A doubly-fed wind farm equivalent modeling method based on multi-characteristic influence factors

Abstract

The application relates to a doubly-fed wind farm equivalent modeling method based on multi-feature influence factors, which comprises the following steps: S1, establishing a doubly-fed wind turbine voltage and current control model; S2, determining key factors influencing the voltage and current output of the doubly-fed wind turbine, i.e. feature influence factors, to form a data set; S3, based on the obtained data set, using a maximum-minimum distance method to perform primary clustering on a doubly-fed wind farm group; S4, using the primary clustering center and the curve similarity of each wind turbine to perform secondary clustering, and then performing doubly-fed wind farm equivalent operation after the secondary clustering. The method is beneficial to quickly and effectively classifying a large number of wind turbines in a wind farm and improving model accuracy.

Description

Technical Field

[0001] This invention belongs to the field of new energy power generation grid connection technology, specifically involving an equivalent modeling method for doubly fed wind farms based on multiple characteristic influencing factors. Background Technology

[0002] Currently, with the large-scale grid connection of wind turbines and the increased penetration rate of wind power, the fault characteristics of the power grid have been fundamentally affected. Traditional relay protection systems in the power industry are no longer fully applicable to the large-scale grid connection of wind turbines. Wind farm power sources differ from traditional synchronous generators. Large doubly-fed induction generator (DFIG) wind farms consist of dozens or even hundreds of turbines. The operating states of each turbine in a wind farm are not entirely identical, making it impossible to accurately represent the characteristics of the entire wind farm under the combined action of numerous turbines using a single turbine. Therefore, to analyze the impact of wind turbine transient characteristics under fault conditions on the safe and stable operation of the power system, it is necessary to construct a wind farm fault transient model.

[0003] Large-scale doubly-fed induction generator (DFIG) wind farms consist of dozens or even hundreds of wind turbines, including the turbines themselves and other electrical facilities. In simulation calculations, building a complete model of the wind turbines and then creating detailed models of each turbine to form the entire wind farm model is a common approach. While this model can accurately reflect the transient characteristics of the wind turbine cluster operating in grid-connected conditions under fault conditions, its complexity and computational burden make it unsuitable for practical engineering applications. Therefore, it is necessary to establish an equivalent model with a simple structure, easy computation, and accurate reflection of the transient behavior of the wind farm.

[0004] During grid faults, numerous factors influence the transient characteristics of turbine faults. Traditional methods employ a single clustering index for equivalent modeling, using factors such as geographical location, wind speed, and whether the crowbar is switched on or off as clustering indices. While this improves simulation speed, it introduces inaccuracies in characteristic simulation. Equivalent modeling of wind farms primarily involves two aspects: turbine clustering and turbine cluster equivalence. Turbine clustering requires determining characteristic influencing factors as its clustering index, while turbine cluster equivalence addresses the equivalence of electrical parameters of doubly-fed induction generator (DFIG) turbines and parameters of the wind farm's collection network. Transient equivalent modeling of DFIG wind farms has become crucial for analyzing wind farm output characteristics. It is significant for subsequent research on wind turbine fault ride-through control strategies, the impact of wind power integration on the stability of the receiving-end grid, and improving the grid connection efficiency of large-scale wind power systems. Therefore, establishing a simple and accurate equivalent model that reflects the transient behavior characteristics of wind farms has become an urgent problem to solve. Summary of the Invention

[0005] The purpose of this invention is to provide an equivalent modeling method for doubly fed wind farms based on multiple feature influencing factors. This method is beneficial for quickly and effectively classifying a large number of wind turbine units in a wind farm, thereby improving the accuracy of the model.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: an equivalent modeling method for doubly-fed wind farms based on multiple feature influencing factors, comprising the following steps:

[0007] S1. Establish a voltage and current control model for a doubly fed wind turbine;

[0008] S2. Identify the key factors affecting the voltage and current output of the doubly fed wind turbine, i.e., the characteristic influencing factors, and form a dataset;

[0009] S3. Based on the obtained dataset, the maximum-minimum distance method is used to perform the initial clustering of the doubly fed wind farm group;

[0010] S4. Perform secondary clustering using the initial cluster centers and the curve similarity of each wind turbine. After secondary clustering, perform equivalent calculations of the doubly-fed wind farm.

[0011] Furthermore, in step S1, the voltage and current control model of the doubly fed wind turbine is a control model that operates under low voltage fault conditions.

