.yaml file

[Qutianyuan0124]

Discussed in #547

^{Originally posted by prith-gs December 8, 2022}
Can I add a new (custom) wake deficit model on my own and use it to calculate AEP for a wind farm?

[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]

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?

Hi @Qutianyuan0124 , thanks for the question! See my response over on #979

[misi9170]

Closing this issue, as I believe it is resolved in #979

[Qutianyuan0124]

hello， Recently, I have been researching how to optimize wind turbine layouts. When I use the Example layout method for optimization, I find that I cannot find the optimal layout. However, when I use the Generic random search for optimization, I notice that the turbine layout is very irregular and aesthetically unpleasing. Do you have any methods to ensure the turbines are arranged neatly and aligned with the dominant wind direction? I have tried using the SLSQP method, but I am unable to integrate it into the Generic random search code. Do you have any good suggestions for this? sincerely， qutianyuan ***@***.*** From: misi9170 Date: 2024-09-25 03:42 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] .yaml file (Issue #978) Closing this issue, as I believe it is resolved in #979 ― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Qutianyuan0124]

Assuming that my dominant wind direction is from the west, how can I arrange the turbines neatly while ensuring a certain improvement in energy production? The image below shows my optimized layout, which appears very scattered. ***@***.*** 发件人： 瞿天元 发送时间： 2024-09-27 16:35 收件人： NREL/floris 主题： Re: Re: [NREL/floris] .yaml file (Issue #978) hello， Recently, I have been researching how to optimize wind turbine layouts. When I use the Example layout method for optimization, I find that I cannot find the optimal layout. However, when I use the Generic random search for optimization, I notice that the turbine layout is very irregular and aesthetically unpleasing. Do you have any methods to ensure the turbines are arranged neatly and aligned with the dominant wind direction? I have tried using the SLSQP method, but I am unable to integrate it into the Generic random search code. Do you have any good suggestions for this? sincerely， qutianyuan ***@***.*** From: misi9170 Date: 2024-09-25 03:42 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] .yaml file (Issue #978) Closing this issue, as I believe it is resolved in #979 ― Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[misi9170]

Hi @Qutianyuan0124 ,

No, we don't currently have any in-built methods that generate "neat" layouts that also maximize AEP. We recently added a layout optimizer LayoutOptimizationGridded that does indeed generate gridded layouts (see #976, or this example)---however, the intention with that is to pack as many turbines as possible into a given boundary area, which is a proxy for maximizing AEP, rather than taking a fixed number of turbines and arranging them in a neat way to maximize AEP.

We'd be interested in what you find if you decide to build your own layout optimizer that maximizes AEP while keeping turbines in a neat arrangement!

Misha

[Qutianyuan0124]

I'm trying this example, when an error prompt is as follows: the from floris. Optimization. Layout_optimization. Layout_optimization_gridded import ( ModuleNotFoundError: No module named 'floris.optimization.layout_optimization.layout_optimization_gridded' Do I need to update floris or modify the configuration file. Also, I would be very grateful if you could tell me the path of the original code file about the optimized layout ***@***.*** From: misi9170 Date: 2024-09-28 05:01 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] .yaml file (Issue #978) Hi @Qutianyuan0124 , No, we don't currently have any in-built methods that generate "neat" layouts that also maximize AEP. We recently added a layout optimizer LayoutOptimizationGridded that does indeed generate gridded layouts (see #976, or this example)---however, the intention with that is to pack as many turbines as possible into a given boundary area, which is a proxy for maximizing AEP, rather than taking a fixed number of turbines and arranging them in a neat way to maximize AEP. We'd be interested in what you find if you decide to build your own layout optimizer that maximizes AEP while keeping turbines in a neat arrangement! Misha — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Qutianyuan0124]

Hi, I used genetic random search for optimization, but the yield was lower, and I couldn't figure out what was wrong. Attached is my time series data, boundary and code files. Could you please help me check them. ***@***.*** From: misi9170 Date: 2024-09-28 05:01 To: NREL/floris CC: Qutianyuan0124; Mention Subject: Re: [NREL/floris] .yaml file (Issue #978) Hi @Qutianyuan0124 , No, we don't currently have any in-built methods that generate "neat" layouts that also maximize AEP. We recently added a layout optimizer LayoutOptimizationGridded that does indeed generate gridded layouts (see #976, or this example)---however, the intention with that is to pack as many turbines as possible into a given boundary area, which is a proxy for maximizing AEP, rather than taking a fixed number of turbines and arranging them in a neat way to maximize AEP. We'd be interested in what you find if you decide to build your own layout optimizer that maximizes AEP while keeping turbines in a neat arrangement! Misha — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>