CN117057227A - Equivalent modeling method and related device for photovoltaic power station - Google Patents

Abstract

The present invention belongs to the technical field of equivalent modeling methods. In order to solve the technical problem of insufficient parameter setting accuracy in existing photovoltaic power station equivalent methods, a photovoltaic power station equivalent modeling method and related devices are provided. The characteristic vectors of the inverters in the photovoltaic power station are clustered and analyzed to obtain the cluster centers of each cluster. Then the equivalent machine parameters of each cluster are calculated, and the box transformers and collector lines are equivalently processed. Finally , according to the cluster center of each cluster, the equivalent machine parameters, and the equivalent processing results of the box transformer and collector line, the equivalent model of the photovoltaic power station is obtained. Compared with the existing modeling method, there is no need to input the number of equivalent units, and there will be no problem of insufficient accuracy in setting the number of equivalent units when setting parameters. This greatly reduces the probability of deviations during simulation analysis, and effectively improves the construction efficiency. accuracy of model results.

Description

A photovoltaic power station equivalent modeling method and related device

Technical Field

The invention belongs to the technical field of equivalent modeling methods, and relates to an equivalent modeling method for a photovoltaic power station and related devices.

Background Art

In recent years, under the guidance of the dual carbon goals, the proportion of renewable energy represented by photovoltaics in the power system has gradually increased. Among them, photovoltaic power generation, as an important renewable energy, has an important impact on the stable operation of the power system, and has brought a series of challenges in terms of safety and stability to the new power system with new energy as the main body. At present, the capacity of photovoltaic inverters is mostly 1MW-3MW, and there are usually hundreds of inverters in a single station. If the detailed electromechanical model of each inverter is considered when simulating the operation mode of the power grid and fault simulation, the calculation amount is too large, which wastes computing resources and leads to the lack of timeliness of the simulation results, affecting the safety and stability analysis of the power system. Therefore, it is necessary to establish an equivalent model of the photovoltaic power station on the basis of ensuring the accuracy of the simulation to reduce the complexity of the simulation calculation of the whole network.

Most of the existing equivalence methods for photovoltaic power stations require priority determination of grouping, and do not consider parameter identification of low voltage ride-through control parameters of photovoltaic inverters, which leads to insufficient accuracy of parameter setting.

Summary of the invention

In order to solve the technical problem that the existing photovoltaic power station equivalent method has insufficient parameter setting accuracy, the present invention provides a photovoltaic power station equivalent modeling method and related devices.

In order to achieve the above object, the present invention adopts the following technical solutions:

In a first aspect, the present invention provides a photovoltaic power station equivalent modeling method, comprising the following steps:

Perform cluster analysis on the inverters in the photovoltaic power station and obtain the cluster center of each cluster;

Calculate the equivalent machine parameters of each cluster, and perform equivalent processing on the box-type transformer and collector line;

According to the cluster center of each cluster, the equivalent machine parameters, and the equivalent processing results of the box transformer and the collector line, an equivalent model of the photovoltaic power station is obtained.

Preferably, the cluster analysis is specifically:

Step 1: divide the feature vectors used to characterize the inverters in the photovoltaic power station into k clusters, and select k from all feature vectors as initial cluster centers;

Step 2: Take each feature vector as a sample, calculate the distance between each sample and the k initial cluster centers according to the value of the clustering index corresponding to each sample, and then divide each sample into the cluster corresponding to the sub-cluster center with the minimum distance;

Step 3, according to the marginal sample percentage, calculate the number of marginal samples of each cluster, and obtain the marginal samples of each cluster;

Step 4: Consider the edge samples of each cluster and iteratively optimize the initial cluster centers of all clusters until the difference between the sum of the distances of all samples to the initial cluster centers of their clusters between two adjacent iterations meets the preset requirements, thus obtaining the cluster centers of all clusters.

Preferably, in step 2, the clustering index includes an average steady-state voltage value Us during normal operation of the photovoltaic inverter, an average active power value Pe output by the inverter during a fault, and an average reactive power value Qe output by the inverter during a fault.

Preferably, in step 2, the distance between each sample and the centers of the k initial clusters is calculated by the following formula:

Wherein, d _ij ( _xi , c _j ) is the Euclidean distance from sample _xi to the initial cluster center c _j , j = 1, 2, ..., k, u is a variable, _xiu is the value of the u-th clustering index of sample _xi , and c _ju is the value of the u-th clustering index of the initial cluster center c _j ;

In step 4, the iterative optimization is specifically as follows:

1) Recalculate the new initial cluster center c′ _j by the following formula:

Among them, g is the current number of samples in the jth cluster, and _Xm is the sample set consisting of all samples of the photovoltaic power station;

M _aη represents the number of marginal samples in clusters other than cluster a; m _bη is the number of marginal samples of m _b samples in the bth cluster;

Y _aη represents the marginal samples of other clusters except cluster a;

2) Return to step 2 and perform iterative optimization. After each iteration, calculate the sum P of the distances from all samples to the initial cluster center of their cluster clusters until the difference between two adjacent iterative calculations meets the preset requirements. The initial cluster center of each cluster cluster is used as the new cluster center after optimization. P is calculated by the following formula:

Preferably, the equivalent machine parameters include equivalent machine power, equivalent machine low voltage ride-through period control parameters and equivalent machine low voltage ride-through active current recovery control parameters;

The equivalent machine parameters of each cluster are calculated, and the equivalent processing is performed on the box-type transformer and the collector line, specifically:

1) Define all inverters in each cluster as a unit group, add the steady-state active power and steady-state reactive power of all inverters in each unit group respectively, and obtain the active power and reactive power of the equivalent machine, that is, the equivalent machine power;

