Wind power plant position layout optimization method considering wake effect

Abstract

The invention provides a wind power plant position layout optimization method considering a wake effect, which comprises the following steps of: 1, establishing a wind power plant model: designing a wind power plant into a square, and enabling one diagonal line of the wind power plant to be parallel to an X axis; a wind power plant is divided into M cells used for containing wind turbine generators, the cells are square, and one diagonal line of each cell is parallel to the X axis; randomly placing the N wind turbine generators in the corresponding N cells; n < M; step 2, obtaining an optimal wind power plant position layout: under the condition that the number, the wind direction and the wind speed of the wind power plant are determined, optimizing the placement position of each wind power plant by using a particle swarm algorithm by taking the maximum overall output power of the wind power plant as an optimization target; and the optimized placement positions of all the wind turbine generators form an optimal wind power plant position layout. The method is suitable for different wind directions and the situation that wind turbine generators are randomly placed in the wind power plant, and the wake effect is considered to optimize the position layout of the wind power plant.

Description

Technical Field

[0001] The present invention belongs to the field of wind farms, and in particular relates to a method for optimizing the location layout of wind farms taking wake effects into consideration. Background Art

[0002] Wind energy is one of the most commercially promising and dynamic renewable energy sources, known for its clean, low-cost, and inexhaustible supply. Wind power generation offers significant advantages, including significant potential for installed capacity growth, rapidly declining costs, safety, and an inexhaustible energy source. The wake effect occurs when a wind turbine extracts energy from the wind, creating a downstream wake zone with decreasing wind speed. If a downstream wind turbine is located within this wake zone, the input wind speed to the downstream wind turbine will be lower than that of the upstream wind turbine. The wake effect within a wind farm during wind power generation prevents grid-connected wind power systems from achieving higher output power. This not only limits the proportion of wind power connected to the grid but also affects the accuracy of power system frequency control. Furthermore, the wake effect can cause wind turbine oscillations, leading to subsynchronous oscillations in the wind turbine-HVDC Flexible transmission system, which can cause short circuits in the HVDC Flexible transmission system and lead to system failure. Therefore, optimizing the location and layout of wind farms is crucial. Unoptimized wind farm layouts increase the wind farm wake effect and mechanical wear, leading to a decrease in wind farm output power, exacerbating the risk of system frequency instability, turbine oscillations, subsynchronous oscillations in the HVDC Flexible system, and even short-circuit failures. To determine the optimal placement of wind turbines and reduce power fluctuations caused by the wake effect, existing technologies have proposed a dual-objective layout optimization model for multiple wind farms that considers sequential fluctuations in wind power. This minimizes wind power fluctuations, but lacks verification of its applicability under different wind directions. Furthermore, the wind farm power ratio is constant, and the wake effect of wind farm power at different ratios is not considered. Nor is the wind farm wake effect considered in conjunction with factors such as HVDC Flexible system oscillations and short-circuit circuits.

[0003] When designing a wind farm, by considering the geographic location and installation position of each wind turbine, the impact of the wake effect can be mitigated, if not completely eliminated. Minimizing the wake effect within a limited area is a key condition for maximizing the proportion of wind power connected to the grid. Existing optimization strategies have problems determining the optimal number and location of wind turbines in a wind farm. They only directly or indirectly reduce the wake effect at a certain localized location within the wind farm and are not applicable to scenarios where wind turbines are randomly placed within the wind farm. Summary of the Invention

[0004] The purpose of this invention is to provide a wind farm layout optimization method that considers wake effects. This method is applicable to situations with different wind directions and random placement of wind turbines within the wind farm. It is also applicable to situations with different wind farm power ratios. It also comprehensively considers the wake effects on oscillation, short circuit, and other constraints of the flexible DC transmission system to optimize the wind farm layout. The technical solution adopted is:

[0005] A method for optimizing wind farm location layout considering wake effects comprises the following steps:

[0006] Step 1: Establish a wind farm model:

[0007] The wind farm is designed as a square with one diagonal parallel to the X-axis;

[0008] The wind farm is divided into M cells for placing wind turbines, where each cell is a square and one diagonal of the cell is parallel to the X-axis;

[0009] Randomly place N wind turbines in the corresponding N cells; N<M;

[0010] Step 2: Obtain the optimal wind farm location layout:

[0011] When the number of wind turbines, wind direction and wind speed are determined, the wake effect on the oscillation and short-circuit constraints of the flexible DC transmission system is comprehensively considered. With the maximum output power of the wind farm as the optimization goal, the particle swarm algorithm is used to optimize the placement of each wind turbine. The optimized placement of all wind turbines constitutes the optimal wind farm location layout.

