about .yam file

[Qutianyuan0124]

No matter how I configure the turbine radius and rotor diameter in the YAML file of the turbine library, it always defaults to 126m during application. How can I customize the turbine?

[misi9170]

Hi @Qutianyuan0124 ,

Can you share an example of the code that you are running so that we can help you debug it? Ideally, this would be a minimal example including your turbine YAML file (or something that exhibits the behavior you're seeing, with a rotor diameter different to 126m), the main floris input yaml, and a short script demonstrating how you are running FLORIS and the issue you are running into?

[Qutianyuan0124]

***@***.*** Thank you very much for your reply. Below, I will provide a detailed introduction to the difficulties I have encountered and provide a script file for you to better understand my current situation. Firstly, I noticed that the wind turbine type in the gch.yaml file is nrel_5MW. In order to facilitate my practical application, I directly modified the hub height to 160m and the impeller diameter to 200m in the turbininlibrary file. Next, I optimized the layout and visualized it, and found that the impeller diameter did not change and remained at 125.88m. The attached test. py is an example, where min-dist-D=5. During my execution, I found that 5D=629.4m. I'm wondering if my operating method is incorrect, why can't I modify the configuration of the fan. Besides, I am concerned that I may not be able to modify parameters such as power curve and thrust coefficient, which would prevent me from applying them. As described in the picture, I am unable to apply the modified impeller diameter.In addition, in the visualization of the entire hub height section of the wind farm, the impeller diameter cannot be changed, and I feel that this seems to be a problem. Thank you again for your reply. Looking forward to receiving good news. ***@***.*** From: misi9170 Date: 2024-09-09 21:03 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi @Qutianyuan0124 , Can you share an example of the code that you are running so that we can help you debug it? Ideally, this would be a minimal example including your turbine YAML file (or something that exhibits the behavior you're seeing, with a rotor diameter different to 126m), the main floris input yaml, and a short script demonstrating how you are running FLORIS and the issue you are running into? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[misi9170]

Hi @Qutianyuan0124 ,

Unfortunately the files you attached did not come through. Rather than attaching them to an email, could you upload them directly to the discussion on github.com//discussions/979?

In the meantime, I did run a quick check: I modified the rotor_diameter field of floris/turbine_library/NREL_5MW.yaml to 160.0, and then ran the following:

from floris import FlorisModel

fmodel = FlorisModel("inputs/gch.yaml") # Relative path to gch.yaml from examples/ directory

fmodel.run()

print(fmodel.core.farm.rotor_diameters[0]) # Prints [160. 160. 160.]

That seems to be working as expected, but perhaps there is an issue in the optimization code somehow.

Thanks,
Misha

[Qutianyuan0124]

***@***.***， I did the same operation as you, and the result still shows 125.88. Below is the parameters I set.I modified this directly in the yaml file.Can you tell me how you set it up, or maybe I'm not doing it the right way.In addition,Did I have some bugs when installing floris?Looking forward to your reply. Thanks, Qutianyuan ***@***.*** From: misi9170 Date: 2024-09-10 22:08 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi @Qutianyuan0124 , Unfortunately the files you attached did not come through. Rather than attaching them to an email, could you upload them directly to the discussion on github.com//discussions/979? In the meantime, I did run a quick check: I modified the rotor_diameter field of floris/turbine_library/NREL_5MW.yaml to 160.0, and then ran the following: from floris import FlorisModel fmodel = FlorisModel("inputs/gch.yaml") # Relative path to gch.yaml from examples/ directory fmodel.run() print(fmodel.core.farm.rotor_diameters[0]) # Prints [160. 160. 160.] That seems to be working as expected, but perhaps there is an issue in the optimization code somehow. Thanks, Misha — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Qutianyuan0124]