2) The gradient descent algorithm is used to perform multivariate linear regression on the low voltage ride through active current and reactive current of all inverters in each unit group, and the low voltage ride through active current setting value Ip _{set_LV} , low voltage ride through active current control parameters K _1-Ip-LV and K _2-Ip-LV , low voltage ride through reactive current setting value Iq _{set_LV} and low voltage ride through reactive current control parameters K _1-Iq-LV and K _2-Iq-LV are obtained. The inverter low voltage ride through active current Ip _LVRT and inverter low voltage ride through reactive current Iq _LVRT are obtained by the following formula as the control parameters of the equivalent machine during low voltage ride through:

Ip _LVRT ＝K _1-Ip-LV *Vt+K _2-Ip-LV *Ip ₀ +Ip _{set_LV}

Iq _LVRT ＝K _1-Iq-LV *(0.9-Vt)+K _2-Iq-LV *Iq ₀ +Iq _{set_LV}

Wherein, Vt is the terminal voltage during the fault period, Ip ₀ is the active current before entering the low voltage ride through, and Iq ₀ is the reactive current before entering the low voltage ride through;

3) The equivalent machine active current recovery rate parameter is obtained by the following formula: As the control parameters of the active current recovery of the low voltage ride-through of the equivalent machine:

Where m is the total number of inverters in the group. is the active current recovery rate parameter of the i-th inverter

Where, _In is the rated current of the inverter, is the slope of the line segment between the two data points at the steady-state start time t1 and the steady-state end time t2 on the active current curve after the i-th inverter recovers from fault:

in, is the inverter output active current at time t1 after the fault of the i-th inverter is restored, is the inverter output active current at time t2 after the fault of the i-th inverter is restored;

4) The total power consumption ΔP _Teq , equivalent impedance Z _Teq and equivalent capacity S _Teq of the transformer in the box-type transformer after equivalent treatment are obtained by the following formula as the box-type transformer equivalent treatment result:

ΔP _Teq ＝(NI _T ) ² Z _Teq

Z _Teq = Z _T /N

S _Teq = NS _T

Among them, N is the total number of box-type transformers, I _T is the current of each box-type transformer, Z _T is the impedance of the box-type transformer, and S _T is the capacity of a single box-type transformer;

5) The power loss ΔP _eq and total impedance Z _eq of the collector line after equalization are obtained by the following formula as the result of equalization processing of the collector line:

ΔP _eq ＝(FI) ² Z _eq

Where I is the output current of the photovoltaic array, _Ze is the impedance of the e-th collector line, and F is the total number of collector lines.

In a second aspect, the present invention provides a photovoltaic power station equivalent modeling system, comprising:

A cluster analysis module is used to perform cluster analysis on the inverters in the photovoltaic power station and obtain the cluster center of each cluster;

The equivalent module is used to calculate the equivalent machine parameters of each cluster and perform equivalent processing on the box-type transformer and the collector line;

The modeling module is used to obtain an equivalent model of the photovoltaic power station according to the cluster center of each cluster, the equivalent machine parameters, and the equivalent processing results of the box transformer and the collector line.

Preferably, the cluster analysis module includes:

The cluster center submodule is used to divide the feature vectors used to characterize the inverters in the photovoltaic power station into k clusters, and select k from all the feature vectors as initial cluster centers;

The classification submodule is used to take each feature vector as a sample, calculate the distance between each sample and the k initial cluster centers according to the value of the clustering index corresponding to each sample, and then divide each sample into the cluster corresponding to the cluster center with the minimum distance;

The edge sample submodule is used to calculate the number of edge samples of each cluster according to the edge sample percentage, and obtain the edge samples of each cluster;

The iterative optimization submodule is used to consider the edge samples of each cluster and iteratively optimize the initial cluster centers of all clusters until the difference between the sum of the distances of all samples to the initial cluster centers of their clusters between two adjacent iterations meets the preset requirements, thereby obtaining the cluster centers of all clusters.

Preferably, the equivalent machine parameters include equivalent machine power, equivalent machine low voltage ride-through period control parameters and equivalent machine low voltage ride-through active current recovery control parameters;

The equivalent modules include:

The equivalent machine power submodule is used to add the steady-state active power and steady-state reactive power of all inverters in each unit group to obtain the active power and reactive power of the equivalent machine, that is, the equivalent machine power; all inverters in each cluster are defined as a unit group;

The low voltage ride through control submodule is used to perform multivariate linear regression on the low voltage ride through active current and reactive current of all inverters in each unit group by using the gradient descent algorithm, obtain the low voltage ride through active current setting value Ip _{set_LV} , low voltage ride through active current control parameters K _1-Ip-LV and K _2-Ip-LV , low voltage ride through reactive current setting value Iq _{set_LV} and low voltage ride through reactive current control parameters K _1-Iq-LV and K _2-Iq-LV , and obtain the inverter low voltage ride through active current Ip _LVRT and inverter low voltage ride through reactive current Iq _LVRT by the following formula as the control parameters of the equivalent machine during low voltage ride through:

Ip _LVRT ＝K _1-Ip-LV *Vt+K _2-Ip-LV *Ip ₀ +Ip _{set_LV}

Iq _LVRT ＝K _1-Iq-LV *(0.9-Vt)+K _2-Iq-LV *Iq ₀ +Iq _{set_LV}

Wherein, Vt is the terminal voltage during the fault period, Ip ₀ is the active current before entering the low voltage ride through, and Iq ₀ is the reactive current before entering the low voltage ride through;

Then the equivalent machine active current recovery rate parameter is obtained by the following formula: As the control parameters of the active current recovery of the low voltage ride-through of the equivalent machine:

Where m is the total number of inverters in the group. is the active current recovery rate parameter of the i-th inverter

Where, _In is the rated current of the inverter, is the slope of the line segment between the two data points at the steady-state start time t1 and the steady-state end time t2 on the active current curve after the fault of the i-th inverter is restored:

in, is the inverter output active current at time t1 after the fault of the i-th inverter is restored, is the inverter output active current at time t2 after the fault of the i-th inverter is restored;

