Equivalent modeling method for operation evaluation model of new energy station

Abstract

The invention discloses a new energy station operation evaluation model equivalent modeling method, and belongs to the field of power systems. The problem that in the existing equivalent modeling process of the new energy station, the identifiability of data is too low is solved. The method comprises the steps of extracting initial variable indexes of a new energy unit; establishing a factor analysis model of the initial variable indexes; obtaining factor score data as a grouping index; performing similarity measurement on the grouping indexes by using correlation analysis and significance test, and obtaining a correlation relationship of the new energy units to complete group division; and calculating equivalent parameters based on a grouping result, and establishing a new energy station equivalent model. The method is mainly used for carrying out equivalent modeling on the new energy station.

Description

Technical Field

[0001] The invention belongs to the field of power systems. Background Art

[0002] A new energy station often has dozens or even hundreds of units. If a transient model is established for each unit separately, the complexity of the power system simulation model and the simulation calculation time will be greatly increased, and even the problem of "dimensionality curse" will be faced. Therefore, it is necessary to establish an equivalent model of the new energy station in order to evaluate the stable operation of the station.

[0003] In the equivalent modeling of new energy stations, the selected initial variable index is a multidimensional time series. When the initial variable index of all new energy units forms a high-dimensional matrix, the identifiability of the data is too low. The artificially selected temporal sequence and the uncertainty of their own weights make the clustering index of the entire initial variable too abstract, and its internal logic cannot be observed, thereby weakening the objectivity and authenticity of the clustering results. To this end, in view of the problem of low identifiability of the initial variable index, certain measures need to be taken to fully explore the potential information of the variable index, find out the mapping relationship between the index and the operating status of the new energy unit, reasonably quantify the clustering index, and enhance the identifiability and guiding role of the index. Summary of the invention

[0004] The purpose of the present invention is to solve the problem of low data identifiability in the equivalent modeling process of existing new energy stations, and to provide an equivalent modeling method for a new energy station operation evaluation model.

[0005] The equivalent modeling method of the new energy station operation evaluation model comprises the following steps:

[0006] Extract the initial variable indicators of new energy units;

[0007] Establish factor analysis model of initial variable indicators;

[0008] Obtain factor score data as a clustering indicator;

[0009] Use correlation analysis and significance test to measure the similarity of clustering indicators, obtain the correlation of new energy units and complete the cluster division;

[0010] Based on the clustering results, the equivalent parameters are calculated, and the equivalent model of the new energy station is established to complete the modeling.

[0011] Preferably, the initial variable index of the new energy unit includes four output external characteristics P, Q, i d 、i q Transient and steady-state operation data;

[0012] Among them, P, Q, id 、i q They are respectively the active response characteristics of the new energy unit, the reactive response characteristics index, the direct axis component current index of the motor and the quadrature axis component current index of the motor; the four output external characteristics P, Q, i d 、i q The selection basis of transient and steady-state operation data is as follows:

[0013] During the initial steady state, select P, i at a time point d ;

[0014] During the transition period of the protection circuit action, select P, Q, and i at two time points d 、i q ;

[0015] During the transient stability period of voltage drop, select P, Q, and i at two time points d 、i q ;

[0016] Select P, i at a time point during the steady state after the fault is cleared d ;

[0017] Finally, the initial variable indicators of the 20-dimensional new energy unit are obtained.

[0018] Preferably, the factor analysis model is: X = AF;

[0019] X is the standardized initial variable index, F=(F1, F2,…, F m ) T is a common factor vector, the number of common factors is m; A=(a ik ) 21×m is the factor loading matrix, a ik represents the i-th dimension variable X in X i In the kth common factor F k The factor loading on ik The larger the absolute value of X i With F k The greater the degree of correlation;

[0020]

[0021] Where: λ1 to λ m is the first m largest characteristic roots of the correlation matrix R of the initial variable index, λ1 to λ m are the 1st to mth largest characteristic roots respectively; λ1≥λ2≥…λ m >0,u1,u2,…,u m is the characteristic root corresponding to λ1,λ2,…,λ m The feature vector of .

[0022] Preferably, the number of common factors m is determined by the cumulative variance contribution rate Sure.

[0023] Preferably, the expression for obtaining factor score data is:

[0024]

[0025] Where: is the factor score data of all common factors in the common factor vector at each sample point; A′ is the rotated factor loading matrix; R is the correlation matrix of the standardized initial variable indicator X.

