roadmap of occ_airconics #27
Comments
|
Hi @TsingQAQ , I have also had the thought that replacing the external VSP with something like occ_airconics would be an interesting addition to SUAVE; I do plan a For a bit of context: I started occ_airconics as part of my PhD project, for which I am looking into optimizing the top level component configuration for a given mission (perhaps a better explanation can be found here). I have been a little slow in the release of version 0.3, as most of my PhD work will be in it. For this release, I plan to include our methods for evolving aircraft concepts, utilizing knowledge of how each change affects the overall performance. Admittedly, this is perhaps a niche topic, and could be a separate tool, but it just happens that I've included it here. Hopefully that clears it up? As I am the sole contributor at the moment, the project is largely going where my PhD needs it to, but I'm interested in suggestions and pull requests. |
|
Many thanks for your reply. I'm a postgraduate student of grade 1 and thanks to Professor Sobester's book and your greate python code so that I've get started at aerodynamic design and optimization area. In my point of view, Professor Sobester's books is more than just aerodynamic design. From where I study aerodynamic optimization often conductes during the preliminary design phase, some key parameters like wing area will largely remain the same. Since the high fidelity analysis tools and surrogate models are becoming universal, there's a trend(just in my point of view) that the design optimization will begin at the very begining of the conceptual design stage. This is what I could get from Sobester's book, and also the SUAVE's main purpose. And cungratualations of your AIAA paper, I took a glance of that though I'm stiil too naive on this area and to give any useful suggestions on this work, but I'm still very curious about this paper, semms that u're going to optimize through a topology tree to achive any geometry configuration, though today's CFD's capability, more specific the preprocessor's capability, is far less than what CAD can get. Like if we're going to optimize a common wing and tube airplane, that is quite easy cause the topology will not change significantly and we could just modefy a standard mesh script slightly to get all the optimize candidates run. But if we have variety of topologies which hugely differs from each other, how could we be able to analysis all these using high fidelity tools? SUAVE tends to solve this by using physics based empirical equations and low fidelity tools like vertex lattice method. But if we're going to use CAD, our ideal aerodynamic solver then must be CFD right, so I'll really appreaciate that if you could give some hints on these problem. I'm also working on a school project like occ_airconics, though more than a model builder, that one requires design tools, solvers(SU2. etc), optimizers and surrogate models, I've took a survey on existing tools, there're some commercial program like Delft ParaPy which do a similar job. There're some opensource programs on aerodynamic design and optimization area like SUAVE, OpenMDAO, SU2 and so on, so there's an opportunity that a python based model builder will be included! |
Uh oh!
There was an error while loading. Please reload this page.
Hi #Paul, I'm pretty curious about the roadmap of this great program. Seems that you're making effort to connect occ_airconics to SUAVE, is that the main purpose of occ_airconics now? I've tried SUAVE before but unfortunately it do not support python 3.x yet, what is your plan on this problem?
Also, seems that SUAVE're using the NASA OpenVSP as a model builder, it would be wounderful if occ_airconics could be a candidate of that.
The text was updated successfully, but these errors were encountered: