Hello
I wanted to ask if it possible to output the axial induction factor for the turbine in my output files to see what is happening
I am not sure where exactly could i pick the output of the turbines. I mean I read the manual and I knew how things work but what I do not know how to pick which output to be printed to my text file or is it forced to print all of the output related to turbines.
The main question is how to print the axial induction for my turbine.
The axial induction factor should be the Uinf-Urotor/Uinf for the actual disk model but I am not sure if it is like that in the model I am using or not. I am using a 5MW NREL turbine with fast 8.
Are you using AeroDyn v15 within FAST v8? Within the AeroDyn v15 module, the axial induction factor is available as output BαNβAxInd, for node β of blade α. See Appendix E of the draft AeroDyn v15 User’s Guide and Theory Manual for more information: nwtc.nrel.gov/system/files/Aero … 15.04a.pdf.
Dear Dr Jason,
I am using open fast and I am reading the manual you gave me. it seems that there is a way to output the axial induction factor as I can see in the appendix. I am still wondering how to print it out in the output folder. I mean I am using the AeroDyn v15 and i am getting many variables in the <>.out file but it does not have the aif. is there something in the input fast of fast I need to change so that I can print the aif ?
thank you so much for your help
yours
Mostafa
Hey Dr Jason
if I understood correctly, Fast is connecting different modules together.I have more than one input file. so to change in the output for the AeroDyn v15 module, I have to change in the NRELOffshrBsline5MW_Onshore_AeroDyn15.dat output parameters. I am testing it right now but I thought to ask while investigating more in the topic.
thanks
Dear Dr Jason
there is an axial induction factor for each blade for each node and I have 17 nodes. what i need is the axial factor for the rotor averaged value. Is there a way for fast to output this value only? IF not, how can i post process my axial induction factor then to get the rotor averaged axial induction factor?
I was thinking about outputting the axial induction factor for each node for each blade and then calculating the average but that seems to be really stupid as 17 nodes * 3 blades is so many outputs to be post-processed in matlab. I am saying that I will have so many reading of axial induction factor
i would like to ask if this method i am suggesting is proper to get the average axial induction factor.
should i add the aif for the 17 nodes then divided by 17 as if i am getting the average ai for each blade then I get the average aif for each blade and divide by 3 to get the average aif for the rotor.
thank you in advance
mostafa
IS there a way i could edit the adero dyn v15 files so that it could output for me right a way the averaged value . I mean is it possible that i use fast to do the post processing. where exactly should i head to if i want to do something like this ?
The rotor-disk averaged axial induction factor is not a standard output of AeroDyn v15. You could, of course, output the axial induction factors for every node, but as you said, this would be a lot of outputs. For this calculation, it may be preferred if you modify the source code directly to calculate and output what you want.
Please note that if you are averaging axial induction across the rotor, it may be preferred to weight the local axial induction factor by the local radius/circumference because the outer nodes sweep more distance than the inner nodes when the rotor is rotating.
is it possible to explain more how to weight the loca axial induction factor by the local radius ?the method of implementing it .i understand my axial induciton factor changes with the radius.
thank you so much
Here is a simple implementation if r(b,n) is the local radius to node n of blade b and a(b,n) is the local axial induction:
WeightedAvgInd = 0
rSum=0
FOR b = 1,NumBlades
FOR n = 1,NumNodes
WeightedAvgInd = WeightedAvgInd + r(b,n)*a(b,n)
rSum = rSum + r(b,n)
END
END
WeightedAvgInd = WeightedAvgInd/rSum
Best regards,
how can i know the node postion ?is there an Output paramater to know to node radial postion ? how do i know the Maximum number of nodes ?is the node equal to Number of blade input stations?I found that Parameter called station_total saying that i have 49 Input Station for the blade .is that the number of nodes ?
i found a Parameter in elasto dyn Folder called BldNodes (with a Deskription of ''Number of blade nodes ‘’)is that it ?