[0012] Preferably, in step 2, the wind direction includes a constant wind direction and a random wind direction.

[0013] Preferably, in step 2, the output power expression of the wind turbine is:

[0014]

[0015] ρ-air density; R-wind rotor radius; v-wind speed; C p -Power coefficient, which is a function of the tip speed ratio λ and the pitch angle β.

[0016] Preferably, in step 2, within the wake effect tolerance, the number of wind turbines is related to the wind power grid connection ratio.

[0017] Preferably, the wind speed in step 1 obeys Weibull distribution.

[0018] Preferably, step 2 further includes the following steps: simulating the output power of all wind turbines in the wind farm to verify the optimized wind farm location layout.

[0019] Preferably, the proportion of wind farms connected to the power grid is further considered in obtaining the optimal wind farm location layout.

[0020] Preferably, the constraints for obtaining the optimal wind farm location layout include: no wind turbine oscillation, subsynchronous oscillation of the flexible HVDC system, and no short circuit fault.

[0021] Preferably, the optimal wind farm location layout should be obtained within the wake effect tolerance limit.

[0022] Preferably, the proportion of wind farms connected to the grid and the wake effect satisfy the following relationship:

[0023] |λI-H(s,v0)|=0

[0024] Where H(s,v0) varies with the wind power grid connection ratio s and the wake effect; set the characteristic value λ i =σ i +jω i (i=1,2,…,n), the real part reflects the damping intensity, and the imaginary part reflects the frequency drop degree; v0 is the wind speed.

[0025] Compared with the prior art, the advantages of the present invention are:

[0026] 1. The placement of wind turbines can be changed according to different wind directions to reduce the wind turbine wake effect and avoid wind turbine resonance.

[0027] 2. The placement of wind turbines can be changed according to different wind directions, thereby maximizing the overall output power of the wind farm.

[0028] 3. The placement of wind turbines can be changed according to different wind directions. While maximizing the overall output power of the wind farm, the subsynchronous oscillation of the flexible DC transmission system can be suppressed, thereby preventing short-circuit current faults.

[0029] 4. When wind turbines are randomly placed inside the wind farm, the placement is optimized with multiple objectives such as minimizing the wind turbine wake effect, maximizing the output power, and minimizing the short-circuit current.

[0030] 5. The optimization of wind farm location also takes into account wind speed, wind direction and the proportion of wind power connected to the grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The wind farm model diagram established in step 1;

[0032] Figure 2 Optimize the flow chart for WFLO;

[0033] Figure 3(a) shows the WFLO optimization results under constant wind speed and wind direction in this embodiment;

[0034] Figure 3(b) shows the WFLO optimization results at constant wind speed and wind direction using the existing technology;

[0035] Figure 4 This is a comparison chart of output power at different wind turbine positions under constant wind speed and wind direction;

[0036] Figure 5(a) shows the WFLO optimization results for random wind speed and wind direction in this embodiment;

[0037] Figure 5(b) shows the WFLO optimization results under random wind speed and wind direction using the existing technology;

[0038] Figure 6 The output power comparison chart of different wind turbine positions under random wind speed and wind direction;

[0039] Figure 7 This is the Jensen wake model diagram, which shows the impact of the upstream wind turbine's wake effect on the wind speed of the downstream wind turbine;

[0040] Figure 8 It is the characteristic root locus when the wind power access ratio increases from 20% to 60%. DETAILED DESCRIPTION

[0041] The following is a more detailed description of the wind farm location layout optimization method that considers wake effects, using schematic diagrams. These schematic diagrams illustrate preferred embodiments of the present invention. It should be understood that those skilled in the art may modify the present invention described herein while still achieving the beneficial effects of the present invention. Therefore, the following description should be understood as a general guide for those skilled in the art and not as a limitation of the present invention.

