H _b , H _c , I _a , I _b , I _c type. ρ _Xij represents the voltage sag to be calculated and the amplitude value in the X-type mode library is |V _i |, and the phase jump becomes

The correlation of the voltage sag of , the larger the correlation coefficient, the greater the similarity between the corresponding to-be-calculated voltage vector and the corresponding vector in the pattern library.

In addition, the Chebyshev distance is used to measure the similarity of two samples. The larger the measurement result, the less similar the two samples are; the smaller the distance, the opposite. In this embodiment, the distance matrix of each vector in the voltage sag type pattern library in the form of the voltage sag to be calculated and the voltage sag type pattern library in the form of a six-dimensional vector calculated based on the Chebyshev distance is specifically:

和实测电压

的切比雪夫距离

的定义如下：A vector in the voltage sag type pattern library in the form of a six-dimensional vector

and measured voltage

Chebyshev distance

is defined as follows:

In the formula, k represents 1,2,3...,∞;

For the sag type X, the Chebyshev distance matrix of the voltage sag to be calculated and each vector of the pattern library is:

的电压暂降的切比雪夫距离；例如m_X-100-3表示待计算的电压暂降与X类型模式库中幅值为1、相位为-78°的电压暂降的切比雪夫距离。In the formula, m _Xij represents the voltage sag to be calculated and the amplitude of the X-type pattern library is |V _i |, and the phase is

The Chebyshev distance of the voltage sag; for example, m _X-100-3 represents the Chebyshev distance between the voltage sag to be calculated and the voltage sag with an amplitude of 1 and a phase of -78° in the X-type pattern library.

The larger the Chebyshev distance, the smaller the similarity between the corresponding voltage vector to be calculated and the corresponding vector in the pattern library.

Using the correlation coefficient matrix P _X and the Chebyshev distance matrix M _X obtained above, the correlation index can be defined as the correlation coefficient divided by the corresponding Chebyshev distance, which has the advantage of reducing the uncertainty caused by a single criterion, Improve calculation accuracy. In this embodiment, the voltage sag type correlation matrix is constructed, the correlation index is defined, and the maximum correlation is calculated, and the voltage sag type calculated according to the maximum correlation is specifically:

For the sag type X, define the sag type correlation matrix K _X , each element in K _X is equal to the value obtained by dividing each element in the correlation coefficient matrix P _X and the corresponding element in the distance matrix M _X , as follows:

in,

的电压暂降的相关度；例如k_X-100-3表示待计算暂降与X类型模式库中幅值为1、相位跳变为-78°的电压暂降的相关度。相关系数ρ越大，切比雪夫距离m越小，相关度越大，表示待计算暂降与某向量的匹配度越高。In the formula, k _Xij represents the amplitude of the sag to be calculated and the X type pattern library is |V _i |, and the phase jump becomes

For example, k _X-100-3 represents the correlation between the sag to be calculated and the voltage sag whose amplitude is 1 and whose phase jump is -78° in the X-type pattern library. The larger the correlation coefficient ρ is, the smaller the Chebyshev distance m is, and the larger the correlation is, which means that the matching degree between the sag to be calculated and a certain vector is higher.

Find the maximum correlation between the sag to be calculated and each element in the K _X matrix:

k _X-max =max(K _X );

The maximum correlation degree is obtained for each dip type, the maximum value km _max among the maximum correlation degrees of all types is selected, and the related dip types are matched according to km _max . Apply to the rest of the sag types, denote the k _X-max obtained by each sag type as k _A , k _Ca ,..., k _Ic a total of 13, and extract the maximum value among the 13 correlations .

k _max =max(k _A ,k _Ba ,k _Bb ,...,k _Ic ) (23)

According to the maximum value among the 13 correlations, the relevant dip types are matched, and the matching results are shown in Table 2 below.

Table 2 Voltage sag type matching results

This embodiment also provides a hybrid criterion-based voltage sag type computing system, including a memory, a processor, and computer program instructions stored on the memory and capable of being executed by the processor, when the processor executes the computer program instructions , the method steps as described above can be implemented.

This embodiment also provides a computer-readable storage medium on which computer program instructions that can be executed by a processor are stored, and when the processor executes the computer program instructions, the method steps described above can be implemented.