[0026] Preferably, the similarity measurement of the clustering indicators is performed using correlation analysis and significance test, and the implementation method of obtaining the correlation relationship of the new energy units to complete the clustering is:

[0027] Firstly, the correlation coefficient of the factor score data of the new energy units is obtained by using correlation analysis;

[0028] Then, the significance test of the correlation coefficient is carried out to obtain the correlation between the new energy units; among which, the correlation is strong correlation or not strong correlation;

[0029] Finally, by using the correlation between any two new energy units, the strongly correlated new energy units are divided into one group, and the two non-strongly correlated new energy units are divided into different groups, and finally the division of all new energy units in the photovoltaic power station is completed.

[0030] Preferably, the expression of the correlation coefficient is:

[0031]

[0032] In the formula, is the factor score data of the hth common factor in the common factor vector F at each sample point, is the factor score data of the jth common factor in the common factor vector F at each sample point;

[0033] Consider each factor score data of the new energy unit as a sample, c hj For sample and samples The correlation coefficient of For sample and samples The covariance of The samples are and samples The variance of .

[0034] Preferably, equivalent parameters are calculated based on the clustering results to establish an equivalent model of new energy stations. The implementation methods for completing the modeling include:

[0035] After the original n new energy units in the new energy station are divided into s groups, s<n, the new energy units in each group are aggregated into a single new energy unit through single-machine equivalence, and the aggregated single new energy unit is used as an equivalent unit. At the same time, the capacity weighting method and parameter identification method are used to determine the equivalent parameters, so as to realize the equivalent model of the new energy station containing n new energy units into the equivalent model of the new energy station containing only s new energy units, and complete the modeling;

[0036] And the principles followed in the modeling process are:

[0037] a. The total rated capacity of the equivalent model and detailed model of the new energy station is the same;

[0038] b. The grid connection point voltage of the equivalent model and detailed model of the new energy station is the same;

[0039] c. The active power and reactive power output by the equivalent model and detailed model of the new energy station are the same.

[0040] Preferably, the implementation of determining equivalent parameters using the capacity weighting method and the parameter identification method includes:

[0041] The equivalent parameters of the generator and transformer are:

[0042]

[0043] Where: S Geq , S Teq , Z Geq and Z Teq are the apparent power of the equivalent unit, the apparent power of the machine-end transformer, the generator impedance and the machine-end transformer impedance respectively; S Gh and Z Gh are the apparent power and generator impedance of wind turbine h respectively; S Th and Z Th are the apparent power and impedance of the transformer at the machine end of wind turbine h, respectively; f is the number of wind turbines included in the equivalent unit, 1≤h≤f;

[0044] The equivalent parameters of the axis system are:

[0045]

[0046] Where: H teq , H geq , K eq and D eqare the wind turbine inertia time constant, generator rotor inertia time constant, shaft stiffness coefficient and shaft damping coefficient of the equivalent unit respectively; H th , H gh , K h and D h are the wind turbine inertia time constant, generator rotor inertia time constant, shaft stiffness coefficient and shaft damping coefficient of wind turbine h respectively;

[0047] Wind speed and other parameters are:

[0048]

[0049] Where: v eq , A and c peq are the input wind speed, wind turbine swept area and wind energy utilization coefficient of the equivalent unit respectively; v h , A h and c ph are the input wind speed, wind turbine swept area and wind energy utilization coefficient of wind turbine h respectively;

[0050] The equivalent parameters of the photovoltaic array are:

[0051]

[0052] Where: N seq 、N peq are the number of series and parallel cells in the photovoltaic array in the equivalent model; N s 、N p are the number of batteries in series and parallel connection of a single photovoltaic array in the detailed model; λ is the coefficient;

[0053] The equivalent light intensity is obtained by taking the weighted average of the light intensity of each photovoltaic power generation unit;

[0054] The equivalent parameters of the photovoltaic inverter are:

[0055]

[0056] Where: S eq , Z ceq are the capacity and equivalent impedance of the inverter in the equivalent model; S, Z c are the capacity and impedance of a single PV inverter in the detailed model, respectively;

[0057] The capacitance and inductance parameters of the AC side filter are:

[0058]

[0059] Where: C eq , L eqare the filter capacitor and filter inductor on the AC side of the inverter in the equivalent model; C and L are the filter capacitor and filter inductor on the AC side of a single inverter in the detailed model;

[0060] The equivalent parameters of the collector line are:

[0061]

[0062] Where: Z i represents the line impedance at the outlet of the i-th unit; P i The impedance Z i Power; Z eqm′ Represents the equivalent line impedance of the m′th unit branch, i≤m′≤n.