[0042] A method for optimizing wind farm location layout considering wake effects comprises the following steps:

[0043] Step 1: Establish a wind farm model.

[0044] The wind farm is designed as a square with one diagonal parallel to the X-axis.

[0045] The wind farm is divided into M cells for placing wind turbines, where each cell is a square and one diagonal of the cell is parallel to the X-axis;

[0046] Randomly place N wind turbines in corresponding N cells. N < M.

[0047] That is, a wind turbine is placed in one cell.

[0048] The natural flow of wind has random probability characteristics. They have different speeds, densities and directions, so they should be fully considered when planning the layout of wind farms. At any specific wind speed and direction, the layout of wind farms will present a specific arrangement that may produce the best output power. Using the probability density function P U (x), the wind speed follows a specific Weibull distribution as follows:

[0049]

[0050] Where: c is the scale parameter; z is the shape parameter.

[0051] A square wind farm location with a size of l×l is used. The wind farm location diagram is a top view of the wind farm location. In order to ensure that it can face the wind speed to the greatest extent, it is tilted towards the wind direction. The square wind farm location is divided into a certain number of cells to represent the possible placement positions of all wind turbines in the wind farm. The placement positions are numbered as follows: Figure 1 shown.

[0052] from Figure 1 It can be seen that one diagonal of the wind farm is horizontally located on the x-axis. Only when the wind direction is perpendicular to the diagonal axis can the wind farm be considered to have the optimal layout. This makes the wind farm placed in a diagonal form have a wider surface area facing the high wind direction and maximizes the overall output power of the wind farm.

[0053] The system eigenvalue λ is solved according to the following formula.

[0054] |λI-H(s,v0)|=0

[0055] Among them, H(s,v0) varies with the wind power grid connection ratio s and the influence of the wake effect, thus affecting the characteristic root of the system solution. Set the characteristic value λ i =σ i +jω i (i = 1, 2, ..., n), where the real part reflects the damping strength and the imaginary part reflects the degree of frequency drop. When the eigenvalue is a conjugate complex root, the system frequency is at risk of instability. When the real part of the eigenvalue is positive and the damping ratio is negative, the system frequency is completely unstable. Where v0 is the wind speed.

[0056] In view of the increasing proportion of wind power connected to the grid under the influence of the internal wake effect of wind farms, the frequency instability occurs when the grid system state changes. Figure 8 The horizontal axis represents the real part of the characteristic root, and the imaginary axis represents the imaginary part of the characteristic root, indicating the frequency stability state.

[0057] Step 2: Obtain the optimal wind farm location layout:

[0058] When the number of wind turbines, wind direction and wind speed are determined, the particle swarm algorithm is used to optimize the placement of each wind turbine with the maximum output power of the wind farm as the optimization goal; the optimized placement of all wind turbines constitutes the optimal wind farm location layout.

[0059] Wind farm location layout, namely WindFarm Layout, WFLO.

[0060] The particle swarm optimization (PSO) algorithm is used to optimize the layout of randomly placed wind turbines in a traditional wind farm. The square wind farm, which is tilted diagonally towards the incoming wind, has a wider surface area when facing the incoming wind, and the distance between adjacent wind turbines in a unit is maximized, ensuring that the optimized wind farm location layout effectively weakens the negative impact of the wake effect on the wind farm frequency control.

[0061] The program first inputs the basic parameters of the entire wind farm, including wind farm specifications, square wind farm size, wind speed, wind direction, number of wind turbines, wind turbine locations, particle swarm characteristics (functions and parameters), and Jensen wake model data. Initially, a certain number of wind turbines are randomly placed.

[0062] Among them, the wind turbine group includes a generator and a wind turbine (blower) connected thereto.

[0063] The Jensen wake model data refers to the incoming wind speed of the upstream wind turbine.

[0064] Calculate the initial objective function value, run the PSO algorithm in MATLAB to get the next objective function value, compare it with the previous objective function value, and then update it to the new objective function value. Iter is the number of iterations. max is the maximum number of iterations. The proposed wind farm location layout optimization process is as follows: Figure 2 shown.