Next, simulation experiments are carried out in this embodiment to further illustrate the effectiveness of the proposed method.

First of all, this embodiment describes the experimental data as follows:

(1) Simulation data 1.

Perform simulation according to known types of voltage sags and construct multiple sag data. In order to avoid duplication of simulation data and pattern library data, the amplitude and phase of characteristic voltages are adjusted to [0.1, 0.9] and [-40, 10] , the number of samples is 50*50, and 2500 sets of simulation data are formed. At the same time, considering the change of the standard voltage, the cases of F=1, F=0.9 and F=0.8 are taken for verification. The results show that the proposed algorithm has high accuracy.

Table 3 Simulation data 1 Calculation results of voltage sag types

In order to verify the effect of the proposed hybrid criterion, the voltage sag type calculation is carried out under a single criterion that only considers the Pearson correlation coefficient and only the Chebyshev distance. For the former, in the correlation coefficient matrix, the correlation coefficient between the measured voltage sag data and a certain sag type is the largest, and this voltage sag is the corresponding type. For the latter, in the Chebyshev distance matrix, the Chebyshev distance between the measured voltage sag data and a certain sag type is the smallest, and this voltage sag is the corresponding type.

The calculation results are shown in Table 3. When the positive and negative sequence factor F is the standard voltage of 1, for all simulation data, the correct rate of the proposed method is 100%. When the positive and negative sequence factor F is not the standard voltage of 0.9 and 0.8, it still has a high recognition accuracy, and the accuracy is higher than 95%. Considering the case of F = 1 and 0.9, only using the Pearson correlation coefficient or the Chebyshev distance as the criterion, the accuracy of the calculation result of the voltage sag type is about 75% or 80%, which is much lower than the calculation of the mixed criterion Effect.

(2) Simulation data 2.

Build an IEEE 14 node network (as shown in Figure 4) on the PSCAD/EMTDC platform, set two different types of voltage sags, set a two-phase fault and a three-phase fault between the No. 4 bus and No. 9 bus, fault 1 It is a two-phase grounding short-circuit fault of A and B between lines 4-9. The fault start time is 1s, and the fault clearing time is 4s; fault 2 is a three-phase grounding short-circuit fault that occurs between lines 4-9. , the start time is 5.6s, and the fault clearing time is 6.8s; fault 3 is a two-phase grounding fault of A and C between lines 4-9, the start time is 8.6s, and the end time is 10s. The RMS values of the three dip waveforms are shown in Figure 5. Knowing the specific fault types, it can be inferred that the monitored sag types are Class _C , Class A, and Class I _b , respectively.

Its three-phase amplitude and phase are shown in Figure 6:

The monitoring data in Figure 5 was classified by the proposed method, and its maximum correlation was calculated. The results are listed in Table 4. It can be seen from the calculation results that the sag types most related to sag 1, sag 2, and sag 3 are Class C _c , Class A, and Class I _b , respectively. The calculated results are consistent with the actual results, proving the effectiveness of the proposed method.

Table 4 The calculation results of the maximum correlation degree of the simulation data

Next, this embodiment verifies the measured data.

Four sag events in 2019 in the Fuzhou power quality monitoring system were selected for analysis to verify the proposed method. The calculation is performed on the MATLAB/Simulink platform, and the duration of each calculation does not exceed 1s. Calculate the phase amplitude and phase jump value of the four dips, and the vector diagram is shown in Figure 7. The proposed method is used to calculate the similarity between the four groups of measured voltage sag data and various sags in the model library. The results are shown in Table 5.

Table 5 Calculation results of maximum correlation degree of measured data

data 1 data 2 data 3 data 4 k_A 0.937 1.535 2.032 9.788 k_Ca 0.910 1.928 9.107 3.628 k_Cb 1.192 1.805 1.355 6.639 k_Cc 0.916 1.086 1.538 4.283 k_Da 1.142 1.203 1.238 6.176 k_Db 0.886 1.267 2.952 3.546 k_Dc 1.006 2.868 2.234 4.589 k_Ha 0.933 1.115 1.230 3.292 k_Hb 0.982 1.119 1.299 3.178 k_Hc 1.463 1.585 1.281 3.229 k_Ia 1.025 1.214 1.434 3.178 k_Ib 1.129 1.228 1.486 3.191 k_Ic 0.879 1.100 1.341 3.268

According to the calculation result, find the maximum correlation of each column, and the corresponding dip type is the calculation result.