***@***.***, Could you please tell me in which path the yaml file called by the model is stored? I just deleted the Nrel_5Mwe. yaml file in turbine library and the program can still run, so I guess I changed the diameter of the fan impeller in the wrong place. In addition, how to customize a new fan and let the model be applied? Thanks, Qutianyuan ***@***.*** From: misi9170 Date: 2024-09-10 22:08 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi @Qutianyuan0124 , Unfortunately the files you attached did not come through. Rather than attaching them to an email, could you upload them directly to the discussion on github.com//discussions/979? In the meantime, I did run a quick check: I modified the rotor_diameter field of floris/turbine_library/NREL_5MW.yaml to 160.0, and then ran the following: from floris import FlorisModel fmodel = FlorisModel("inputs/gch.yaml") # Relative path to gch.yaml from examples/ directory fmodel.run() print(fmodel.core.farm.rotor_diameters[0]) # Prints [160. 160. 160.] That seems to be working as expected, but perhaps there is an issue in the optimization code somehow. Thanks, Misha — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[misi9170]

Hi Qutianyuan,

If you cloned the FLORIS repository and then installed using pip install -e floris, then the NREL_5MW.yaml model located in floris/floris/turbine_library/ should be the one that will be used. However, if you've found that you can still run the code above after deleting NREL_5MW.yaml, then perhaps FLORIS is installed in a different location. For example, if you ran pip install floris (without the -e flag), then FLORIS will be installed from PyPI to wherever your conda environment is stored, and the NREL_5MW.yaml that is installed with it will be used.

You can check this by running pip show floris from the terminal---does this print the path to your clone, or to somewhere else on your computer?

Cheers,
Misha

[Qutianyuan0124]

Hi Misha， I have solved this problem according to your method, and I plan to reinstall it for better use. Thank you very much for your valuable tips. Thanks， Qutianyuan ***@***.*** From: misi9170 Date: 2024-09-11 10:20 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi Qutianyuan, If you cloned the FLORIS repository and then installed using pip install -e floris, then the NREL_5MW.yaml model located in floris/floris/turbine_library/ is the one that will be used. However, if you've found that you can still run the code above after deleting NREL_5MW.yaml, then perhaps FLORIS is installed in a different location. For example, if you simply run pip install floris (without the -e flag), then FLORIS will be installed from PyPI to wherever your conda environment is stored, and the NREL_5MW.yaml that is installed with it will be used. You can check this by running pip show floris from the terminal---does this print the path to your clone, or to somewhere else on your computer? Cheers, Misha — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Qutianyuan0124]

Hello Misha， Recently, I have encountered some issues regarding Floris visualization that I would like to ask you about. For the "Visualize y cut plane," is this plane a horizontal plane passing through the hub center of the turbine? It seems more like a plane perpendicular to the Y-axis and parallel to the X-Z plane. For the "Visualize cross plane," I feel that this plane passes through the hub center of the first row of turbines and reflects the wake situation. However, I want to check the second and third rows of turbines, but I don’t know how to proceed. Attached are my turbine coordinate files and boundary points. How should I set it up to see the wake distribution for turbines T6 to T10? Your sincerely, Qutianyuan ***@***.*** From: misi9170 Date: 2024-09-11 10:20 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi Qutianyuan, If you cloned the FLORIS repository and then installed using pip install -e floris, then the NREL_5MW.yaml model located in floris/floris/turbine_library/ is the one that will be used. However, if you've found that you can still run the code above after deleting NREL_5MW.yaml, then perhaps FLORIS is installed in a different location. For example, if you simply run pip install floris (without the -e flag), then FLORIS will be installed from PyPI to wherever your conda environment is stored, and the NREL_5MW.yaml that is installed with it will be used. You can check this by running pip show floris from the terminal---does this print the path to your clone, or to somewhere else on your computer? Cheers, Misha ― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[paulf81]

Hi! You're right, a y-cross plane is perpendicular to the y-axis and so you only need to set it's position along the y axis, while the x-cross plane is perpendicular to x and you set the distance downstream (along x-axis), there are examples illustrating both here:

https://nrel.github.io/floris/examples/examples_visualizations/002_visualize_y_cut_plane.html

https://nrel.github.io/floris/examples/examples_visualizations/003_visualize_cross_plane.html