[0065] Particle Swarm Optimization (PSO) is an iterative optimization algorithm similar to genetic algorithms. PSO is initialized with a swarm of random particles (random solutions). It then iterates to find the optimal solution. In each iteration, the particles update themselves by tracking two "extremes." The first is the optimal solution found by the particle itself, which is called the individual extreme value P. best The other extreme value is the optimal solution found by the entire population so far, which is the global extreme value G best .

[0066] When these two optimal values ​​are found, the particle updates its velocity and new position according to the following formula:

[0067]

[0068] i is the i-th particle in the particle swarm; d is the dimension of the particle; rand: a random number between (0,1); c1 and c2 are the self-learning and social learning factors respectively; w is the inertia weight.

[0069] like Figure 7 As shown, according to the Jensen wake model, we can get:

[0070]

[0071] Where R is the radius of the wind wheel; U j is the incoming wind speed of the upstream wind turbine; U i is the wind speed in the downstream wake area; R w is the wake radius, where R+ks=R w , s is the ratio of the downstream distance to the rotor diameter D; C T is the thrust coefficient of the fan; k is the attenuation coefficient;

[0072] The power expression of wind turbine is:

[0073]

[0074] ρ-air density; R-wind rotor radius; v-wind speed; C p - Power coefficient, a function of the tip speed ratio λ and the pitch angle β, is state of the art.

[0075] Step 3: Simulate the output power of all wind turbines in the wind farm to verify the optimized wind farm location layout.

[0076] The parameters and specifications of wind farm and wind turbine are shown in Table 1 and Table 2. The thrust coefficient C T =0.88; the total size of the selected wind farm is 3 km × 3 km. The entire wind farm is divided into 100 possible wind turbine location units.

[0077] Table 1 Wind farm parameters

[0078]

[0079]

[0080] Table 2 Wind turbine parameters

[0081]

[0082] PSO parameters are used for fast response. The particle constraint parameters are inertia weight coefficient w = 0, acceleration coefficient constants c1 = 2.5, c2 = 2.5, and the maximum number of iterations Itermax = 100. When selecting a random number between 0 and 1, the Weibull distribution of formula (1) is used.

[0083] 1) Constant wind speed and direction

[0084] Assuming the wind farm is currently experiencing a constant wind speed of 10 m / s and a constant north wind direction, applying the aforementioned wind farm location layout optimization method yields the optimized WFLO as shown in Figure 3(a), while Figure 3(b) shows the result using the existing optimization method.

[0085] This optimization method ensures that the influence of wake effect on wind turbines downstream of the wind farm is reduced, so as to maximize the active power output of each wind turbine participating in frequency control. By simulating the output power of 30 wind turbines in the wind farm, the existing optimization method is compared with the WFLO optimization method proposed in this paper. Figure 4 shown.

[0086] Among them, the placement locations of 30 wind turbines were obtained by PSO.

[0087] from Figure 4 It can be seen that under the existing optimization method, it is difficult for each wind turbine to output optimal power, and the power generated by different wind turbines in the wind farm fluctuates greatly. The minimum output power occurs at wind turbine numbered 86 (the 27th wind turbine), with an output power of 1.47MW. The output power of wind turbines at positions 16, 25, 82, 47, and 66 (corresponding to wind turbines 20 to 25, respectively) decreases significantly, which is not conducive to frequency control after a high proportion of wind power is connected to the grid. Under the WFLO optimization method proposed in this paper, the output power decreases only at wind turbines 33, 55, 77, and 85 (corresponding to wind turbines 4, 10, 14, and 26, respectively). Otherwise, most wind turbines are able to operate at higher output power, with the lowest output power reaching 2.51MW, significantly reducing the power fluctuations caused by the wake effect on the wind farm.

[0088] 2) Random wind speed and direction

[0089] Considering the random wind speed and variable wind direction, all wind turbines are located within the wind farm and have the ability to rotate according to the variable wind direction, making the wind farm closer to the actual natural scene.

[0090] By using the aforementioned optimization method, all situations under a wind direction can be quickly solved. At any wind direction angle, the optimal layout pattern can be obtained through simulation, as shown in Figure 5(a). Figure 5(b) shows the result of using the existing optimization method.

[0091] As shown in Figure 5(a), there are four possible random wind directions. The layout pattern in Figure 5(a) is for one of the random wind directions.

[0092] In this embodiment, the optimal layout pattern is solved for any random wind direction in FIG5(a).