[0063] A new energy station operation evaluation model equivalent modeling device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the new energy station operation evaluation model equivalent modeling method.

[0064] Advantages of the present invention:

[0065] In the equivalent modeling method of the new energy station operation evaluation model proposed in the present invention, the strong correlation and unidentifiability of the initial variable indicators of the new energy units selected in the equivalent modeling will affect the accuracy of the clustering results. By performing factor analysis on the initial variable indicators, the complex and redundant original data can be converted into a few common factors with practical significance. Compared with the initial variable data of the new energy units, the reliability and validity of the factor score data have been improved. Using it as the clustering indicator of the new energy units for clustering can improve the accuracy of the clustering results; at the same time, the mapping relationship reflected by the factor data can enhance the identifiability and guiding role of the indicators and reduce the complexity of the problem analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a flow chart of the equivalent modeling method of the new energy station operation evaluation model of the present invention;

[0067] Figure 2 This is a schematic diagram of the topological structure of the collector line of a new energy station, where: Figure 2 a is a schematic diagram of radial topology structure. Figure 2 b is a schematic diagram of trunk topology;

[0068] Figure 3 Schematic diagram of the conversion from trunk topology to radial topology. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0071] Specific implementation method 1. Combination Figure 1 As shown, the equivalent modeling method of the new energy station operation evaluation model described in this embodiment includes the following steps:

[0072] Extract the initial variable indicators of new energy units;

[0073] Establish factor analysis model of initial variable indicators;

[0074] Obtain factor score data as a clustering indicator;

[0075] Use correlation analysis and significance test to measure the similarity of clustering indicators, obtain the correlation of new energy units and complete the cluster division;

[0076] Based on the clustering results, the equivalent parameters are calculated, and the equivalent model of the new energy station is established to complete the modeling.

[0077] In this embodiment, in equivalent modeling, the strong correlation and unidentifiability of the selected initial variable indicators of the new energy units will affect the accuracy of the clustering results. By performing factor analysis on the initial variable indicators, the complex and redundant original data can be converted into a few common factors with practical significance. Compared with the initial variable data of the new energy units, the reliability and effectiveness of the factor score data have been improved. Using it as a clustering indicator for the new energy units can improve the accuracy of the clustering results; at the same time, the mapping relationship reflected by the factor data can enhance the identifiability and guiding role of the indicators and reduce the complexity of the problem analysis.

[0078] Before determining the grouping index, it is necessary to select the initial variable index that can fully characterize the operating status of the new energy unit in the entire operating condition. The unit output external performance comprehensively reflects the operating status of the new energy unit. Therefore, based on the output characteristics of the new energy unit, the active response characteristics and reactive response characteristics of the new energy unit that are easy to obtain in actual engineering are selected to analyze the operating status of each new energy unit under the three-phase short-circuit fault condition. At the same time, it is also necessary to select a suitable sampling time point to meet the requirements of electromechanical transient stability analysis under all operating conditions.

[0079] Specifically, the initial variable indicators of the new energy unit include the four output external characteristics of the new energy unit P, Q, i d 、i q The transient and steady-state operation data; among them, P, Q, i d 、i q They are respectively the active response characteristic of the new energy unit, the reactive response characteristic index, the direct axis component current index of the motor and the quadrature axis component current index of the motor;

[0080] Four output external characteristics P, Q, i d 、i q The selection basis of transient and steady-state operation data is as follows:

[0081] During the initial steady state, select P, i at a time point d ;

[0082] During the transition period of the protection circuit action, select P, Q, and i at two time points d 、i q ;

[0083] During the transient stability period of voltage drop, select P, Q, and i at two time points d 、i q ;

[0084] Select P, i at a time point during the steady state after the fault is cleared d ;

[0085] Finally, the initial variable indicators of the 20-dimensional new energy unit are obtained.