[Qutianyuan0124]

But what does downstream_dist=500.0 mean, how do I look at the second and third rows, and what should I set downstream_dist to ***@***.*** From: paulf81 Date: 2024-09-13 23:03 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi! You're right, a y-cross plane is perpendicular to the y-axis and so you only need to set it's position along the y axis, while the x-cross plane is perpendicular to x and you set the distance downstream (along x-axis), there are examples illustrating both here: https://nrel.github.io/floris/examples/examples_visualizations/002_visualize_y_cut_plane.html https://nrel.github.io/floris/examples/examples_visualizations/003_visualize_cross_plane.html — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Qutianyuan0124]

For example, I want to check the section of the four columns of 1, 2, 3 and 4 fans, how should I set it ***@***.*** From: paulf81 Date: 2024-09-13 23:03 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi! You're right, a y-cross plane is perpendicular to the y-axis and so you only need to set it's position along the y axis, while the x-cross plane is perpendicular to x and you set the distance downstream (along x-axis), there are examples illustrating both here: https://nrel.github.io/floris/examples/examples_visualizations/002_visualize_y_cut_plane.html https://nrel.github.io/floris/examples/examples_visualizations/003_visualize_cross_plane.html — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[misi9170]

Hi @Qutianyuan0124,

I don't quite understand your layout, but if you want to visualize the cross planes at the turbine locations, you can use calculate_cross_plane with downstream_dist set to the location of the turbines. You can use this code to get started:

import matplotlib.pyplot as plt

from floris import FlorisModel
from floris.flow_visualization import visualize_cut_plane

fmodel = FlorisModel("../inputs/gch.yaml")

x_locations = [0, 600, 1200, 1800, 0, 600, 1200, 1800]
y_locations = [0, 0, 0, 0, 500, 500, 500, 500]

fmodel.set(
    layout_x=x_locations,
    layout_y=y_locations,
    wind_directions=[270],
    wind_speeds=[8],
    turbulence_intensities=[0.06],
)

# Collect the cross plane at the location of each downstream turbine
fig, ax = plt.subplots(4, 1, figsize=(3, 10))
for i, x in enumerate(x_locations[:4]):
    cross_plane = fmodel.calculate_cross_plane(
        y_resolution=100,
        z_resolution=100,
        downstream_dist=x,
    )

    # Plot the flow field
    visualize_cut_plane(
        cross_plane, ax=ax[i], min_speed=3, max_speed=9, label_contours=True,
        title=f"Cross plane at x={x} m",
    )

plt.show()

Note that here, the downstream_dist is simply the x-location of the turbines because the wind direction is 270 degrees and the wind farm is oriented with the x and y axes. For other wind directions or layouts, you'll need to do a little bit of geometry to get the correct downstream_dist.

[Qutianyuan0124]

Hello, Today is our country's Mid-Autumn Festival, also wish you a happy Mid-Autumn Festival, family reunion. Also I would like to ask what downstream_dist should be set to in your example if the wind direction is 0 degrees ***@***.*** From: misi9170 Date: 2024-09-16 20:56 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi @Qutianyuan0124, I don't quite understand your layout, but if you want to visualize the cross planes at the turbine locations, you can use calculate_cross_plane with downstream_dist set to the location of the turbines. You can use this code to get started: import matplotlib.pyplot as plt from floris import FlorisModel from floris.flow_visualization import visualize_cut_plane fmodel = FlorisModel("../inputs/gch.yaml") x_locations = [0, 600, 1200, 1800, 0, 600, 1200, 1800] y_locations = [0, 0, 0, 0, 500, 500, 500, 500] fmodel.set( layout_x=x_locations, layout_y=y_locations, wind_directions=[270], wind_speeds=[8], turbulence_intensities=[0.06], ) # Collect the cross plane at the location of each downstream turbine fig, ax = plt.subplots(4, 1, figsize=(3, 10)) for i, x in enumerate(x_locations[:4]): cross_plane = fmodel.calculate_cross_plane( y_resolution=100, z_resolution=100, downstream_dist=x, ) # Plot the flow field visualize_cut_plane( cross_plane, ax=ax[i], min_speed=3, max_speed=9, label_contours=True, title=f"Cross plane at x={x} m", ) plt.show() Note that here, the downstream_dist is simply the x-location of the turbines because the wind direction is 270 degrees and the wind farm is oriented with the x and y axes. For other wind directions or layouts, you'll need to do a little bit of geometry to get the correct downstream_dist. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Qutianyuan0124]