The box-type transformer submodule is used to obtain the total power consumption ΔP _Teq , equivalent impedance Z _Teq and equivalent capacity S _Teq of the transformer in the box-type transformer after equivalent value processing as the box-type transformer equivalent value processing result through the following formula:

ΔP _Teq ＝(NI _T ) ² Z _Teq

Z _Teq = Z _T /N

S _Teq = NS _T

Among them, N is the total number of box-type transformers, I _T is the current of each box-type transformer, Z _T is the impedance of the box-type transformer, and S _T is the capacity of a single box-type transformer;

The collector line submodule is used to obtain the collector line power loss ΔP _eq and total impedance Z _eq after equivalent value processing through the following formula as the collector line equivalent value processing result:

ΔP _eq ＝(FI) ² Z _eq

Where I is the output current of the photovoltaic array, _Ze is the impedance of the e-th collector line, and F is the total number of collector lines.

In a third aspect, the present invention proposes a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the steps of the above method when executed by a processor.

Compared with the prior art, the present invention has the following beneficial effects:

The present invention proposes a photovoltaic power station equivalent modeling method, which is based on clustering analysis of characteristic vectors used to characterize inverters, obtaining the cluster center of each cluster cluster, and then calculating the equivalent machine parameters of each cluster cluster, and performing equivalent processing on the box transformer and the collector line. Compared with the existing modeling method, it does not need to input the number of equivalent units, and there will be no problem of insufficient accuracy in the number of equivalent units and parameter settings, which greatly reduces the probability of deviations in simulation analysis and effectively improves the accuracy of modeling results. It has been verified in practice that the equivalent model obtained by the modeling method of the present invention has a very high degree of consistency when compared with the detailed model, which also fully demonstrates the effectiveness and accuracy of the modeling method of the present invention.

Furthermore, when performing cluster analysis in the present invention, edge samples are taken into account in the K-means clustering algorithm, and the cluster center of each cluster is confirmed through iterative optimization. It has been verified that the cluster center obtained after optimization is highly consistent with the actual situation, and the existing cluster analysis method is fully optimized.

Furthermore, when performing cluster analysis in the present invention, the selected clustering indicators include the average steady-state voltage during normal operation of the photovoltaic inverter, the average active power output of the inverter during faults, and the average reactive power output of the inverter during faults, which are closer to the actual operating conditions of the photovoltaic power station, and fully consider the indicators that may be affected by the actual operating conditions of the photovoltaic power station, so that the modeling method of the present invention is more consistent with the actual operating conditions.

Furthermore, in the present invention, when calculating the equivalent machine parameters, it is proposed to use multivariate linear regression based on the gradient descent algorithm to identify the control parameters during the low voltage ride-through period for the clustered data, so that the equivalent model can reflect the characteristics of the detailed model in detail.

Furthermore, the present invention also proposes a photovoltaic power station equivalent modeling system, which implements the above-mentioned equivalent modeling method through a cluster analysis module, an equivalent module and a modeling module. Through a modular structure, an extremely accurate equivalent model of a photovoltaic power station can be obtained.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

FIG1 is a schematic diagram of a process flow of Embodiment 1 of the present invention;

FIG2 is a schematic diagram of a flow chart of a second embodiment of the present invention;

FIG3 is a schematic diagram of a flow chart of Embodiment 3 of the present invention;

FIG4 is a schematic diagram of hierarchical clustering numbers drawn in Embodiment 3 of the present invention;

5 is a schematic diagram showing the distribution of three clustering indicators of a photovoltaic power station after cluster center optimization in Embodiment 3 of the present invention;

FIG6 is a comparison diagram of reactive power between the equivalent model and the detailed model obtained in Example 3 of the present invention;

FIG7 is a comparison diagram of active power between the equivalent model and the detailed model obtained in Example 3 of the present invention;

FIG8 is a connection diagram of Embodiment 4 of the present invention;

FIG9 is a schematic diagram of the connection of a cluster analysis module in an embodiment of equivalent modeling of a photovoltaic power station according to the present invention;

FIG. 10 is a schematic diagram showing the connection of equivalent modules in an embodiment of equivalent modeling of a photovoltaic power station according to the present invention.

Among them, 401-cluster analysis module, 501-equivalent module, 601-modeling module, 701-cluster center submodule, 801-classification submodule, 901-edge sample submodule, 1001-iterative optimization submodule, 1101-equivalent machine power submodule, 1201-low voltage ride-through control submodule, 1301-box transformer submodule, 1401-collection line submodule.

DETAILED DESCRIPTION

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.

It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not require further definition and explanation in the subsequent drawings.

In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. indicate an orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", which does not mean that the structure must be completely horizontal, but can be slightly tilted.

In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

The present invention is further described in detail below in conjunction with the accompanying drawings:

In view of the problems existing in the prior art, the present invention proposes a multi-clustering indicator photovoltaic power station equivalence method considering low voltage ride-through. The solution of the present invention is described in detail through multiple embodiments of the present invention as follows.

Embodiment 1

As shown in FIG1 , this is a basic embodiment of a multi-clustering index photovoltaic power station equivalent method of the present invention, and the specific steps are as follows:

S101, cluster analysis is performed on the characteristic vectors used to characterize the inverters in the photovoltaic power station to determine the cluster center of each cluster and select clustering indicators, which generally include steady-state voltage indicators, active power indicators during faults, and reactive power indicators during faults.

S102, calculating the equivalent machine parameters of each cluster, and performing equivalent processing on the box transformer and the collector line, and obtaining the equivalent model of the photovoltaic power station according to the cluster center, equivalent machine parameters, box transformer and the collector line equivalent processing results of each cluster.