[0086] As an example, select P and i at a time point during the initial steady state. d , marked as 1 under the indicator, specifically P1, i d1 ;

[0087] During the transition period of the protection circuit action, select P, Q, and i at two time points d 、i q , the indicators are marked as 2 and 3; specifically P2, Q2, i d2 、i q2 、P3、Q3、i d3 、i q3 ;

[0088] During the transient stability period of voltage drop, select P, Q, and i at two time points d 、i q , the indicators are marked as 4 and 5; specifically P4, Q4, i d4 、i q4 、P5、Q5、i d5 、i q5 ;

[0089] Select P, i at a time point during the steady state after the fault is cleared d , marked as 6 under the indicator; specifically P6, i d6 ;

[0090] Therefore, the final 20-dimensional initial variable indicators of the new energy unit are obtained.

[0091] In the multi-machine equivalent modeling of new energy stations, there is a strong correlation and unidentifiability between the 20-dimensional initial variable indicators of each selected new energy unit. Therefore, it is necessary to perform effective data processing on the selected initial variable indicators of new energy units, eliminate the strong correlation between variable indicators, and improve the effectiveness of new energy unit grouping; fully explore the intrinsic essential information of indicator data, map the actual operating status of new energy units, enhance the identifiability and guiding role of variable indicators, and provide a reasonable basis for the grouping of new energy units.

[0092] Factor analysis is a data mining technique that studies the extraction of common factors from a group of variables and belongs to the category of multivariate statistics. Starting from studying the dependency relationship within the correlation matrix, it groups variables based on correlation and represents each group with an unobservable comprehensive variable, thereby reducing the dimensions of the initial variables with complex relationships into a few comprehensive factor variables while minimizing information loss. The factor analysis method can be used to reduce the dimensions of the initial variable data, eliminate the correlation between variables, and extract representative factors with practical meanings, thereby realizing the mapping of the relationship between data and physical meaning.

[0093] Therefore, before establishing the factor analysis model of the clustering index, it is necessary to perform data standardization on the selected 20-dimensional initial variable matrix. In the big data analysis project, the complex and redundant multi-source data are often not comparable under the same conditions due to the different dimensions and levels. It is necessary to uniformly standardize the original data to eliminate the analysis bias caused by the data itself. The basic principle is to subtract the mean value of the data from the data value, and then divide it by its standard deviation to obtain data with a mean of 0 and a standard deviation of 1 that obeys the standard normal distribution, that is, the 0-1 standardization method. The index data processed by the standardization method can be used for the analysis, calculation and comparison of subsequent indicators.

[0094] Furthermore, a factor analysis model is constructed, specifically:

[0095] The factor analysis model is: X = AF;

[0096] X is the standardized initial variable index, F=(F1, F2,…, F m ) T is a common factor vector, the number of common factors is m; A=(a ik )21×m is the factor loading matrix, a ik represents the i-th dimension variable X in X i In the kth common factor F k The factor loading on ik The larger the absolute value of X i With F k The greater the degree of correlation;

[0097] In the process of solving the factor analysis model, determining factor loading, factor rotation analysis and calculating factor scores are the three core steps of the factor analysis method. Specific analysis around these three elements is also the key to data processing of new energy unit group indicators. Specifically,

[0098] First, factor loading determination

[0099] The calculation methods of factor loading matrix include principal component analysis, maximum likelihood estimation, principal factor method, etc. The present invention adopts principal component analysis to determine the factor loading matrix. The basic idea is to transform several original variables into another group of several unrelated variables by means of coordinate transformation. The factor loading matrix A obtained by principal component analysis is:

[0100]

[0101] Where: λ1 to λ m is the first m largest characteristic roots of the correlation matrix R of the initial variable index, λ1 to λ m are the 1st to mth largest characteristic roots respectively; λ1≥λ2≥…λ m >0,

[0102] u1,u2,…,u m is the characteristic root corresponding to λ1,λ2,…,λ m The eigenvector of , the number of common factors m is determined by the cumulative variance contribution rate The number of common factors is limited and determined by the standard of cumulative variance contribution rate reaching 90%. The factor loading matrix is ​​determined by this method. While reducing the number of variables to achieve dimensionality reduction of the initial variable matrix, the information amount of the selected common factors can also effectively represent the information of the initial variables, minimizing the loss of information.