my Problem is that when i opened the AeroDyn v15 source Output files i found out the following
INTEGER(IntKi), PARAMETER :: B1N1AxInd = 520
INTEGER(IntKi), PARAMETER :: B1N2AxInd = 521
INTEGER(IntKi), PARAMETER :: B1N3AxInd = 522
INTEGER(IntKi), PARAMETER :: B1N4AxInd = 523
INTEGER(IntKi), PARAMETER :: B1N5AxInd = 524
INTEGER(IntKi), PARAMETER :: B1N6AxInd = 525
INTEGER(IntKi), PARAMETER :: B1N7AxInd = 526
INTEGER(IntKi), PARAMETER :: B1N8AxInd = 527
INTEGER(IntKi), PARAMETER :: B1N9AxInd = 528
there is only 9 Output Parameters per blade which mean there is only 9 nodes !!! however that is totally a different number compared to the parameters mentioned above .so how many nodes do i have ?is it possible that i do not get outputed the axial induction factor for all of the nodes ?also how do i know the postion of the node the radial postion to be exact ?
I have another question
in the aerodyn 15 file
i outputed “B1N1AxInd” and set NBlOuts to 1 and BlOutNd to 1 and i get in my Output file the variable but all of my axial induction factor is Zero
i do not know why
secondly
according to the 5mw turbine , the pitch angle can be changed with a Maximum Absolute Blade Pitch Rate of 8 degree per second. What i am doing is using open Loop function in simulink and sending the commands with pitch angles. The idea is i am doing a sinsoidal Motion with my blade .TO explain it more , i want my blade to follow this equation, pitch angle of blade 1 =Ampsin(Omegasimulationtime)
my amp is the Amplitude (1 degree or 2 degree ) and the Omega is 2*pi/period the period is the period of my sin function.my question is what frequency or period should i be using that i can get realistic results and to be able to implement it with the turbine?
NREL does not have the resources to change the source code for you. You’ll have to study the source code yourself if you wish to make the changes you want.
To answer some of your questions:
The number of aerodynamic blade analysis nodes is specified via input NumBlNds in the AeroDyn blade input file.
AeroDyn v15 is limited to outputting only 9 of the NumBlNds. Which nodes can be output is specified via AeroDyn inputs NBlOuts and BlOutNd in the AeroDyn primary input file.
When AeroDyn is coupled to FAST with blade flexibility enabled, the radial position of each aerodynamic analysis node will change every time step. The positions of the nodes are available within the source code, but are not standard outputs from AeroDyn.
The axial induction may in fact be zero at the blade root or tip due to the hub- and tip-loss induction factors, when enabled.
I’m not sure what you mean by “realistic results” because it sounds like the sinusoidal blade-pitch motion you are prescribing is quite idealistic. The maximum pitch rate of the NREL 5-MW turbine is set at 8 deg/s, but normal pitch rates would be much less than this.
thank you so much for your help Doctor
secondly i want ask few things.if i output the 9 of the numblnd to calculate the average rotor axial induction, does that mean my results does not count as an accurate as i am not using the whole nodes of the blade ? secondly i do not understand what you meant by blade flexibility is that an option(i am checking the manuals right now but i thought to ask you )? also if the nodes are changing their position at each time step ,does this mean the output coming out from aerodyn for the axial induction factor is for different positions ?like at each interval the axial induction factor is for a node at a different position ? am i understanding it correctly?
another general question
for a student who is not the best with fast and has no experience with openfoam and he wants to try using SOWFA do you think it is possible ? i have experience with another LES but not SOWFA.is such thing even possible?I am trying to learn fast really quickly
I would think you could get an accurate representation of the induction if you distribute the 9 output nodes along the blade. Of course, outputting data at 9 nodes will be less accurate than outputting data at all blade nodes, but the induction should be fairly smoothly distributed along the blade.
By “flexibility”, I mean structural deflection when running AeroDyn coupled to ElastoDyn or BeamDyn within FAST. Both ElastoDyn and BeamDyn can model blade deflection and these have an aero-elastic interaction with AeroDyn. “Yes,” if the blade structural degrees of freedom are enabled, the local radius to each aerodynamic output node will change in time as a result of the blade deflection.