[0012] Furthermore, step S1 specifically includes the following steps:

[0013] S1.1, The output voltage model of the doubly fed wind turbine during operation is established as follows:

[0014] U N '=αU+Z[nI N +(n-1)I N-1 +…+I1];

[0015] In the formula, U N ' is the wind turbine voltage at point N after the fault, α is the voltage drop factor, Z is the impedance between adjacent wind turbine lines, I is the current in each branch of the wind turbine, U is the voltage at the grid connection point of the wind turbine, and n is the total number of wind turbines at point N from the grid connection point.

[0016] S1.2, The short-circuit current model of the doubly-fed induction generator output during the fault period is established as follows:

[0017]

[0018] In the formula, K represents the fault voltage drop amplitude, and U s ω is the fan outlet voltage when there is no fault. s For synchronous speed, L' s For the transient inductance of the stator winding, k r ω is the rotor coupling coefficient. r For rotor speed, For rotor flux linkage, T r ' is the transient component decay constant; It consists of three components: Part 1 The second part is the power frequency periodic component of the short-circuit current generated by the steady-state component of the stator flux linkage. It is the aperiodic component of the short-circuit current generated by the transient component of the stator flux linkage and continuously decaying, Part Three It is the non-periodic component of the short-circuit current generated by the transient component of the rotor flux linkage and continuously decaying; the value of the short-circuit current output by the doubly fed wind turbine during the fault is related to the distance between the wind turbine location and the point of common coupling, the voltage drop amplitude, the wind speed, and the maximum allowable current value of the converter. The distance between the point of common coupling, the voltage drop amplitude, the wind speed, and the maximum allowable current value of the converter are defined as characteristic influencing factors.

[0019] Furthermore, step S2 specifically includes the following steps:

[0020] S2.1 Determine the key factors affecting the voltage output of the doubly fed wind turbine. Under fault conditions, the degree of voltage drop at the outlet of the wind turbine at different locations in the wind farm is mainly affected by the spatial location of the wind turbine in the wind farm.

[0021] S2.2 Determine the key factors affecting the current output of the doubly fed wind turbine. The factors that determine the magnitude of the output current are the port voltage before the unit failure, the generated power, the wind speed, and the port voltage during the failure.

[0022] Furthermore, in step S3, the Euclidean distance between each sample in the dataset is first calculated, and then the maximum-minimum distance method is used to perform the initial clustering of wind turbines in the doubly-fed wind farm.

[0023] Furthermore, in step S4, a second clustering is performed using the curve similarity between the initial cluster centers and each wind turbine. The Fréchet distance of the curve similarity is:

[0024]

[0025] In the formula, A and B are the two curves to be calculated, d is the metric function, t is time, and α and β are the matrix data contained in the two curves, i.e., the characteristic influence factors.

[0026] Compared with the prior art, the present invention has the following beneficial effects: the method is conducive to quickly and effectively classifying a large number of wind turbine units in a wind farm, improving the accuracy of the model, and is of great significance for improving the stability of the receiving-end power grid and improving the grid connection efficiency of large-scale wind power systems. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating the method implementation of an embodiment of the present invention;

[0028] Figure 2 This is a comparison chart of the equivalent results of the wind farm and the detailed model simulation in the embodiments of the present invention. Detailed Implementation

[0029] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0030] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0031] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0032] like Figure 1 As shown, this embodiment provides an equivalent modeling method for doubly-fed wind farms based on multiple feature influence factors, including the following steps:

[0033] S1. Establish a voltage and current control model for a doubly fed wind turbine;

[0034] S2. Identify the key factors affecting the voltage and current output of the doubly fed wind turbine, i.e., the characteristic influencing factors, and form a dataset;

[0035] S3. Based on the obtained dataset, the maximum-minimum distance method is used to perform the initial clustering of the doubly fed wind farm group;

[0036] S4. Perform secondary clustering using the initial cluster centers and the curve similarity of each wind turbine. After secondary clustering, perform equivalent calculations of the doubly-fed wind farm.

[0037] In this embodiment, the voltage and current control model of the doubly fed wind turbine is a control model that operates under low voltage fault conditions. Step S1 specifically includes the following steps:

[0038] S1.1, The output voltage model of the doubly fed wind turbine during operation is established as follows:

[0039] U N '=αU+Z[nI N +(n-1)I N-1 +…+I1];

[0040] In the formula, U N' is the wind turbine voltage at point N after the fault, α is the voltage drop factor, Z is the impedance between adjacent wind turbine lines, I is the current in each branch of the wind turbine, U is the voltage at the wind turbine grid connection point, and n is the total number of wind turbines at point N from the grid connection point.