[0103] Second, factor rotation analysis

[0104] The purpose of establishing a factor analysis model is not only to extract common factors and group the original variables by correlation, but more importantly to explain the physical meaning of each extracted common factor in order to make scientific analysis and judgment on practical problems. At this time, the factor loading matrix can be subjected to a varimax orthogonal rotation based on the non-uniqueness of the factor loading matrix. The varimax orthogonal rotation is to multiply the factor loading matrix by an orthogonal matrix on the right to maximize the variance of the square of each column of the factor loading, that is, the factor loading is polarized, thereby obtaining the rotated factor loading matrix. After the varimax orthogonal rotation, each original variable has a larger factor loading only on one common factor, and its absolute value tends to 1, while the factor loading on the remaining common factors is smaller, and its absolute value tends to 0. Factor loading a ik Reflects the variable X i With the common factor F k The greater the absolute value, the greater the weight of the common factor, and the higher the correlation between the variable and the common factor. The present invention rotates the factors of the extracted initial variable indicators of each new energy unit, and can judge the correlation between each initial variable indicator and the common factor according to the rotated factor load matrix. Name the common factors according to the actual operating status of the new energy unit represented by the initial variable indicators, give each common factor an actual physical meaning, and form a mapping relationship between the factors and the operating status of the new energy unit.

[0105] 3. Factor score calculation

[0106] In the analysis of practical problems, it is often necessary to use the value of common factors to evaluate the status of samples in the model. In the factor analysis model, since the factor loading matrix is ​​not reversible, it is impossible to directly solve the value of the common factor. In this case, it is necessary to measure the common factors, express the common factors as linear combinations of the original variables, and calculate the factor scores.

[0107] Furthermore, the expression for obtaining factor score data is:

[0108]

[0109] Where: is the factor score data of all common factors in the common factor vector at each sample point; A′ is the rotated factor loading matrix; R is the correlation matrix of the standardized initial variable indicator X.

[0110] After determining the clustering indicators of each new energy unit, it is necessary to use reasonable methods to cluster each new energy unit based on the clustering indicators. In order to eliminate the reliance on clustering algorithms during the clustering process, it is necessary to use the data characteristics of the selected clustering indicators to measure the similarity of the clustering indicators, obtain the iconic dividing points between the new energy units, and perform clustering based on significant feature information. Specifically, the clustering indicators are measured for similarity using correlation analysis and significance tests, and the correlation between the new energy units is obtained to complete the group division in the following way:

[0111] First, the correlation coefficient of the factor score data of the new energy unit is obtained by correlation analysis; the specific correlation coefficient

[0112] In the formula, is the factor score data of the hth common factor in the common factor vector F at each sample point, is the factor score data of the jth common factor in the common factor vector F at each sample point;

[0113] Consider each factor score data of the new energy unit as a sample, c hj For sample and samples The correlation coefficient of For sample and samples The covariance of The samples are and samples The variance of .

[0114] Then, the significance test of the correlation coefficient is carried out to obtain the correlation between the new energy units; among which, the correlation is strong correlation or non-strong correlation;

[0115] Finally, by using the correlation between any two new energy units, the strongly correlated new energy units are divided into one group, and the two non-strongly correlated new energy units are divided into different groups, and finally the division of all new energy units in the photovoltaic power station is completed.

[0116] In this preferred implementation, the similarity measurement of the clustering indicators is completed through correlation analysis and significance test, and the correlation coefficient of the new energy unit factor score data can be obtained to reflect the correlation between the new energy units, so as to cluster the new energy units using clear correlation.

[0117] Furthermore, based on the clustering results, equivalent parameters are calculated to establish an equivalent model of new energy stations. The implementation methods for completing the modeling include:

[0118] After the original n new energy units in the new energy station are divided into s groups, s<n, the new energy units in each group are aggregated into a single new energy unit through single-machine equivalence, and the aggregated single new energy unit is used as an equivalent unit. At the same time, the capacity weighting method and parameter identification method are used to determine the equivalent parameters, so as to realize the equivalent model of the new energy station containing n new energy units into the equivalent model of the new energy station containing only s new energy units, and complete the modeling;

[0119] The principles followed during the modeling process are:

[0120] a. The total rated capacity of the equivalent model and detailed model of the new energy station is the same;

[0121] b. The grid connection point voltage of the equivalent model and detailed model of the new energy station is the same;

[0122] c. The active power and reactive power output by the equivalent model and detailed model of the new energy station are the same.

