If you have powerful simulators, we can substantially reduce the wind tunnel testing which is a big advantage. We shall test only those configuration on planes which can not be simulated.
Spot on. Hence supercomputers. I wonder if we can use the combined capacities of the IMD and ISRO supercomputers for this. Perhaps on a contractual basis of sorts.
You conveniently ignored the fact that Indian performances have been trending upward. Or the fact that ACM ICPC isn't even advertised all that heavily in India
Probably not. DRDO apparently recently set up their own supercomputer lab, with capability in petaflops, so maybe they are using this.
We have sufficient computing power as well but what is necessary is the storage of RAW data got from simulation of results of wind tunnel testing and actual flight. As we keep updating our software, we can get better and better simulation result and minimize wind tunnel testing and refine the design to reduce the design freezing time.
I didn't ignore the upward trend. I said we are getting there. But not fast enough unfortunately. ICPC is quite popular now. And even if ICPC wasn't, ICPC level of problems should be. None of the Roadside Institute of Engineering and Management types teach that kind of stuff. Our B.Sc programs are also lopsided with a 3 year period instead of 4. The old mentality of Engineering entrance fails pursuing BSc still exists sadly. Higher entrants in BSc programs don't indicate that the quality of the same has improved, simply because top companies here seldom hire BSc grads and if they do then they are paid less than the Tech grads.
I see this as a god given opportunity for a bunch of universities to team up and work on at least 4 different topics: Faster data pipelines for supercomputing grids, numerical simulation of fluid dynamics (CFDs), numerical simulation of mechanics of materials, numerical simulation of aerothermodynamics.
What you said is right. Thankfully the GoI and DRDO seem to recognise this. They've been involving many universities. As I understand it, AJIT and SHAKTI were made with the intent of replacing x86 and ARM chips in a lot of areas in India including like education (Arduinos and Raspberry Pis), and supercomputers. Apparently we will see those chips in commercial applications within 2 years. According to Ministry of IT, they aim to get AJIT and SHAKTI into supercomputers not too soon after. Supposedly other work is being done for supercomputers.
Shifting to RISC-V will be a great leap forward by India. We can use our custom implementations of RISC-V for our specific needs. And I do hope that Indian homegrown consumer tech can take this forward. If Android can be ported over to RISC-V, Indian fabless chipmakers don't have to pay boatloads of money to ARM. Windows already runs on ARM and I'm sure its a matter of time before it runs on RISC-V.
Yeah it will be massive. And the fact that they are targetting education institutions first is a good sign. An indigenous RPI equivalent will help getting computers to more poor people, so more people can experiment with these computers first. Any low level assembly type things they learn on it will make it easy for them to stick to RISC-V, but make moving to ARM and x86 harder. Somewhat like microsoft's own strategy. They definitely stand a chance if the funding is maintained.
They need to fix the requirements for 1st batch of mwf and get cranking. All extra capabilities like retractable probe and cft can come with next batches. Otherwise IAF will demand all extra capabilities into first production batch itself and delay orders.
Most of simulations happen in daily R&D (for purely scientific purpose) and designing stage. Once your whole plane design is accepted by your clients, the next stage is use wind tunnel to test your design thoroughly and real flying test of prototype. There is no much simulation can do in this stage unless you find a huge error in original design. In this case, it is generally meaning the you need to back to re-design the whole thing.
Wrong. Simulation doesn't replace the wind test or flying test but the manual mathematic calculation stage in past.
You are wrong once again. Manual mathematics is a very small part of design validation such as compliance of area rilling etc. It is also done on computer now.It doesn't consume big time. It is followed by simulation where those parts of design are tested which can be simulated such as boundary layer separation.
Wind tunnel testing is also computer controlled and aided to understand even minute changes. In addition, 3D printing of parts enable in understanding aerodynamics and many models can be made to find the most optimal ones for use.
Any airflow pattern or other physics simulation is kind of mathimatical calculation. The result of wind tunnel test is also presented in numbers.
That was why I said - manual mathematics calculation was replaced by computer simulation.
No, all these parts of design should have already been simulated even before the whole program started. Once the clients put down their requirements, the designers should pull out the individual parts designs and focus on the whole plane integration and validation.
| Thread starter | Similar threads | Forum | Replies | Date |
|---|---|---|---|---|
|
ADA Tejas Mark-II/Medium Weight Fighter | Knowledge Repository | 6 | |
|
|
AERO INDIA 2021 | Science and Technology | 308 | |
|
ADA Tejas (LCA) News and Discussions | Indian Air Force | 17457 | |
| P | ADA DRDO and HAL Delays a threat to National Security | Internal Security | 20 |