S103, obtaining an equivalent model of the photovoltaic power station according to the cluster center of each cluster, the equivalent machine parameters, and the equivalent processing results of the box transformer and the collector line.

Embodiment 2

As shown in FIG. 2 , a more detailed embodiment of a photovoltaic power station equivalent modeling method of the present invention is shown, and the specific steps are as follows:

S201, select multiple clustering indicators. Generally, the selected clustering indicators include a steady-state voltage indicator, an active power indicator during a fault, and a reactive power indicator during a fault. However, in fact, the present invention can select other indicators as clustering indicators. The three indicators in Example 1 can make the equivalent method of the present invention closer to the actual working condition.

Obtaining the steady-state voltage index can characterize the terminal voltage distribution during normal operation of the unit. The active power index during the fault period refers to the active power output by each inverter during the fault period, which can characterize the active power output characteristics of the unit during the low voltage ride-through period. The reactive power index during the fault period refers to the reactive power output by each inverter during the fault period, which can characterize the reactive power output characteristics of the unit during the low voltage ride-through period.

Cluster analysis is performed on the inverters in the photovoltaic power station to obtain the cluster centers of each cluster.

S202, setting the number of clusters of the photovoltaic power station to k, and correspondingly obtaining k clusters.

S203, all inverters in the photovoltaic power station are characterized by feature vectors, each feature vector is used as a sample, k samples are selected from all samples as initial cluster centers, and k initial cluster centers c=c ₁ ,c ₂ ,…c _k are obtained.

S204, calculating the distance between each sample and the k initial cluster centers according to the value of each clustering index corresponding to each sample, and dividing each sample into the cluster corresponding to the initial cluster center with the minimum distance.

S205, respectively calculating the number of edge samples of each cluster, and then obtaining edge samples.

S206, combining the edge samples of each cluster, iteratively optimize the initial cluster centers of all clusters. After each iterative calculation, calculate the sum P of the distances from all samples to the initial cluster centers of their clusters, until the P difference between two adjacent iterative calculations meets the preset requirements, and obtain the new cluster center of each cluster after optimization.

S207, calculate the equivalent machine parameters of each cluster of the photovoltaic power station: (1) equivalent machine power; (2) equivalent machine low voltage ride-through control parameters; (3) equivalent machine low voltage ride-through active current recovery control parameters. And perform equivalent processing on the box transformer and the collector line respectively.

S208, obtaining an equivalent model of the photovoltaic power station based on the optimized cluster center, the equivalent machine parameters of each cluster, and the equivalent processing results of the box transformer and the collector line.

Embodiment 3

As shown in FIG3 , it is a more specific embodiment of a photovoltaic power station equivalent modeling method of the present invention. Taking a photovoltaic power station with 30 inverters as an example, the specific steps are as follows:

S301, selecting the average steady-state voltage Us of the photovoltaic inverter during normal operation, the average active power Pe output by each inverter during a fault, and the average reactive power Qe output by each inverter during a fault as clustering indicators.

S302, draw a hierarchical clustering tree as shown in Figure 4, in which the ordinate represents the Euclidean distance and the abscissa represents the serial number of the inverter in the photovoltaic power station. Taking 0.01 as the clustering boundary, the number of clusters of the photovoltaic power station is determined to be k=2 according to the clustering boundary.

In step S302, the clustering boundary line may also be adjusted according to the actual working conditions. In addition, the number of clusters may also be determined by other methods besides the hierarchical clustering tree. Even if the hierarchical clustering tree method is used, the specific clustering method may be adjusted.

S303: taking all feature vectors characterizing inverters in the photovoltaic power station as samples, randomly selecting k samples from all samples as initial cluster centers, and obtaining k initial cluster centers c=c ₁ ,c ₂ ,…c _k .

S304, assuming that the number of all samples is n, and n samples are recorded as x _i , i = 1, 2, ... n. The Euclidean distance, i.e., the straight-line distance, from sample x _i to the centers of the k initial clusters is calculated by the following formula. Since three clustering indicators are selected in step S301, the dimension w of sample x _i is 3, and the calculation formula of the Euclidean distance d _ij (x _i , c _j ) is as follows:

Wherein, d _ij ( _xi , c _j ) is the Euclidean distance from sample _xi to the initial cluster center c _j , j = 1, 2,…, k, u is a variable, _xiu is the value of the u-th clustering index of sample _xi , and c _ju is the value of the u-th clustering index of the initial cluster center c _j .

S305, calculating the number of edge samples m _aη of m _a samples in the a th cluster by the following formula:

m _a ＝ηm _a

In the third embodiment, because the number of clusters is k=2, a is equal to 1 or 2, and η is the percentage of edge samples, which can be set.

Then in each cluster, the m _aη samples with the largest distance from the center of the initial cluster are regarded as edge samples, denoted as X _jη .

S306, recalculate the new initial cluster center c _j ′ by the following formula:

Among them, g is the current number of samples in the jth cluster, and _Xm is the sample set consisting of all samples of the photovoltaic power station.

M _aη represents the number of marginal samples in clusters other than cluster a; m _bη is the number of marginal samples of m _b samples in the bth cluster.

Y _aη represents the marginal samples of other clusters except cluster a.

Repeat steps S304 to S306 for iterative optimization, and after each repetition, calculate the sum of the distances P from all samples to the initial cluster center of the cluster to which they belong, until the difference in P between two adjacent iterative calculations meets the preset requirements, and use the initial cluster center of each cluster as the new cluster center after optimization. P is calculated by the following formula:

When performing iterative optimization, if you find that the P difference between two iterations is too large, you can also try to adjust the edge sample percentage η, reduce the edge sample percentage η, and then perform iterative optimization.