[0123] Furthermore, the implementation methods of determining equivalent parameters using the capacity weighting method and the parameter identification method include:

[0124] (1) Calculation method of equivalent parameters of generator and transformer

[0125] The equivalent parameters of the generator and transformer are:

[0126]

[0127] Where: S Geq , S Teq , Z Geq and Z Teq are the apparent power of the equivalent unit, the apparent power of the machine-end transformer, the generator impedance and the machine-end transformer impedance respectively; S Gh and Z Gh are the apparent power and generator impedance of wind turbine h respectively; S Th and Z Th are the apparent power and impedance of the transformer at the machine end of wind turbine h, respectively; f is the number of wind turbines included in the equivalent unit, 1≤h≤f;

[0128] (2) Calculation method of equivalent parameters of shaft system

[0129] The equivalent parameters of the axis system are:

[0130]

[0131] Where: H teq , H geq , K eq and D eqare the wind turbine inertia time constant, generator rotor inertia time constant, shaft stiffness coefficient and shaft damping coefficient of the equivalent unit respectively; H th , H gh , K h and D h are the wind turbine inertia time constant, generator rotor inertia time constant, shaft stiffness coefficient and shaft damping coefficient of wind turbine h respectively;

[0132] (3) Calculation method of wind speed and other parameters

[0133] Wind speed and other parameters are:

[0134]

[0135] Where: v eq , A and c peq are the input wind speed, wind turbine swept area and wind energy utilization coefficient of the equivalent unit respectively; v h , A h and c ph are the input wind speed, wind turbine swept area and wind energy utilization coefficient of wind turbine h respectively;

[0136] (4) Calculation method of equivalent parameters of photovoltaic array

[0137] According to the principle of equivalent modeling, the output voltage of the photovoltaic array remains unchanged before and after the equivalent value, so the number of series-connected cells in the equivalent photovoltaic array remains unchanged; and since the total output power is the same before and after the equivalent value, when all photovoltaic power generation units are equalized to one, the output current of the photovoltaic array needs to be increased to λ times the original value. Therefore,

[0138] The equivalent parameters of the photovoltaic array are:

[0139]

[0140] Where: N seq 、N peq are the number of series and parallel cells in the photovoltaic array in the equivalent model; N s 、N p are the number of batteries in series and parallel connection of a single photovoltaic array in the detailed model; λ is the coefficient;

[0141] The equivalent light intensity is obtained by taking the weighted average of the light intensity of each photovoltaic power generation unit.

[0142] (5) Calculation method of equivalent parameters of photovoltaic inverter

[0143] Since the rated capacity of the equivalent inverter should be the sum of the rated capacities of all photovoltaic inverters, the impedance should be the equivalent impedance of all photovoltaic inverters connected in parallel. Since the inverter models are the same in the same photovoltaic power station, the multiplication method can be used for parameter calculation. Therefore, the equivalent parameters of the photovoltaic inverter are:

[0144]

[0145] Where: S eq , Z ceq are the capacity and equivalent impedance of the inverter in the equivalent model; S, Z c are the capacity and impedance of a single PV inverter in the detailed model; (6) the calculation method of the capacitance and inductance parameters of the AC side filter

[0146] The control structure of the inverter remains unchanged before and after the equalization, and the control parameters remain consistent. Therefore, the capacitance and inductance parameters of the AC side filter are:

[0147]

[0148] Where: C eq , L eq are the filter capacitor and filter inductor on the AC side of the inverter in the equivalent model; C and L are the filter capacitor and filter inductor on the AC side of a single inverter in the detailed model;

[0149] (7) Calculation method of equivalent parameters of collector line

[0150] There are mainly two types of collection lines in new energy stations: radial topology and trunk topology. Figure 2 a is a schematic diagram of the radial topology structure. Each new energy unit is connected to the PCC point through the line. The collection system of this topology structure can directly calculate the equivalent impedance using the equal power loss method. Figure 2 b is a schematic diagram of the trunk topology structure. Except for the first section of the new energy unit connected to the PCC point through the line, the other units are connected to the end of the line of the previous unit through the line. The power collection system of this topology can be converted into a radial topology first, and then the equivalent impedance is calculated using the equal power loss method. First, use the equal power loss method to calculate the equivalent impedance of each new energy unit on the trunk line, and convert it into a radial topology connection. Figure 3 As shown in the figure, there are n new energy units connected in a trunk topology at the PCC point. After being converted into a radial topology, the equivalent impedance of each unit is calculated according to the following formula:

[0151] The equivalent parameters of the collector line are:

[0152]

[0153] Where: Z irepresents the line impedance at the outlet of the i-th unit; P i The impedance Z i Power; Z eqm′ Represents the equivalent line impedance of the m′th unit branch, i≤m′≤n.

