Equivalent Modeling Method for Large Wind Farms Based on the Power Loss Equivalence Principle

Abstract

This invention discloses an equivalent modeling method for large-scale wind farms based on the power loss equivalence principle. First, a model of the actual wind farm is built. Then, initial values ​​of the characteristic state variables of each wind turbine are calculated based on the power-wind speed and power-speed curves of each turbine. Next, based on the matrix formed by the initial values, the K-means clustering algorithm is used to calculate the wind farm grouping results. Finally, the equivalent parameters are calculated to perform single-unit equivalent modeling of the turbines in the same group, completing the equivalent modeling of the entire wind farm. This wind farm equivalent modeling method fully considers the requirements of the power loss equivalence principle, has low complexity, and can be used in practical engineering implementation.

Description

Technical Field

[0001] This invention relates to the equivalent modeling parameters of wind farms, specifically to an equivalent modeling method for large-scale wind farms based on the principle of power loss equivalence. Background Technology

[0002] With the continuous increase in installed capacity of wind farms, studying the stability of large-scale wind power grid connection is of great value. Wind farms are characterized by small single-unit capacity, complex wiring structures, and high single-unit model order. To reduce the system order while retaining the corresponding model accuracy, it is necessary to simplify the wind farm using equivalent methods. However, there is currently no unified and mature methodology for wind farm modeling, especially in the analysis of wind farm grid connection stability. The commonly used method is the single-unit equivalent method, which makes many assumptions, ignores many influencing factors, and suffers from insufficient accuracy. Summary of the Invention

[0003] The purpose of this invention is to provide an equivalent modeling method for large-scale wind farms based on the principle of power loss equivalence, laying the foundation for optimizing the small-signal stability analysis of wind farm grid-connected systems.

[0004] The technical solution to achieve the purpose of this invention is as follows: a method for equivalent modeling of large-scale wind farms based on the principle of power loss equivalence, comprising the following steps:

[0005] Step 1: Build an actual wind farm model. The wind farm mainly uses doubly fed wind turbines. The wind speed received by the wind turbines in different areas is different, and the wind turbines in the same area are affected by the wake to varying degrees.

[0006] Step 2: Select wind speed, slip rate, and active power as characteristic state variables of the wind turbine. Given the wind speed received by the wind turbine, obtain the active power generated from the wind turbine power-wind speed curve, obtain the wind turbine speed from the power-speed curve, and then calculate the slip rate. Use the values ​​obtained above as the initial values ​​of the characteristic state variables.

[0007] Step 3: Generate a state variable matrix from the initial values ​​obtained in Step 2, and apply the K-means clustering algorithm to cluster the wind farm. The value of k is estimated based on the size of the wind farm area and the actual environment, and its value is the final equivalent number of machines.

[0008] Step 4: Calculate the equivalent machine parameters of the fans in the same group according to the equivalent parameter calculation method;

[0009] Step 5: Based on the calculation results in Step 4, complete the equivalent modeling of the wind farm.

[0010] Furthermore, in step 3, the K-means clustering algorithm is applied to cluster the wind farms. The specific method is as follows:

[0011] 1) Randomly select k data points from the state variable matrix as the mean vector of the initial cluster;

[0012] 2) For each remaining data point in the state variable matrix, assign it to the nearest cluster based on its distance to the mean vectors of each cluster. The distance is calculated using the Euclidean distance formula. Let the i-th variable in the variable matrices X and Y be x. i and y i The formula for calculating distance d is as follows:

[0013]

[0014] 3) Update the mean vectors of the k newly generated clusters;

[0015] 4) Repeat steps 2) and 3) until the objective function converges, x i For sample data, c i Let be the centroid, specifically the mean of all data. Then the objective function E is as follows:

[0016]

[0017] Furthermore, in step 4, the equivalent machine parameters of the fans within the same group are calculated according to the equivalent parameter calculation method. The specific method is as follows:

[0018] If there are N wind turbines in this group, then the capacity weighting coefficient σ i for:

[0019]

[0020] In the formula, S i Let P be the capacity of the i-th wind turbine in the wind farm. mi and P ei Let S be the input power and electromagnetic power of the i-th fan, then the equivalent fan capacity S is... eq Input power P meq and electromagnetic power P eeq as follows:

[0021]

[0022] Let X i Let X be the internal parameters of the i-th wind turbine, including electrical and mechanical parameters. The electrical parameters include the generator's internal parameters (rotor resistance and rotor inductance), and the mechanical parameters include the generator's time inertia and the blade's time inertia. Then, the internal parameters X of the equivalent machine are... eq as follows:

[0023]