SOWFA is much more difficult to learn than FAST. You mention experience with LES; any experience with OpenFOAM would be useful to learn to SOWFA.
i have a very small question.
first, I am speechless from your help and I greatly appreciate it.I owe you a favor and i hope to meet you to thank you in person.
secondly, My file has 17 nodes I can output only 9 nodes right? i get to choose any of these 9 nodes right from the 17 node ?
also which nodes should i pick up ? i mean should i choose the first 9 nodes after each other or should I pick one and leave one I mean what combination of nodes should I pick ? or do you think it does not matter? I was thinking that picking a combination of nodes to cover the whole blade to be outputted is the best idea. i mean for example instead of outputting nodes :1 2 3 4 5 6 7 8 9 to be 1 3 5 7 9 11 13 15 17 .I meant to make sure that I have nodes covering the whole blade or what do you think?i am not sure I study mechatronics engineering and i do not have much experience with aerodynamics but i am using my common sense hoping i can get some output .
Also if i turn on the blade structural degrees of freedom, how is the radius changing? I mean are the nodes moving linearly? I mean for example node 0 is at position 0 at time step 0 and node 1 is at position 1.In the next time step, does node 0 get positioned at position 1 and node 1 get positioned at position 2 or how does it work ? do you have any idea how it works?I mean this information will be very very very helpful and i can use it to just post-processing the data instead of editing in the source code.The problem im not an expert in fortran and tried to understand the aerodyn codes but it does not seem to be a simple task.
thank you so so much
mostafa
Correct, outputting nodes 1 3 5 7 9 11 13 15 17 would be a good distribution of the output nodes.
The deflection is not linear, but if you disable the structural degrees of freedom in ElastoDyn, then the blades will be rigid and the aerodynamic analysis node radial positions you set in the AeroDyn input file will be fixed over time.
a very small question ! i have opened ElastoDyn file and i am looking for the manual of it.My problem is in the structural degrees of freedom i do not find a certain paramter called the structural degrees of freedom .i can only find these parameters to switch on and off but there is not parameter called structural degree of freedom
could you please tell me more about which parameter you ?
i am sure that you are talking about one of these parameters but which one exactly:
True FlapDOF1 - First flapwise blade mode DOF (flag)
True FlapDOF2 - Second flapwise blade mode DOF (flag)
True EdgeDOF - First edgewise blade mode DOF (flag)
Yes, I mean the disable all of the blade structural DOFs i.e. FlapDOF1 = FlapDOF2 = EdgeDOF = False.
Of course, if you wish to model a rigid turbine, you may as well use the standalone driver for AeroDyn v15 rather than coupled to FAST. Running the standalone driver is much simpler than running AeroDyn coupled to FAST. Documentation on the standalone driver for AeroDyn v15 is provided along with the AeroDyn documentation.
Is closing these structural degrees of freedom affect in my loads and power calculations? I mean i will get less realistic results?
Also as you can see in the picture this is a representation of blade with nodes. When we said that the nodes are moving, are they moving in which direction? In the aerodyn file you enter the initial postion of the radial distance to each node starting from 0 to 63.as my rotor radius is 63.when you said that the nodes move in the radial direction you mean the direction in the picture or the direction along the rotor radius (meaning that the intial radial distances of the nodes are changing? )
If my second assumption is correct (concerning that my intial radial distance are changing like the node at postion 62 m will move a bit down to62.1), is it correct to say that the nodes are moving up and down is that right?due to contraction and extraction of the blade? Is it true
If its true, according to what parameter does it extend or contract? What parameters says when the node to move up or down?what parameter say that the blade is extending or not?
Also is their a maximum to the amount of extention of the blade?
Is it possible to assume that my nodes are moving all linearly with a constant value. I mean lets say that all my nodes move down by 0.01 and at another instant all nodes goes up by 0.01?
There is a reason why i donot want to close the structural dof. I have checked on the loads with them on and off and there seemed to be a difference between structural dof on and off.
What i hope to do is to assum that the nodes are moving linearly in the radial direction with a consant value it contract or extend according to a certain parameter i would love to know. With this information i can post process my data assuming its moving linearly.