[0154] Specific implementation method 2: A new energy station operation evaluation model equivalent modeling device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the new energy station operation evaluation model equivalent modeling method.

[0155] Specific implementation method three: a computer-readable storage device, wherein the storage device stores a computer program, and when the computer program is executed, the equivalent modeling method of the new energy station operation evaluation model is implemented.

[0156] Verification test:

[0157] After establishing the new energy station equivalent model, it is necessary to verify the validity of the model. The error evaluation index of the new energy station equivalent model is defined based on the output characteristics of the new energy station detailed model as follows:

[0158]

[0159] Where: W i , W i * They respectively represent the active output or reactive output of the detailed model of the new energy station and the equivalent model at the outlet of the new energy station; n represents the number of sampling points.

[0160] It has been verified that the error evaluation index of the present invention is relatively small, which proves the effectiveness of the present invention.

[0161] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. The equivalent modeling method of the new energy station operation evaluation model is characterized by: The method comprises the following steps: Extract the initial variable indicators of new energy units; Establish factor analysis model of initial variable indicators; Obtain factor score data as a clustering indicator; Use correlation analysis and significance test to measure the similarity of clustering indicators, obtain the correlation of new energy units and complete the cluster division; Based on the clustering results, the equivalent parameters are calculated, and the equivalent model of the new energy station is established to complete the modeling.

2. The equivalent modeling method of the new energy station operation evaluation model according to claim 1 is characterized in that: The initial variable indicators of the new energy unit include the four output external characteristics of the new energy unit P, Q, i d 、i q Transient and steady-state operation data; Among them, P, Q, i d 、i q They are respectively the active response characteristics of the new energy unit, the reactive response characteristics index, the direct axis component current index of the motor and the quadrature axis component current index of the motor; the four output external characteristics P, Q, i d 、i q The selection basis of transient and steady-state operation data is as follows: During the initial steady state, select P, i at a time point d ; During the transition period of the protection circuit action, select P, Q, and i at two time points d 、i q ; During the transient stability period of voltage drop, select P, Q, and i at two time points d 、i q ; Select P, i at a time point during the steady state after the fault is cleared d ; Finally, the initial variable indicators of the 20-dimensional new energy unit are obtained.

3. The equivalent modeling method of the new energy station operation evaluation model according to claim 1 is characterized in that: The factor analysis model is: X = AF; X is the standardized initial variable index, F=(F1, F2,…, F m ) T is a common factor vector, the number of common factors is m; A=(a ik ) 21×m is the factor loading matrix, a ik represents the i-th dimension variable X in X i In the kth common factor F k The factor loading on ik The larger the absolute value of X i With F k The greater the degree of correlation; Where: λ1 to λ m is the first m largest characteristic roots of the correlation matrix R of the initial variable index, λ1 to λ m are the 1st to mth largest characteristic roots respectively; λ1≥λ2≥…λ m >0,u1,u2,…,u m is the characteristic root corresponding to λ1,λ2,…,λ m The feature vector of .

4. The equivalent modeling method of the new energy station operation evaluation model according to claim 3 is characterized in that: The number of common factors m is determined by the cumulative variance contribution rate Sure.

5. The equivalent modeling method of the new energy station operation evaluation model according to claim 1 is characterized in that: The expression for obtaining factor score data is: Where: is the factor score data of all common factors in the common factor vector at each sample point; A′ is the rotated factor loading matrix; R is the correlation matrix of the standardized initial variable indicator X.

6. The equivalent modeling method of the new energy station operation evaluation model according to claim 1 is characterized in that: The similarity of clustering indicators is measured by using correlation analysis and significance test, and the correlation of new energy units is obtained to complete the clustering division in the following way: Firstly, the correlation coefficient of the factor score data of the new energy units is obtained by using correlation analysis; Then, the significance test of the correlation coefficient is carried out to obtain the correlation between the new energy units; among which, the correlation is strong correlation or not strong correlation; Finally, by using the correlation between any two new energy units, the strongly correlated new energy units are divided into one group, and the two non-strongly correlated new energy units are divided into different groups, and finally the division of all new energy units in the photovoltaic power station is completed.