[0041] S1.2, The short-circuit current model of the doubly-fed induction generator output during the fault period is established as follows:

[0042]

[0043] In the formula, K represents the fault voltage drop amplitude, and U s ω is the fan outlet voltage when there is no fault. s For synchronous speed, L' s For the transient inductance of the stator winding, k r ω is the rotor coupling coefficient. r For rotor speed, For rotor flux linkage, T r ' is the transient component decay constant; It consists of three components: Part 1 The second part is the power frequency periodic component of the short-circuit current generated by the steady-state component of the stator flux linkage. It is the aperiodic component of the short-circuit current generated by the transient component of the stator flux linkage and continuously decaying, Part Three It is the non-periodic component of the short-circuit current generated by the transient component of the rotor flux linkage and continuously decaying; the value of the short-circuit current output by the doubly fed wind turbine during the fault is related to the distance between the wind turbine location and the point of common coupling, the voltage drop amplitude, the wind speed, and the maximum allowable current value of the converter. The distance between the point of common coupling, the voltage drop amplitude, the wind speed, and the maximum allowable current value of the converter are defined as characteristic influencing factors.

[0044] The location of the wind turbine affects the line impedance, i.e., the value of n in step S1.1. If the wind turbine is very close to the grid connection point, the value of n is very small, and the sum of the impedances is also very small; conversely, if the wind turbine is very far from the grid connection point, n will be very large, and the sum of the impedances will increase.

[0045] The reference frames for the output voltage model during doubly-fed induction generator (DFIG) operation and the output short-circuit current model during a fault are different. The current model is studied for a single turbine. The voltage model is for the entire wind farm; due to the influence of the collector lines, the voltage at the turbine outlet will be inconsistent. Un and Us in both the voltage and current models are the same for the same turbine.

[0046] Step S2 specifically includes the following steps:

[0047] S2.1 Determine the key factors affecting the voltage output of doubly-fed wind turbines. Under fault conditions, the degree of voltage drop at the outlet of wind turbine units at different locations within the wind farm is mainly affected by the spatial location of the wind turbine units within the wind farm.

[0048] S2.2 Determine the key factors affecting the current output of the doubly fed wind turbine. The factors that determine the magnitude of the output current are the port voltage before the unit failure, the generated power, the wind speed, and the port voltage during the failure.

[0049] In step S3, the Euclidean distance between samples in the dataset is first calculated, and then the maximum-minimum distance method is used to perform the initial clustering of wind turbines in the doubly fed wind farm.

[0050] In step S4, secondary clustering is performed using the curve similarity between the initial cluster centers and each wind turbine. The Frechet distance of the curve similarity is:

[0051]

[0052] In the formula, A and B are the two curves to be calculated, d is the metric function, t is time, and α and β are the matrix data contained in the two curves, i.e., the characteristic influence factors.

[0053] In this embodiment, Matlab / Simulink is used for simulation. The wind farm is set to consist of 16 doubly-fed induction generators, and a constant wind speed scenario of 15 m / s is set, with the wind direction from north to south. A fault occurs on the grid side, which occurs at 0.3 s and lasts for 0.6 s. The fault voltage drops to 0.5 pu. The port voltage, power output, wind speed, and port voltage data of each wind turbine before the fault are collected as initial values.

[0054] When performing maximum-minimum distance clustering, changing the value of the criterion theta yields the following clustering results:

[0055] 1. Theta = 0.5

[0056] Category 1: 1, 2, 3, 5, 6, 9, 13;

[0057] Category 2: 4, 7, 8, 10, 11, 12, 14, 15, 16.

[0058] 2. Theta = 0.3

[0059] Category 1: 1, 5; Cluster center is 1;

[0060] Category 2: 8, 11, 12, 14, 15, 16; Cluster center is 16;

[0061] The third category consists of clusters 2, 3, 4, 6, 7, 9, 10, and 13; the cluster center is 3.

[0062] 3. Theta = 0.2

[0063] Category 1: 1, 5;

[0064] Category 2: 12, 16;

[0065] Category 3: 2, 3, 6, 9, 13;

[0066] Category 4: 4, 7, 8, 10, 11, 14, 15.