[0024] The DC link voltage of the equivalent fan is the same as that of a single fan. The DC link capacitance should be the sum of the DC link capacitances of all fans. Furthermore, the filter capacitance at the grid connection point of the equivalent fan should also be the sum of the capacitances of all fans. Let the DC link capacitance of the i-th fan be C. i Equivalent capacitor C eq for:

[0025]

[0026] For a trunk-type connection structure, assuming the wind farm has m rows and n columns, let ΔU be the voltage drop between the outlet voltage of the i-th wind turbine in the l-th row (belonging to group k) and the grid connection point voltage. li The current is I li The impedance of the j-th segment is Z. lj If the grid connection voltage is U, then the power loss ΔS of the i-th wind turbine belonging to group k in row l is... li for:

[0027]

[0028] The total power loss of the wind turbines belonging to group k in row l is:

[0029]

[0030] The equivalent power loss ΔS of the l-th row of wind turbines belonging to group k lk-eq for:

[0031]

[0032] Based on the principle of equivalent line losses before and after the equivalent value, the calculation is performed by combining the above two equations to obtain the equivalent impedance Z of the wind turbines belonging to group k in the l-th row. lk-eq for:

[0033]

[0034] Therefore, the equivalent power loss ΔS of the kth group of wind turbines in the entire wind farm is obtained. k-eq for:

[0035]

[0036] The equivalent impedance Z of the k-th group of wind turbines k-eq for:

[0037]

[0038] A large-scale wind farm equivalent modeling system based on the power loss equivalence principle is provided. Based on the aforementioned large-scale wind farm equivalent modeling method, the system realizes the equivalent modeling of large-scale wind farms based on the power loss equivalence principle.

[0039] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs equivalent modeling of a large wind farm based on the power loss equivalence principle using the aforementioned large wind farm equivalent modeling method.

[0040] A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it realizes equivalent modeling of a large wind farm based on the principle of power loss equivalence, according to the aforementioned equivalent modeling method for large wind farms.

[0041] Compared with the prior art, the significant advantages of this invention are: 1) It fully considers the influence of specific factors such as wind speed, terrain, and wind turbine characteristics on wind farms, and has high accuracy in equivalence measurement, fully considering the power loss equivalence principle; 2) The parameter equivalence method has low complexity and can be used for practical engineering implementation. Attached Figure Description

[0042] Figure 1 This is a flowchart of an equivalent modeling method for large-scale wind farms based on the principle of power loss equivalence.

[0043] Figure 2 This is the equivalent circuit topology diagram of a single wind turbine.

[0044] Figure 3 This is a control block diagram of a single wind turbine rotor-side converter.

[0045] Figure 4 This is a circuit diagram of an equivalent wind farm.

[0046] Figure 5 This is a comparison chart of the simulated waveforms of an equivalent wind farm and the simulated waveforms of an actual wind farm. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] like Figure 1 As shown, the equivalent modeling method for large-scale wind farms based on the power loss equivalence principle includes the following steps:

[0049] Step 1: Build an actual wind farm model. The equivalent circuit topology of a single wind turbine is shown below. Figure 2 As shown, a doubly-fed induction generator (DFIG) wind turbine is used. The turbine functions through maximum power point tracking (MPPT), rotor-side control, and grid-side control. The control block diagram of the turbine rotor-side converter is shown below. Figure 3As shown, constant active power and constant reactive power control methods are adopted. The wind speed received by the wind turbines in different areas is different, and the degree of influence of wake on the wind turbines in the same area is also different. The actual wind speed received should take the wake effect into account.

[0050] Step 2: Select wind speed, slip, and active power as characteristic state variables of the wind turbine. The wind speed received by the wind turbine is known in Step 1. The active power generated is obtained from the wind turbine power-wind speed curve. The wind turbine speed is obtained from the power-speed curve. Finally, the slip can be calculated. The values ​​obtained above are used as the initial values ​​of the characteristic state variables.

[0051] Step 3: Generate a state variable matrix from the initial values ​​obtained in Step 2, and apply the K-means clustering algorithm to cluster the wind farm. The value of k is 4, which is equivalent to 4 equivalent machines.

[0052] Step 4: Calculate the equivalent machine parameters of the fans in the same group according to the equivalent parameter calculation method.

[0053] Step 5: Based on the calculation results in Step 4, connect the four equivalent turbines in parallel to obtain the circuit diagram of the equivalent wind farm, as shown below. Figure 4 Complete the equivalent modeling of the wind farm.

[0054] Example

[0055] To verify the effectiveness of the method of the present invention, the following experiment was conducted.

[0056] The specific parameters of the wind turbines are shown in Table 1. The wind speeds received by the four equivalent wind turbines are 8 m / s, 11 m / s, 12 m / s, and 14 m / s, respectively. Comparing the waveforms of active power, reactive power, and voltage between the actual wind farm and the equivalent wind farm, as shown in Table 1... Figure 5 As shown in the figure, the error is small and the equivalent accuracy is high, making it suitable for problems such as the grid connection stability analysis of wind farms.