As shown in FIG5 , it is a schematic diagram of the distribution of the three indicators of the photovoltaic power station units. It can be seen that the present invention adopts the above-mentioned K-means algorithm considering edge samples to perform cluster analysis on the feature vectors used to characterize the inverters in the photovoltaic power station, and the effect is as expected.

S307, calculate the equivalent machine parameters of each cluster of the photovoltaic power station: (1) equivalent machine power; (2) equivalent machine low voltage ride-through control parameters; (3) equivalent machine low voltage ride-through active current recovery control parameters. The inverters in each cluster are regarded as a unit group, and k unit groups are obtained after completing the cluster analysis. Parameters (2) and (3) are obtained by identifying the control parameters of the photovoltaic inverters in the unit group. The specific calculation method of each parameter is as follows:

1) Calculation of equivalent machine power

The steady-state active power and steady-state reactive power of all inverters in each unit group are added together to obtain the active power of the equivalent machine and the reactive power of the equivalent machine, that is, the equivalent machine power of each unit group is obtained.

2) Calculation of control parameters during low voltage ride-through of equivalent machines

2.1) Identification of control parameters during inverter low voltage ride-through

The gradient descent algorithm is used to perform multivariate linear regression on the low voltage ride through active current and reactive current of all inverters in each unit group.

(a) Obtain the data to be identified of all inverters in each group, including the active current Ip _LVRT during low voltage ride-through, the active current Ip ₀ before entering low voltage ride-through (as the initial active current), and the terminal voltage Vt during the fault period, and form a matrix X:

Where m represents the number of inverters in the group;

Let Y＝[Ip _LVRT ⁽¹⁾ ,Ip _LVRT ⁽²⁾ ,…,Ip _LVRT ^(m) ] ^T ,

Among them, _K1-Ip-LV and _K2-Ip-LV are two low voltage ride through active current control parameters, and Ip _{set_LV} is the low voltage ride through active current setting value.

(b) Calculate the loss function And the loss function gradient

(c) Setting initial parameters The iteration step size ψ and iteration accuracy ε are iteratively calculated according to the following formula:

Until After completing the iterative calculation, the corresponding low voltage ride-through control parameter identification results are obtained: Ip _{set_LV} , K _1-Ip-LV and K _2-Ip-LV .

2.2) The same method as step 2.1) is used to obtain another part of the control parameter identification results during the low voltage ride through period: two low voltage ride through reactive current control parameters K _1-Iq-LV and K _2-Iq-LV , and a low voltage ride through reactive current setting value Iq _{set_LV} .

2.3) Calculate the inverter low voltage ride-through active current Ip _LVRT by the following formula:

Ip _LVRT ＝K _1-Ip-LV *Vt+K _2-Ip-LV *Ip ₀ +Ip _{set_LV}

The inverter low voltage ride-through reactive current Iq _LVRT is calculated by the following formula:

Iq _LVRT ＝K _1-Iq-LV *(0.9-Vt)+K _2-Iq-LV *Iq ₀ +Iq _{set_LV}

Among them, Iq ₀ is the reactive current before entering the low voltage ride through.

The active current Ip _LVRT and the reactive current Iq _LVRT are the control parameters during the low voltage ride-through period of the equivalent machine.

3) Calculation of active current recovery control parameters for low voltage ride-through of equivalent machine

3.1) In the active current recovery stage, the slope of the active current curve after the fault recovery of each inverter is obtained by the following formula:

in, is the slope of the line segment between the two data points at the steady-state start time t1 and the steady-state end time t2 on the active current curve after the i-th inverter recovers from fault: is the inverter output active current at time t2 after the fault of the i-th inverter is restored, is the active current output by the inverter at time t1 after the fault of the i-th inverter is restored.

3.2) Calculate the active current recovery rate parameter of the i-th inverter by the following formula:

Where _In is the rated current of the inverter.

3.3) For all inverters in each group, the output active current of the inverter at time t1 after the fault of the i-th inverter is restored is calculated as follows: Active current recovery rate parameter for the i-th inverter The weighted average is calculated by the following formula to obtain the active current recovery rate parameter of the equivalent machine: As the control parameters of the active current recovery of the low voltage ride-through of the equivalent machine:

S308, perform equivalent processing on the box-type transformer and the collector line by the following method

1) Equivalent processing of box-type transformers

For the unit transformer, the total power loss _ΔPT before equalization is:

Where N is the total number of box-type transformers, _IT is the current of each box-type transformer, and _ZT is the impedance of the box-type transformer.

After equalization, the total transformer loss ΔP _Teq is:

ΔP _Teq ＝(NI _T ) ² Z _Teq

The equivalent impedance Z _Teq is obtained as:

Z _Teq = Z _T /N

The equivalent capacity S _Teq is:

S _Teq = NS _T

Among them, _ST is the capacity of a single box transformer.

After the equivalent processing of the box-type transformer, the equivalent impedance and equivalent capacity of the box-type transformer are obtained.

2) Equivalent treatment of collector lines

According to the principle that the total loss before and after the equalization is unchanged, for the radial topology, that is, each power generation unit is connected in parallel at the grid connection point, considering all photovoltaic power generation units, the line power loss ΔP before the equalization is:

Where F is the total number of collector lines, _Ie is the current of the e-th collector line, _Ze is the impedance of the e-th collector line, and I is the output current of the photovoltaic array. Under the same lighting conditions and temperature, the output current of all photovoltaic arrays is I, and the power loss of the collector line after equalization is _ΔPeq :

ΔP _eq ＝(FI) ² Z _eq

Wherein, _Zeq is the total impedance of the collector line after equalization.

To ensure that the transmission line losses are the same, the total impedance of the collector line Z _eq after equalization is:

Among them, the power loss ΔP _eq of the collector line after equalization and the total impedance Z _eq of the collector line after equalization are the results after the collector line is equalized.