7. The equivalent modeling method of the new energy station operation evaluation model according to claim 6 is characterized in that: The expression of the correlation coefficient is: In the formula, is the factor score data of the hth common factor in the common factor vector F at each sample point, is the factor score data of the jth common factor in the common factor vector F at each sample point; Consider each factor score data of the new energy unit as a sample, c hj For sample and samples The correlation coefficient of For sample and samples The covariance of The samples are and samples The variance of .

8. The equivalent modeling method of the new energy station operation evaluation model according to claim 1 is characterized in that: Based on the clustering results, the equivalent parameters are calculated and the equivalent model of the new energy station is established. The implementation methods of completing the modeling include: After the original n new energy units in the new energy station are divided into s groups, s<n, the new energy units in each group are aggregated into a single new energy unit through single-machine equivalence, and the aggregated single new energy unit is used as an equivalent unit. At the same time, the capacity weighting method and parameter identification method are used to determine the equivalent parameters, so as to realize the equivalent model of the new energy station containing n new energy units into the equivalent model of the new energy station containing only s new energy units, and complete the modeling; And the principles followed in the modeling process are: a. The total rated capacity of the equivalent model and detailed model of the new energy station is the same; b. The grid connection point voltage of the equivalent model and detailed model of the new energy station is the same; c. The active power and reactive power output by the equivalent model and detailed model of the new energy station are the same.

9. The equivalent modeling method of the new energy station operation evaluation model according to claim 8 is characterized in that: The implementation methods of determining equivalent parameters using the capacity weighting method and parameter identification method include: The equivalent parameters of the generator and transformer are: Where: S Geq , S Teq , Z Geq and Z Teq are the apparent power of the equivalent unit, the apparent power of the machine-end transformer, the generator impedance and the machine-end transformer impedance respectively; S Gh and Z Gh are the apparent power and generator impedance of wind turbine h respectively; S Th and Z Th are the apparent power and impedance of the transformer at the machine end of wind turbine h, respectively; f is the number of wind turbines included in the equivalent unit, 1≤h≤f; The equivalent parameters of the axis system are: Where: H teq , H geq , K eq and D eq are the wind turbine inertia time constant, generator rotor inertia time constant, shaft stiffness coefficient and shaft damping coefficient of the equivalent unit respectively; H th , H gh , K h and D h are the wind turbine inertia time constant, generator rotor inertia time constant, shaft stiffness coefficient and shaft damping coefficient of wind turbine h respectively; Wind speed and other parameters are: Where: v eq , A and c peq are the input wind speed, wind turbine swept area and wind energy utilization coefficient of the equivalent unit respectively; v h , A h and c ph are the input wind speed, wind turbine swept area and wind energy utilization coefficient of wind turbine h respectively; The equivalent parameters of the photovoltaic array are: Where: N seq 、N peq are the number of series and parallel cells in the photovoltaic array in the equivalent model; N s 、N p are the number of batteries in series and parallel connection of a single photovoltaic array in the detailed model; λ is the coefficient; The equivalent light intensity is obtained by taking the weighted average of the light intensity of each photovoltaic power generation unit; The equivalent parameters of the photovoltaic inverter are: Where: S eq , Z ceq are the capacity and equivalent impedance of the inverter in the equivalent model; S, Z c are the capacity and impedance of a single PV inverter in the detailed model, respectively; The capacitance and inductance parameters of the AC side filter are: Where: C eq , L eq are the filter capacitor and filter inductor on the AC side of the inverter in the equivalent model; C and L are the filter capacitor and filter inductor on the AC side of a single inverter in the detailed model; The equivalent parameters of the collector line are: Where: Z i represents the line impedance at the outlet of the i-th unit; P i The impedance Z i Power; Z eqm′ Represents the equivalent line impedance of the m′th unit branch, i≤m′≤n.

10. A new energy station operation evaluation model equivalent modeling device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes a computer program to implement an equivalent modeling method for a new energy station operation evaluation model as described in any one of claims 1 to 9.

Images