[0067] The three types of wind turbines obtained with Theta = 0.3 were selected as the results of the initial clustering of wind turbines in the doubly-fed wind farm using the maximum-minimum distance method.

[0068] Secondary clustering is performed using the curve similarity between the initial cluster centers and each wind turbine. The return value of the secondary clustering is ans; the smaller the ans value, the higher the similarity. The return values ​​of the secondary clustering are shown in the table below:

[0069]

[0070] The clustering was adjusted based on the curve similarity (ans) values, resulting in the following secondary clustering results:

[0071] Category 1: 1, 5;

[0072] Category 2: 8, 11, 12, 13, 14, 15, 16;

[0073] Category 3: 2, 3, 4, 6, 7, 9, 10.

[0074] By using clustering results to perform equivalence, the wind farm is equivalent to three wind turbines. Through simulation, it is found that the accuracy of the doubly fed wind farm equivalence modeling method based on multiple feature influence factors is relatively high, and the error is smaller than that of the traditional single multiplication model of equivalence.

[0075] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for equivalent modeling of doubly-fed wind farms based on multiple feature influencing factors, characterized in that, Includes the following steps: S1. Establish a voltage and current control model for a doubly fed wind turbine; S2. Identify the key factors affecting the voltage and current output of the doubly fed wind turbine, i.e., the characteristic influencing factors, and form a dataset; S3. Based on the obtained dataset, the maximum-minimum distance method is used to perform the initial clustering of the doubly fed wind farm group; S4. Perform secondary clustering using the curve similarity between the initial cluster centers and each wind turbine, and then perform equivalent calculations for the doubly-fed wind farm after the secondary clustering. Step S1 specifically includes the following steps: S1.1, The output voltage model of the doubly fed wind turbine during operation is established as follows: ； In the formula, U N ’ α is the voltage of the wind turbine at point N after the fault, Z is the impedance between adjacent wind turbine lines, I is the current of each branch of the wind turbine, U is the voltage of the wind turbine at the grid connection point, and n is the total number of wind turbines at point N from the grid connection point. S1.2, The short-circuit current model of the doubly-fed induction generator output during the fault period is established as follows: In the formula, K represents the fault voltage drop amplitude. For synchronous speed, L' s For the transient inductance of the stator winding, The rotor coupling coefficient is... For rotor speed, For rotor flux linkage, The transient component decay constant; It consists of three components: Part 1 The second part is the power frequency periodic component of the short-circuit current generated by the steady-state component of the stator flux linkage. It is the aperiodic component of the short-circuit current generated by the transient component of the stator flux linkage and continuously decaying, Part Three It is the non-periodic component of the short-circuit current generated by the transient component of the rotor flux linkage and continuously decaying; the short-circuit current output by the doubly fed wind turbine during the fault is related to the distance between the wind turbine location and the point of common coupling, the voltage drop amplitude, the wind speed, and the maximum allowable current value of the converter. The distance between the point of common coupling, the voltage drop amplitude, the wind speed, and the maximum allowable current value of the converter are defined as characteristic influencing factors. In step S4, secondary clustering is performed using the curve similarity between the initial cluster centers and each wind turbine. The Fréchet distance of the curve similarity is: In the formula, A and B are the two curves to be calculated, d is the metric function, t is time, and α and β are the matrix data contained in the two curves, i.e., the characteristic influence factors.

2. The method for equivalent modeling of doubly-fed wind farms based on multiple feature influencing factors according to claim 1, characterized in that, In step S1, the voltage and current control model of the doubly fed wind turbine is a control model that operates under low voltage fault conditions.

3. The method for equivalent modeling of doubly-fed wind farms based on multiple feature influencing factors according to claim 1, characterized in that, Step S2 specifically includes the following steps: S2.1 Determine the key factors affecting the voltage output of the doubly fed wind turbine. Under fault conditions, the degree of voltage drop at the outlet of the wind turbine at different locations in the wind farm is mainly affected by the spatial location of the wind turbine in the wind farm. S2.2 Determine the key factors affecting the current output of the doubly fed wind turbine. The factors that determine the magnitude of the output current are the port voltage before the unit failure, the generated power, the wind speed, and the port voltage during the failure.

4. The method for equivalent modeling of doubly-fed wind farms based on multiple feature influencing factors according to claim 3, characterized in that, In step S3, the Euclidean distance between samples in the dataset is first calculated, and then the maximum-minimum distance method is used to perform the initial clustering of wind turbines in the doubly fed wind farm.

Images