S309, obtaining an equivalent model of the photovoltaic power station based on the optimized cluster center, the equivalent machine parameters of each cluster, and the equivalent processing results of the box transformer and the collector line.

For the photovoltaic power station including 30 inverters in Example 3, the number of clusters is 2. Considering that the photovoltaic power station enters low voltage ride-through due to a grid failure, the equivalent result obtained by the equivalent method in Example 3 is used as an equivalent model for comparison with the detailed model. Figure 6 is a reactive power comparison diagram, and Figure 7 is an active power comparison diagram. It can be seen from the comparison results of Figures 6 and 7 that the equivalent model obtained by the present invention is highly consistent with the detailed model.

The photovoltaic power station equivalent modeling method of the present invention proposes a K-means algorithm based on edge samples to cluster the multi-cluster indicators of the photovoltaic power station, and uses multivariate linear regression based on the gradient descent algorithm to identify the control parameters of the clustered data during the low voltage ride-through period, and determines the equivalent machine control parameters, thereby establishing the electromechanical transient equivalent model of the photovoltaic power station.

Embodiment 4

As shown in FIG. 8 , it is a basic embodiment of a photovoltaic power station equivalent modeling system of the present invention, which includes a cluster analysis module 401 , an equivalent module 501 and a modeling module 601 connected in sequence.

The cluster analysis module 401 is used to perform cluster analysis on the feature vectors used to characterize the inverters in the photovoltaic power station to obtain the cluster center of each cluster;

The equivalent module 501 is used to calculate the equivalent machine parameters of each cluster and perform equivalent processing on the box transformer and the collector line;

The modeling module 601 is used to obtain an equivalent model of the photovoltaic power station according to the cluster center of each cluster, the equivalent machine parameters, and the equivalent processing results of the box transformer and the collector line.

The cluster analysis module 401, the equivalent module 501 and the modeling module 601 are connected by data to realize the transmission between the processing of each module.

In other embodiments of a photovoltaic power station equivalent modeling system of the present invention, the cluster analysis module 401 and the equivalent module 601 may be further divided into multiple submodules for respectively specifically implementing the preferred steps of the equivalent modeling method in the second embodiment and the third embodiment.

As a preferred embodiment of the fourth embodiment, as shown in FIG9 , the cluster analysis module includes a cluster center submodule 701 , a classification submodule 801 , an edge sample submodule 901 and an iterative optimization submodule 1001 which are connected in sequence.

The cluster center submodule 701 is used to divide the feature vectors used to characterize the inverters in the photovoltaic power station into k clusters, and select k from all the feature vectors as cluster centers;

The classification submodule 801 is used to take each feature vector as a sample, calculate the distance between each sample and the k initial cluster centers according to the value of the clustering index corresponding to each sample, and then classify each sample into the cluster corresponding to the cluster center with the minimum distance;

The edge sample submodule 901 is used to calculate the number of edge samples of each cluster according to the edge sample percentage, and obtain the edge samples of each cluster;

The iterative optimization submodule 1001 is used to consider the edge samples of each cluster and iteratively optimize the initial cluster centers of all clusters until the difference between the sum of the distances of all samples to the initial cluster centers of their clusters between two adjacent iterations meets the preset requirements, thereby obtaining the cluster centers of all clusters.

As shown in FIG. 10 , the equivalent module includes an equivalent machine power submodule 1101 , a low voltage ride through control submodule 1201 , a box transformer submodule 1301 and a collector line submodule 1401 .

The equivalent machine power submodule 1101 is used to add the steady-state active power and steady-state reactive power of all inverters in each group of units to obtain the active power and reactive power of the equivalent machine, that is, the equivalent machine power; all inverters in each cluster are defined as a group of units;

The low voltage ride through control submodule 1201 is used to perform multivariate linear regression on the low voltage ride through active current and reactive current of all inverters in each unit group by using a gradient descent algorithm, obtain the low voltage ride through active current setting value Ip _{set_LV} , low voltage ride through active current control parameters K _1-Ip-LV and K _2-Ip-LV , low voltage ride through reactive current setting value Iq _{set_LV} and low voltage ride through reactive current control parameters K _1-Iq-LV and K _2-Iq-LV , and obtain the inverter low voltage ride through active current Ip _LVRT and inverter low voltage ride through reactive current Iq _LVRT as the control parameters during the low voltage ride through of the equivalent machine by the following formula:

Ip _LVRT ＝K _1-Ip-LV *Vt+K _2-Ip-LV *Ip ₀ +Ip _{set_LV}

Iq _LVRT ＝K _1-Iq-LV *(0.9-Vt)+K _2-Iq-LV *Iq ₀ +Iq _{set_LV}

Wherein, Vt is the terminal voltage during the fault period, Ip ₀ is the active current before entering the low voltage ride through, and Iq ₀ is the reactive current before entering the low voltage ride through;

Then the equivalent machine active current recovery rate parameter is obtained by the following formula: As the control parameters of the active current recovery of the low voltage ride-through of the equivalent machine:

Where m is the total number of inverters in the group. is the active current recovery rate parameter of the i-th inverter

I _n is the rated current of the inverter, is the slope of the line segment between the two data points at the steady-state start time t1 and the steady-state end time t2 on the active current curve after the i-th inverter recovers from fault:

is the inverter output active current at time t1 after the fault of the i-th inverter is restored, is the inverter output active current at time t2 after the fault of the i-th inverter is restored;

The box-type transformer submodule 1301 is used to obtain the total power consumption ΔP _Teq , equivalent impedance Z _Teq and equivalent capacity S _Teq of the transformer in the box-type transformer after equivalent value processing as the box-type transformer equivalent value processing result through the following formula:

ΔP _Teq ＝(NI _T ) ² Z _Teq

Z _Teq = Z _T /N

S _Teq = NS _T

Among them, N is the total number of box-type transformers, I _T is the current of each box-type transformer, Z _T is the impedance of the box-type transformer, and S _T is the capacity of a single box-type transformer;

The collector line submodule 1401 is used to obtain the collector line power loss ΔP _eq and the total impedance Z _eq after equalization as the collector line equalization processing result through the following formula:

ΔP _eq ＝(FI) ² Z _eq

Where I is the output current of the photovoltaic array, _Ze is the impedance of the e-th collector line, and F is the total number of collector lines.

