Need help on designing a conceptual HALE UAV

[jag0097wtf]

Hi everyone,

I am a High School student (final year) and am doing an extra-curricular project. Having complete autonomy on what we do the project on, I choose to design a High-Altitude Long-Endurance (HALE) UAV, such as the QinetiQ Zephyr [http://www.airforce-technology.com/projects/zephyr/] or these designs from the NASA competition - [http://commentspro.com/lead/172000] .

Having very basic knowledge of aeronautical engineering, I don't know where to begin the designing process. I have got Open VSP and know that I can simply mockup a few ideas there, and calculate basic data like lift coefficient, drag etc, but in reality, I don't think that won't tell me much about improving my designs. I have also been reading a NASA paper on a similar problem [http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20070017849.pdf], which turns out to be similar to what I want to design. However, they have used complex numerical optimisation methods to get an optimal design, and since I have no experience of MATLAB, or anyting of that sort, I can't do this.

If, in a reasonably short space of time, it is possible to understand the basics of numerical optimisation like MATLAB, and use it to come up with basic designs, I would like to do this, and would like to know how to get started. If this is beyond the scope of learning within, say 2-weeks, then can anyone suggest how to go about the project without the computational route.

I can do a bit of basic Python programming, and have used PTC Creo for product design, but nothing else.

Please tell me if you need to know anything else.

I hope someone can help, Thanks!

 

[Stephen Tashi]

I can't get the link to the NASA paper to work.

Unless you are required to do an optization problem, why not just compare the performance of several different designs?

 

[Borg]

I can't get the link to the NASA paper to work.

He accidently included the ending bracket on his URLs. Just remove the trailing ] and they'll work.

 

[Stephen Tashi]

I got the link to the NASA paper to work but it gave me a 1 page blank document.

 

[jag0097wtf]

I can't get the link to the NASA paper to work.

Unless you are required to do an optization problem, why not just compare the performance of several different designs?

Cheers for the reply.

I checked the NASA link, and it seems to work for me (should bring up a pdf), not sure why it isn't working for you.

No I don't have to do an optimization problem, but that is what I thought I would need to do when I pick a design and want to improve it.

But going back a few steps, how do I come up with several valid designs, that would be aerodynamically correct? I can easily draw up a few designs based on what I have seen on such aircraft, but how would I figure out if they would actually fly in real life? And this kind of leads on to the second thing - how do I compare the performance in the first place? As of now, assuming I have mocked up a few very basic and preliminary designs, using Open VSP, I can work out data such as the wetted area, which would help me calculate the drag for example. Would basic data such as drag, coefficient of lift etc. be sufficient? And if I wanted to go further, how would I evaluate things such as engine performance, or selecting the right airfoil?

I know that's a lot of questions, hope they are clear enough.

 

[Stephen Tashi]

But going back a few steps, how do I come up with several valid designs, that would be aerodynamically correct? I can easily draw up a few designs based on what I have seen on such aircraft, but how would I figure out if they would actually fly in real life? And this kind of leads on to the second thing - how do I compare the performance in the first place? As of now, assuming I have mocked up a few very basic and preliminary designs, using Open VSP, I can work out data such as the wetted area, which would help me calculate the drag for example. Would basic data such as drag, coefficient of lift etc. be sufficient? And if I wanted to go further, how would I evaluate things such as engine performance, or selecting the right airfoil?

I know about mathematical optimization - when the function to be optimized is given. You are asking questions that need to be answered by an aerornautical engineer and I'm not one. You'll have better luck getting answers to technical questions on the forum if you ask one specific question per thread. If this project must be completed in a few months, your goals are too ambitious. You won't be able learn all engineering within that time. You might find some powerful software that does many of the calculations for you. Looking for software packages is a different task than learning the physics and math behind what they are doing. I suggest you set your goals so that they involve only a few things beyond what you already know how to do.

 

[jag0097wtf]

I know about mathematical optimization - when the function to be optimized is given. You are asking questions that need to be answered by an aerornautical engineer and I'm not one. You'll have better luck getting answers to technical questions on the forum if you ask one specific question per thread. If this project must be completed in a few months, your goals are too ambitious. You won't be able learn all engineering within that time. You might find some powerful software that does many of the calculations for you. Looking for software packages is a different task than learning the physics and math behind what they are doing. I suggest you set your goals so that they involve only a few things beyond what you already know how to do.

Thanks for your help!

So at the moment, I will probably look at calculating basic things, such as drag of coefficient of lift, and is I have time, may go into more advanced things.