For the measured data, there is currently no standard algorithm for calculation in the monitoring series, and only visual judgment can be used to estimate the actual type and compare it with the calculation results. Taking data 2 as an example, phase b has not fallen below the sag threshold specified by the standard, phase a is almost equal to the threshold, it can be considered that phases a and b have not fallen, and phase c has dropped significantly to 0.274pu. Visually, waveform 2 is a typical D. Class _c voltage sag, the calculation result of the proposed method is correct. Similar judgments can be made for data 3. The a phase has almost no drop, the _b and c phases drop to 0.682pu and 0.513pu respectively, and the b phase has a large phase jump. Visual judgment.

In order to verify the advanced nature of the proposed method, the calculation results of the proposed method are compared with the calculation results of several mainstream classification algorithms in the world. Using the SCA method, the SPA method, the SVA method and the TP-TA method, the sag types of data 1-data 4 were calculated, and the comparison results are shown in Table 6.

Table 6 Measured data Voltage sag types

Sag Type data 1 data 2 data 3 data 4 Visual results H_c D_c C_a A The proposed method H_c D_c C_a A SCA Act - D_c C_a - SPA law - C_b D_c - SVA method I_ac C_b D_c A TP-TA method - D_c C_c -

Since the SCA method, the SPA method and the TP-TA method do not take into account the voltage swell and the symmetric sag, and the actual types of the data 1 and 4 are the H _c -type swell and the A-type symmetric sag, respectively, the Three methods cannot calculate the dip type for data 1 and 4. Using the SPA method and the SVA method to calculate the sag type of data 2, the results are all C _b type, which is obviously wrong; the sag type of data 3 is calculated, only the result of the SCA method is correct. The phase jump of data 2 and data 3 is the main reason for the error of the existing method. It can be seen from the comparison that the method proposed in this patent can effectively overcome the misjudgment caused by the phase jump, and has a good application prospect.

In summary, in this embodiment, the voltage sag caused by the startup of the large motor and the excitation of the transformer is a slow recovery type sag, and the voltage amplitude is constantly changing during the sag process, and it is difficult to define it by the ABC classification method. A calculation method of voltage sag types based on Pearson correlation coefficient and Chebyshev distance is presented, and the proposed algorithm is suitable for rectangular sags caused by grid faults. By converting the three-phase voltage containing amplitude and phase information into a six-dimensional vector that is easy to calculate, a six-dimensional vector pattern library is established, the maximum correlation of the vectors is calculated, and the type of voltage sag is identified. This patent is verified by simulation data and measured data. For the measured data, the algorithm can accurately calculate the sag type, which provides a basis for analyzing the impact of sag events on sensitive equipment. When the positive and negative sequence factor F is the standard voltage, the accuracy of the proposed method is 100% for all simulation data. When the positive and negative sequence factor F is not a standard voltage, it still has a high recognition accuracy, and the accuracy is higher than 95%. The proposed method has high accuracy and can overcome the shortcomings of traditional methods that are sensitive to shallow dips and phase jumps.

As will be appreciated by those skilled in the art, the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

The above are only preferred embodiments of the present invention, and are not intended to limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications to equivalent changes. Example. However, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solutions of the present invention still belong to the protection scope of the technical solutions of the present invention.

Claims (6)

1. A voltage sag type calculation method based on a mixed criterion is characterized by comprising the following steps:

establishing a voltage sag type mode library according to the type of the voltage sag and the type conversion characteristic of the voltage sag transmitted by the transformer;

extracting the real part and the imaginary part of the three-phase voltage to form a voltage sag type mode library in a six-dimensional vector form;

establishing a correlation coefficient matrix of each element in the voltage sag type pattern library in the form of the voltage sag to be calculated and a six-dimensional vector by applying a Pearson correlation coefficient, and measuring and calculating the similarity between the voltage to be matched and each characteristic voltage in the voltage sag type pattern library in the form of the six-dimensional vector;

calculating a distance matrix of each vector in a voltage sag type mode library in a voltage sag and six-dimensional vector form to be calculated based on the Chebyshev distance;

and constructing a voltage sag type correlation matrix, defining a correlation index, calculating the maximum correlation, and calculating the voltage sag type according to the maximum correlation.