Another embodiment of the present invention provides a computer device. The computer device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of the modules/units in the above-mentioned device embodiments are implemented.

The computer program may be divided into one or more modules/units, and the one or more modules/units are stored in the memory and executed by the processor to accomplish the present invention.

The computer device may be a desktop computer, a notebook computer, a palm computer, a cloud server, etc. The computer device may include, but is not limited to, a processor and a memory.

The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

The memory may be used to store the computer programs and/or modules, and the processor implements various functions of the computer device by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory.

If the module/unit integrated in the computer device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims (10)

1. A photovoltaic power station equivalent modeling method, characterized in that it includes the following steps: Perform cluster analysis on the feature vectors used to characterize the inverters in the photovoltaic power station to obtain the cluster centers of each cluster; Calculate the equivalent machine parameters of each cluster, and perform equivalent processing on the box-type transformer and collector line; According to the cluster center of each cluster, the equivalent machine parameters, and the equivalent processing results of the box transformer and the collector line, an equivalent model of the photovoltaic power station is obtained. 2. According to claim 1, a photovoltaic power station equivalent modeling method is characterized in that the cluster analysis is specifically: Step 1: divide the feature vectors used to characterize the inverters in the photovoltaic power station into k clusters, and select k from all feature vectors as initial cluster centers; Step 2: Take each feature vector as a sample, calculate the distance between each sample and the k initial cluster centers according to the value of the clustering index corresponding to each sample, and then divide each sample into the cluster corresponding to the sub-cluster center with the minimum distance; Step 3, according to the marginal sample percentage, calculate the number of marginal samples of each cluster, and obtain the marginal samples of each cluster; Step 4: Consider the edge samples of each cluster and iteratively optimize the initial cluster centers of all clusters until the difference between the sum of the distances of all samples to the initial cluster centers of their clusters between two adjacent iterations meets the preset requirements, thus obtaining the cluster centers of all clusters. 3. A photovoltaic power station equivalent modeling method according to claim 2, characterized in that: in step 2, the clustering indicators include the steady-state voltage average value Us during normal operation of the photovoltaic inverter, the active power average value Pe output by the inverter during the fault period, and the reactive power average value Qe output by the inverter during the fault period. 4. According to claim 2, a photovoltaic power station equivalent modeling method is characterized in that: in step 2, the distance between each sample and the centers of the k initial clusters is calculated by the following formula: Wherein, d _ij ( _xi , c _j ) is the Euclidean distance from sample _xi to the initial cluster center c _j , j = 1, 2, ..., k, u is a variable, _xiu is the value of the u-th clustering index of sample _xi , and c _ju is the value of the u-th clustering index of the initial cluster center c _j ; In step 4, the iterative optimization is specifically as follows: 1) Recalculate the new initial cluster center c′ _j by the following formula: Among them, g is the current number of samples in the jth cluster, and _Xm is the sample set consisting of all samples of the photovoltaic power station; M _aη represents the number of marginal samples in clusters other than cluster a; m _bη is the number of marginal samples of m _b samples in the bth cluster; Y _aη represents the marginal samples of other clusters except cluster a; 2) Return to step 2 and perform iterative optimization. After each iteration, calculate the sum P of the distances from all samples to the initial cluster center of their cluster clusters until the difference in P between two adjacent iterative calculations meets the preset requirements. The initial cluster center of each cluster cluster is used as the new cluster center after optimization. P is calculated by the following formula: 5. A photovoltaic power station equivalent modeling method according to claim 2, characterized in that: The equivalent machine parameters include the equivalent machine power, the equivalent machine low voltage ride-through period control parameters and the equivalent machine low voltage ride-through active current recovery control parameters; The equivalent machine parameters of each cluster are calculated, and the equivalent processing is performed on the box-type transformer and the collector line, specifically: 1) Define all inverters in each cluster as a unit group, add the steady-state active power and steady-state reactive power of all inverters in each unit group respectively, and obtain the active power and reactive power of the equivalent machine, that is, the equivalent machine power; 2) The gradient descent algorithm is used to perform multivariate linear regression on the low voltage ride through active current and reactive current of all inverters in each unit group, and the low voltage ride through active current setting value Ip _{set_LV} , low voltage ride through active current control parameters K _{1_Ip_LV} and K _{2_Ip_LV} , low voltage ride through reactive current setting value Iq _{set_LV} and low voltage ride through reactive current control parameters K _{1_Iq_LV} and K _{2_Iq_LV} are obtained. The inverter low voltage ride through active current Ip _LVRT and inverter low voltage ride through reactive current Iq _LVRT are obtained by the following formula as the control parameters of the equivalent machine during low voltage ride through: Ip _LVRT ＝K _{1_Ip_LV} *Vt+K _{2_Ip_LV} *Ip ₀ +Ip _{set_LV} Iq _LVRT ＝K _{1_Iq_LV} *(0.9-Vt)+K _{2_Iq_LV} *Iq ₀ +Iq _{set_LV} Wherein, Vt is the terminal voltage during the fault period, Ip ₀ is the active current before entering the low voltage ride through, and Iq ₀ is the reactive current before entering the low voltage ride through; 3) The equivalent machine active current recovery rate parameter is obtained by the following formula: As the control parameters of the active current recovery of the low voltage ride-through of the equivalent machine: Where m is the total number of inverters in the group. is the active current recovery rate parameter of the i-th inverter Where, _In is the rated current of the inverter, is the slope of the line segment between the two data points