 

[berkeman]

Do you need to build and test the UAV, or just submit a paper describing the design? Do you have access to a wind tunnel?

 

[arunsjunk]

See if you can get an academic copy of ASWING. You need a software that can assess the fluid structure interaction of flexible wings inherent to HALE-type aircraft. At the very least, you could use AVL to understand the stability & control aspects of your aircraft design. If you want a Python solution then perhaps you could use something like this open source software: http://www3.imperial.ac.uk/aeroelastics/software .

To design HALE-type vehicles may take you more than 2 weeks, but understanding how these aeroelastic systems work could be a great learning experience. You might be benefited by reading one (or both) of these books:

1. https://www.amazon.com/dp/0486691896/?tag=pfamazon01-20
- older book, but excellent and still the best reference on aeroelasticity

2. https://www.amazon.com/dp/0470858400/?tag=pfamazon01-20
- modern aeroelastic design book

If you have to go the experimental route, wind tunnel testing highly flexible wings is a difficult process, but could provide you insight into rigid-wing aerodynamics. For aeroelastic designs, it may actually be cheaper/better to build/fly because of the difficulties in experimentally representing aeroelastic behavior in a wind tunnel (e.g., flight vehicle may have a different, potentially more damaging behaviors than what is observed in a wind tunnel test). In any case, experimental testing is necessary to validate your understanding of theory, especially in the case of aerodynamics where simulation tools are still relatively inaccurate for unsteady aerodynamics, e.g., flexible wing oscillating under gusts. This is an ongoing area of research.

 

[jag0097wtf]

Do you need to build and test the UAV, or just submit a paper describing the design? Do you have access to a wind tunnel?

No I don't need to build and physically test it, and I don't have access to a wind tunnel as well, so that wouldn't be possible. So for now, I need to just submit a paper describing the design.

 

[jag0097wtf]

See if you can get an academic copy of ASWING. You need a software that can assess the fluid structure interaction of flexible wings inherent to HALE-type aircraft. At the very least, you could use AVL to understand the stability & control aspects of your aircraft design. If you want a Python solution then perhaps you could use something like this open source software: http://www3.imperial.ac.uk/aeroelastics/software .

To design HALE-type vehicles may take you more than 2 weeks, but understanding how these aeroelastic systems work could be a great learning experience. You might be benefited by reading one (or both) of these books:

1. https://www.amazon.com/dp/0486691896/?tag=pfamazon01-20
- older book, but excellent and still the best reference on aeroelasticity

2. https://www.amazon.com/dp/0470858400/?tag=pfamazon01-20
- modern aeroelastic design book

If you have to go the experimental route, wind tunnel testing highly flexible wings is a difficult process, but could provide you insight into rigid-wing aerodynamics. For aeroelastic designs, it may actually be cheaper/better to build/fly because of the difficulties in experimentally representing aeroelastic behavior in a wind tunnel (e.g., flight vehicle may have a different, potentially more damaging behaviors than what is observed in a wind tunnel test). In any case, experimental testing is necessary to validate your understanding of theory, especially in the case of aerodynamics where simulation tools are still relatively inaccurate for unsteady aerodynamics, e.g., flexible wing oscillating under gusts. This is an ongoing area of research.

Thanks for your help. The Open source software from Imperial looks promising, do you know where I can find a tutorial or explanation of how to use and apply the code? Due to the limitations in equipment and time frame, the extent I can go with this project will be a basic paper describing the design process, and then the final design, and how it is better than the rest in terms of better endurance, or payload capability etc. as required.

I have found that NASA have some simulation programs on their website: http://www.grc.nasa.gov/WWW/Wright/airplane/tunnlint.html

But these seem very basic, and only for teaching how they work, I doubt I will be able to use it effectively.

 

[arunsjunk]

The Open source software from Imperial looks promising, do you know where I can find a tutorial or explanation of how to use and apply the code? Due to the limitations in equipment and time frame, the extent I can go with this project will be a basic paper describing the design process, and then the final design, and how it is better than the rest in terms of better endurance, or payload capability etc. as required.

There probably isn't a tutorial, however you could email the author(s), e.g. Palacios, for an example, e.g., test driver, you could learn from. If you started from scratch, then you would need to have a structural model of the aircraft that you could use as input to the Python/C++ code. For a high school level effort, just explaining (and simulating an example of) the physics of the fluid-structure interaction of a large-span vehicle should suffice, and a quantitative analysis into why a lightweight (structurally efficient) vehicle is best for better endurance, e.g., Breguet equation, would also be beneficial (motivates the study).