[0057] Table 1 Main Parameters of the Fan

[0058]

[0059] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0060] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for equivalent modeling of large-scale wind farms based on the principle of power loss equivalence, characterized in that, The steps are as follows: Step 1: Build an actual wind farm model. The wind turbines in this wind farm include doubly fed wind turbines. The wind turbines in different areas receive different wind speeds, and the wind turbines in the same area are affected by wakes to varying degrees. Step 2: Select wind speed, slip rate, and active power as characteristic state variables of the wind turbine. Given the wind speed received by the wind turbine, obtain the active power generated from the wind turbine power-wind speed curve, obtain the wind turbine speed from the power-speed curve, and then calculate the slip rate. Use the values ​​obtained above as the initial values ​​of the characteristic state variables. Step 3: Generate a state variable matrix from the initial values ​​obtained in Step 2, and apply the K-means clustering algorithm to cluster the wind farm. The value of k is estimated based on the size of the wind farm area and the actual environment, and its value is the final equivalent number of machines. Step 4: Calculate the equivalent machine parameters of the fans within the same group according to the equivalent parameter calculation method. The specific method is as follows: If there are N wind turbines in this group, then the capacity weighting coefficient σ i for: In the formula, S i Let P be the capacity of the i-th wind turbine in the wind farm. mi and P ei Let S be the input power and electromagnetic power of the i-th fan, then the equivalent fan capacity S is... eq Input power P meq and electromagnetic power P eeq as follows: Let X i Let X be the internal parameters of the i-th wind turbine, including electrical and mechanical parameters. The electrical parameters include the generator's internal parameters (rotor resistance and rotor inductance), and the mechanical parameters include the generator's time inertia and the blade's time inertia. Then, the internal parameters X of the equivalent machine are... eq as follows: The DC link voltage of the equivalent fan is the same as that of a single fan. The DC link capacitance should be the sum of the DC link capacitances of all fans. Furthermore, the filter capacitance at the grid connection point of the equivalent fan should also be the sum of the capacitances of all fans. Let the DC link capacitance of the i-th fan be C. i Equivalent capacitor C eq for: For a trunk-type connection structure, assuming the wind farm has m rows and n columns, let ΔU be the voltage drop between the outlet voltage of the i-th wind turbine in the l-th row (belonging to group k) and the grid connection point voltage. li The current is I li The impedance of the j-th segment is Z. lj If the grid connection voltage is U, then the power loss ΔS of the i-th wind turbine belonging to group k in row l is... li for: The total power loss of the wind turbines belonging to group k in row l is: The equivalent power loss ΔS of the l-th row wind turbine belongs to group k. lk-eq for: Based on the principle of equivalent line losses before and after the equivalent value, the calculation is performed by combining the above two equations to obtain the equivalent impedance Z of the wind turbines belonging to group k in the l-th row. lk-eq for: Therefore, the equivalent power loss ΔS of the kth group of wind turbines in the entire wind farm is obtained. k-eq for: The equivalent impedance Z of the k-th group of wind turbines k-eq for: Step 5: Based on the calculation results in Step 4, complete the equivalent modeling of the wind farm.

2. The method for equivalent modeling of large-scale wind farms based on the principle of power loss equivalence as described in claim 1, characterized in that, In step 3, the K-means clustering algorithm is applied to cluster the wind farms. The specific method is as follows: 1) Randomly select k data points from the state variable matrix as the mean vector of the initial cluster; 2) For each remaining data point in the state variable matrix, assign it to the nearest cluster based on its distance to the mean vectors of each cluster. The distance is calculated using the Euclidean distance formula. Let the i-th variable in the variable matrices X and Y be x. i and y i The formula for calculating distance d is as follows: 3) Update the mean vectors of the k newly generated clusters; 4) Repeat steps 2) and 3) until the objective function converges, x i For sample data, c i Let be the centroid, specifically the mean of all data. Then the objective function E is as follows:

3. A large-scale wind farm equivalent modeling system based on the power loss equivalence principle, characterized in that, Based on the equivalent modeling method for large wind farms according to any one of claims 1-2, equivalent modeling of large wind farms based on the principle of power loss equivalence is realized.

4. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it performs equivalent modeling of a large wind farm based on the power loss equivalence principle according to the equivalent modeling method for large wind farms as described in any one of claims 1-2.

5. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it realizes equivalent modeling of a large wind farm based on the power loss equivalence principle, according to the equivalent modeling method for large wind farms as described in any one of claims 1-2.

Images