at the steady-state start time t1 and the steady-state end time t2 on the active current curve after the i-th inverter recovers from fault: in, is the inverter output active current at time t1 after the fault of the i-th inverter is restored, is the inverter output active current at time t2 after the fault of the i-th inverter is restored; 4) The total power consumption ΔP _Teq , equivalent impedance Z _Teq and equivalent capacity S _Teq of the transformer in the box-type transformer after equivalent treatment are obtained by the following formula as the box-type transformer equivalent treatment result: ΔP _Teq ＝(NI _T ) ² Z _Teq Z _Teq = Z _T /N S _Teq = NS _T Among them, N is the total number of box-type transformers, I _T is the current of each box-type transformer, Z _T is the impedance of the box-type transformer, and S _T is the capacity of a single box-type transformer; 5) The power loss ΔP _eq and total impedance Z _eq of the collector line after equalization are obtained by the following formula as the result of equalization processing of the collector line: ΔP _eq ＝(FI) ² Z _eq Where I is the output current of the photovoltaic array, _Ze is the impedance of the e-th collector line, and F is the total number of collector lines. 6. A photovoltaic power station equivalent modeling system, characterized by comprising: A cluster analysis module is used to perform cluster analysis on the characteristic vectors used to characterize the inverters in the photovoltaic power station to obtain the cluster center of each cluster; The equivalent module is used to calculate the equivalent machine parameters of each cluster and perform equivalent processing on the box-type transformer and the collector line; The modeling module is used to obtain an equivalent model of the photovoltaic power station according to the cluster center of each cluster, the equivalent machine parameters, and the equivalent processing results of the box transformer and the collector line. 7. A photovoltaic power station equivalent modeling system according to claim 6, characterized in that the cluster analysis module comprises: The cluster center submodule is used to divide the feature vectors used to characterize the inverters in the photovoltaic power station into k clusters, and select k from all the feature vectors as initial cluster centers; The classification submodule is used to take each feature vector as a sample, calculate the distance between each sample and the k initial cluster centers according to the value of the clustering index corresponding to each sample, and then divide each sample into the cluster corresponding to the cluster center with the minimum distance; The edge sample submodule is used to calculate the number of edge samples of each cluster according to the edge sample percentage, and obtain the edge samples of each cluster; The iterative optimization submodule is used to consider the edge samples of each cluster and iteratively optimize the initial cluster centers of all clusters until the difference between the sum of the distances of all samples to the initial cluster centers of their clusters between two adjacent iterations meets the preset requirements, thereby obtaining the cluster centers of all clusters. 8. According to claim 7, a photovoltaic power station equivalent modeling system is characterized in that: the equivalent machine parameters include equivalent machine power, equivalent machine low voltage ride-through period control parameters and equivalent machine low voltage ride-through active current recovery control parameters; The equivalent modules include: The equivalent machine power submodule is used to add the steady-state active power and steady-state reactive power of all inverters in each unit group to obtain the active power and reactive power of the equivalent machine, that is, the equivalent machine power; all inverters in each cluster are defined as a unit group; The low voltage ride through control submodule is used to perform multivariate linear regression on the low voltage ride through active current and reactive current of all inverters in each unit group by using the gradient descent algorithm, obtain the low voltage ride through active current setting value Ip _{set_LV} , low voltage ride through active current control parameters K _{1_Ip_LV} and K _{2_Ip_LV} , low voltage ride through reactive current setting value Iq _{set_LV} and low voltage ride through reactive current control parameters K _{1_Iq_LV} and K _{2_Iq_LV} , and obtain the inverter low voltage ride through active current Ip _LVRT and inverter low voltage ride through reactive current Iq _LVRT by the following formula as the control parameters of the equivalent machine during low voltage ride through: Ip _LVRT ＝K _{1_Ip_LV} *Vt+K _{2_Ip_LV} *Ip ₀ +Ip _{set_LV} Iq _LVRT =K _{1_Iq_LV} *(0.9-Vt)+K _{2_Iq_LV} *Iq ₀ +Iq _{set_LV} Wherein, Vt is the terminal voltage during the fault period, Ip ₀ is the active current before entering the low voltage ride through, and Iq ₀ is the reactive current before entering the low voltage ride through; Then the equivalent machine active current recovery rate parameter is obtained by the following formula: As the control parameters of the active current recovery of the low voltage ride-through of the equivalent machine: Where m is the total number of inverters in the group. is the active current recovery rate parameter of the i-th inverter Where, _In is the rated current of the inverter, is the slope of the line segment between the two data points at the steady-state start time t1 and the steady-state end time t2 on the active current curve after the fault of the i-th inverter is restored: in, is the inverter output active current at time t1 after the fault of the i-th inverter is restored, is the inverter output active current at time t2 after the fault of the i-th inverter is restored; The box-type transformer submodule is used to obtain the total power consumption ΔP _Teq , equivalent impedance Z _Teq and equivalent capacity S _Teq of the transformer in the box-type transformer after equivalent value processing as the box-type transformer equivalent value processing result through the following formula: ΔP _Teq ＝(NI _T ) ² Z _Teq Z _Teq = Z _T /N S _Teq = NS _T Among them, N is the total number of box-type transformers, I _T is the current of each box-type transformer, Z _T is the impedance of the box-type transformer, and S _T is the capacity of a single box-type transformer; The collector line submodule is used to obtain the collector line power loss ΔP _eq and total impedance Z _eq after equivalent value processing through the following formula as the collector line equivalent value processing result: ΔP _eq ＝(FI) ² Z _eq Where I is the output current of the photovoltaic array, _Ze is the impedance of the e-th collector line, and F is the total number of collector lines. 9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 5 when executing the computer program. 10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 5.