2. The voltage sag type calculation method based on the mixing criterion according to claim 1, wherein the Pearson correlation coefficient is applied to establish a correlation coefficient matrix of each element in the voltage sag to be calculated and the voltage sag type pattern library in a six-dimensional vector form, and the similarity between the voltage to be matched and each characteristic voltage in the voltage sag type pattern library in the six-dimensional vector form is measured and calculated respectively as follows:

representing a certain element in the six-dimensional vector form voltage sag type pattern library as:

in the formula, V_s1、V_s3、V_s5Three-phase voltage real part, V, representing any sag data s in the pattern library_s2、V_s4、V_s6Representing the corresponding imaginary three-phase voltage components;

a certain measured voltage sag vector

Expressed as:

in the formula, V_m1、V_m3、V_m5Three-phase voltage real part, V, representing the measured voltage sag data m_m2、V_m4、V_m6Representing the corresponding imaginary three-phase voltage components;

the measured voltage sag vector

The correlation coefficient with the certain element is:

in the formula,

and

respectively the average values of the two groups of vectors;

the correlation coefficient matrix of the voltage to be measured and the pattern library is obtained by calculation by adopting the method as follows:

in which the subscript X of the matrix of correlation coefficients represents the different dip types, p_X-i-jRepresenting the voltage sag to be calculated and the magnitude of | V in the X-type pattern library_iI, phase jump to

The voltage sag dependency of; the larger the correlation coefficient is, the stronger the correlation is, which indicates that the similarity between the corresponding voltage vector to be calculated and the corresponding vector in the pattern library is larger.

3. The voltage sag type calculation method based on the mixing criterion according to claim 1, wherein the distance matrix of each vector in the voltage sag type pattern library to be calculated based on the Chebyshev distance calculation and the six-dimensional vector form is specifically:

certain vector in voltage sag type pattern library in six-dimensional vector form

And a measured voltage

Chebyshev distance of

Is defined as follows:

wherein k represents 1,2,3., ∞;

for the sag type X, the chebyshev distance matrix of the voltage sag to be calculated and each vector of the pattern library is:

in the formula, m_X-i-jThe representation represents the voltage sag to be calculated and the magnitude of | V in the X-type pattern library_iIn phase of

The chebyshev distance of the voltage sag.

4. The voltage sag type calculation method based on the mixing criterion according to claim 1, wherein the step of constructing a voltage sag type correlation matrix, defining a correlation index, and calculating the maximum correlation comprises the following steps of:

for sag type X, a sag type correlation matrix K is defined_X，K_XEach element in (1) is equal to the moment of the correlation coefficientArray P_XMatrix M of elements and distances_XThe value obtained by dividing the corresponding element in (1) is as follows:

wherein,

in the formula, k_X-i-jRepresenting the sag to be calculated and the amplitude of | V in the X-type mode library_iI, phase jump to

The correlation of voltage sag of; the larger the correlation coefficient rho is, the smaller the Chebyshev distance m is, the larger the correlation degree is, and the higher the matching degree of the sag to be calculated and a certain vector is;

calculating the sag and K to be calculated_XMaximum correlation of each element in the matrix:

k_X-max＝max(K_X)；

the maximum correlation degree of each sag type is obtained, and the maximum value k in the maximum correlation degrees of all types is selected_maxAnd according to k_maxThe associated dip type is matched.

5. A voltage sag type calculation system based on a mixing criterion, comprising a memory, a processor and computer program instructions stored on the memory and executable by the processor, the computer program instructions, when executed by the processor, being capable of implementing the method steps of any one of claims 1 to 4.

6. A computer-readable storage medium, having stored thereon computer program instructions executable by a processor, the computer program instructions, when executed by the processor, being capable of carrying out the method steps according to any one of claims 1 to 4.

Images