[0093] However, due to possible complex weather conditions, existing optimization methods place most wind turbines along the boundaries of wind farms, which cannot effectively control the power fluctuations of wind farms and are easily affected by the wake effect.

[0094] As can be seen from Figure 5(a), in the face of random wind speeds and variable wind directions, the optimized doubly fed wind turbines (wind turbines) are placed very evenly. Compared with Figure 5(b), the overcrowding between adjacent wind turbines is avoided, which helps to reduce the wake effect and suppress wind farm power fluctuations.

[0095] from Figure 6 It can be seen that the power generated by each wind turbine is relatively unstable when using the existing WFLO optimization method. The minimum output power occurs at turbine 66 (the 19th wind turbine), with a power of 0.86 MW. The output power of wind turbines 69, 82, and 85 (corresponding to the 18th, 22nd, and 26th wind turbines, respectively) suddenly decreases, causing significant fluctuations in the overall power of the wind farm. However, using the optimization method proposed in this paper, the output power of each wind turbine is almost unchanged, and all are able to maintain a high output power, with the minimum output power reaching 2.58 MW, which is beneficial to the system's frequency control.

[0096] The above description is merely a preferred embodiment of the present invention and does not limit the present invention in any way. Any person skilled in the art who, without departing from the scope of the present invention, makes any equivalent substitution, modification, or other changes to the technical solution and technical content disclosed in the present invention shall be deemed to be within the scope of the present invention and still fall within the scope of protection of the present invention.

Claims

1. A wind farm location layout optimization method considering wake effect, characterized in that: The following steps are involved: Step 1: Establish a wind farm model: The wind farm is designed as a square with one diagonal parallel to the X-axis; The wind farm is divided into M cells for placing wind turbines, where each cell is a square and one diagonal of the cell is parallel to the X-axis; Randomly place N wind turbines in the corresponding N cells; N<M; Step 2: Obtain the optimal wind farm location layout: When the number of wind turbines, wind direction and wind speed are determined, the particle swarm algorithm is used to optimize the placement of each wind turbine with the maximum output power of the wind farm as the optimization goal; the optimized placement of all wind turbines constitutes the optimal wind farm location layout.

2. The wind farm location layout optimization method considering wake effect according to claim 1, characterized in that: In step 2, the wind direction includes constant wind direction and random wind direction.

3. The wind farm location layout optimization method considering wake effect according to claim 1, characterized in that: In step 2, the output power expression of the wind turbine is: ρ-air density; R-rotor radius; v-wind speed; Cp-power coefficient, which is a function of the tip speed ratio λ and the pitch angle β.

4. The method for optimizing wind farm location layout considering wake effect according to claim 1, characterized in that: In step 2, within the tolerance of wake effect, the number of wind turbines is related to the proportion of wind power grid-connected.

5. The method for optimizing wind farm location layout considering wake effect according to claim 1, characterized in that: In step 1, the wind speed follows the Weibull distribution.

6. The method for optimizing wind farm location layout considering wake effect according to claim 1, characterized in that: The step 2 further includes the following steps: simulating the output power of all wind turbines in the wind farm to verify the optimized wind farm location layout.

7. The method for optimizing wind farm location layout considering wake effect according to claim 1, characterized in that: The proportion of wind farms connected to the grid is further considered in obtaining the optimal wind farm location layout.

8. The method for optimizing wind farm location layout considering wake effect according to claim 1, characterized in that: The constraints for obtaining the optimal wind farm location layout include: no wind turbine oscillation, subsynchronous oscillation of the HVDC Flexible system, and short-circuit faults.

9. The method for optimizing wind farm location layout considering wake effect according to claim 1, characterized in that: The optimal wind farm location layout must be obtained within the wake effect tolerance limit.

10. The method for optimizing wind farm location layout considering wake effect according to claim 1, characterized in that: The proportion of wind farms connected to the grid and the wake effect satisfy the following relationship: |λI-H(s,v0)|=0 Among them, H(s,v0) varies with the wind power grid connection ratio s and the wake effect; Set the characteristic value λ = σ + jω, i = 1, 2, ..., n, the real part reflects the damping strength, and the imaginary part reflects the degree of frequency drop; v0 is the wind speed.

Images