I don't quite understand this geometric operation you said, can you demonstrate it? Your sincerely, qutianyuan ***@***.*** From: misi9170 Date: 2024-09-16 20:56 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] about .yam file (Discussion #979) Hi @Qutianyuan0124, I don't quite understand your layout, but if you want to visualize the cross planes at the turbine locations, you can use calculate_cross_plane with downstream_dist set to the location of the turbines. You can use this code to get started: import matplotlib.pyplot as plt from floris import FlorisModel from floris.flow_visualization import visualize_cut_plane fmodel = FlorisModel("../inputs/gch.yaml") x_locations = [0, 600, 1200, 1800, 0, 600, 1200, 1800] y_locations = [0, 0, 0, 0, 500, 500, 500, 500] fmodel.set( layout_x=x_locations, layout_y=y_locations, wind_directions=[270], wind_speeds=[8], turbulence_intensities=[0.06], ) # Collect the cross plane at the location of each downstream turbine fig, ax = plt.subplots(4, 1, figsize=(3, 10)) for i, x in enumerate(x_locations[:4]): cross_plane = fmodel.calculate_cross_plane( y_resolution=100, z_resolution=100, downstream_dist=x, ) # Plot the flow field visualize_cut_plane( cross_plane, ax=ax[i], min_speed=3, max_speed=9, label_contours=True, title=f"Cross plane at x={x} m", ) plt.show() Note that here, the downstream_dist is simply the x-location of the turbines because the wind direction is 270 degrees and the wind farm is oriented with the x and y axes. For other wind directions or layouts, you'll need to do a little bit of geometry to get the correct downstream_dist. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[misi9170]

Hi @Qutianyuan0124 ,

Happy Mid-Autumn Festival!

After playing around with trying to compute downstream distances using the wind direction, I actually found it was easier to just reach down into the data stored on the fmodel to set the downstream distances for arbitrary wind directions. In the example below, you can set wind_direction to whatever you would like (0, 270, 45, etc) and see the cross plane for each of the 8 turbines in this layout.

import matplotlib.pyplot as plt
import numpy as np

from floris import FlorisModel
from floris.flow_visualization import visualize_cut_plane


fmodel = FlorisModel("../inputs/gch.yaml")

x_locations = [0, 600, 1200, 1800, 0, 600, 1200, 1800]
y_locations = [0, 0, 0, 0, 500, 500, 500, 500]

n_turbines = len(x_locations)

wind_direction = 45

fmodel.set(
    layout_x=x_locations,
    layout_y=y_locations,
    wind_directions=[wind_direction],
    wind_speeds=[8],
    turbulence_intensities=[0.06],
)
fmodel.run()
print(fmodel.get_turbine_powers())

# Collect the cross plane at the location of each downstream turbine
fig, ax = plt.subplots(n_turbines, 1, figsize=(3, n_turbines))
for i in range(n_turbines):
    dd = fmodel.core.grid.x_sorted.mean(axis=(2,3))[0,i]
    cross_plane = fmodel.calculate_cross_plane(
        y_resolution=100,
        z_resolution=100,
        downstream_dist=dd,
    )

    ii = fmodel.core.grid.sorted_coord_indices[0,i]

    # Plot the flow field
    visualize_cut_plane(
        cross_plane, ax=ax[ii], min_speed=3, max_speed=9, label_contours=True,
        title="Cross plane T{1:03d}".format(dd, ii),
    )

fig.suptitle("Wind direction: {0:.1f} deg.".format(wind_direction))

plt.show()