Great video, as always! Can you believe that during my degree in Physics (in Brazil), we only learned how to solve things using a computer in two classes and NEVER had to use these techniques throughout the entire degree? It blows my mind that we spend all of our time solving things by hand, because, if you think about it, if I'm able to program a numerical solution to the problem, then I definitely understood the math the professor is teaching. Education need to remodel itself ASAP!
Haven't your degree a course in computational physics? I'm from Europe it's quite common throwing a course in the first year where they teach you to use a computer algebra system and some numerical techniques. Definitely, it's worth learning even if they didn't teach you.
@@SimulatingPhysics We did have a course on numerical methods. My point is we never had to use the things we learned in that course throughout the rest of the degree. My professors used to say we're been trained to do research and would comment that they never did anything by hand on their own research (but they didn't notice the obvious problem: if research can be done using a computer and we were never "forced" to use numerical techniques, then we were not actually trained to do research...)
"optimal" is a bit of a stretch - you just tried 3 control inputs (as you say at the end, not criticising here, just providing further ideas). next step would be to parameterise the whole trajectory depending on the inputs, which you could e.g. set to be constant over small time intervals (very deep rabbit hole about different methods like single/multiple shooting, collocation, differential dynamic programming, adaptive meshing, the list goes on). then define an objective function (e.g. maximum height) and find the optimum with a gradient based optimiser like IPOPT (or i guess the method from scipy.optimize would work too). this way it's also easy to handle constraints (e.g. maximum thrust and acceleration), nonconstant atmospheric pressure, more realistic drag models with (crude approximations of) supersonic effects, etc and still get the optimal solution (up to discretisation error, and i guess local minima, but probably these won't be an issue for this system).
Great video man ! You bring together my two favorite things in the world, Physics & Python. Your content quality is literally awesome. Now I'll have to binge watch all your videos. Keep it up!
Hi, Mr. Problem Solver, I understood that this is a good way to model how to use fuel to make rocket fly far , however what about changing weather, not stable atmosphere, changing friction coefficient b in real time and real mass of fuel by doing real physical experiment on the street ?
What about max q (max dynamic pressure)? I think you want max q to occur where atmospheric pressure is low. It would be neat if your plots showed max q.
Great problem. But I had a little trouble following it because thrust is not mentioned. To make the formula easier to understand, can dm/dt can be replaced with thrust?
Itd be interesting to see how you would chamge the equations if you modeled friction as a function of position (consider it proportionaly to air density, which decreases as position increases).
It would be great to see a gradient of friction (in function of the altitude) instead of setting it constant... Maybe, it has been left to the reader XD Great content, first time on your channel !! +1 sub
Great video! Can you highlight more on the standardization of equations and how to get an intuitive idea of it? Also, a bit more explanation on the system of equations S will be much appreciated in any future videos. Keep up the good work.
Of course dude! I was thinking on making a video on just this topic in the future. If you're curious about the S thing, I think I explain it a little better in my SciPy Tutorial video where I spend time talking about solving ODEs in python!
I think this is a demonstration of "gravity drag." The idea is that the longer you burn (turn the rocket on), the less efficient the ascent. With n=5, the rocket approximately doesn't start burning at the beginning, and then burn all of its fuel at t=T_0. Which means less burn time, less gravity drag, more efficient.
I have to clarify, gravity drag has an effect only when the rocket is pointing vertically. The less time it burns vertically, less gravity drag. So to minimize gravity drag, you need to point horizontally ASAP or deplete your fuel ASAP when pointing upwards. (You need to horizontal speed to achieve orbit anyway)
@@tiagovportela You are correct in the statement that F=ma is not applicable in mass variant systems, but the equation F=dp/dt, which I'm using here, is indeed valid in mass variant systems.
Excellent video! Absolutely splendid! Thank you very very much for making those videos🙏. Unique and amazing content. Very very Helpful. You’ll not believe how much I love your videos and how much I learn from you!! You’re the best!!
Hello..I am a PhD student in physics from Iraq..I hope you can help me find codes in the Python program to study the Fe(II)particle (ising model 2D)to determine the spin crossover of the electrons and find the energy..with many thanks to you.
Great video, as always! Can you believe that during my degree in Physics (in Brazil), we only learned how to solve things using a computer in two classes and NEVER had to use these techniques throughout the entire degree? It blows my mind that we spend all of our time solving things by hand, because, if you think about it, if I'm able to program a numerical solution to the problem, then I definitely understood the math the professor is teaching. Education need to remodel itself ASAP!
Haven't your degree a course in computational physics? I'm from Europe it's quite common throwing a course in the first year where they teach you to use a computer algebra system and some numerical techniques. Definitely, it's worth learning even if they didn't teach you.
@@SimulatingPhysics We did have a course on numerical methods. My point is we never had to use the things we learned in that course throughout the rest of the degree. My professors used to say we're been trained to do research and would comment that they never did anything by hand on their own research (but they didn't notice the obvious problem: if research can be done using a computer and we were never "forced" to use numerical techniques, then we were not actually trained to do research...)
+1 for The Road to Reality on the shelf
why does this channel have so fuckin low number of subscribers??
you are creating very good contetnt bro
Exactly
Hahaha you gotta spread the word then!
Pretty dense stuff. I had to watch it throughout the day, but overall I enjoyed it :D
UNLESS WE OVERCOME THE ROCKET EQUATION WE ARE NEVER GOING TO GET TO THE STARS ARE WE ! ! ! ? ? ?🤔🤔🤔🤔🤔🤔🤔
"optimal" is a bit of a stretch - you just tried 3 control inputs (as you say at the end, not criticising here, just providing further ideas).
next step would be to parameterise the whole trajectory depending on the inputs, which you could e.g. set to be constant over small time intervals (very deep rabbit hole about different methods like single/multiple shooting, collocation, differential dynamic programming, adaptive meshing, the list goes on).
then define an objective function (e.g. maximum height) and find the optimum with a gradient based optimiser like IPOPT (or i guess the method from scipy.optimize would work too). this way it's also easy to handle constraints (e.g. maximum thrust and acceleration), nonconstant atmospheric pressure, more realistic drag models with (crude approximations of) supersonic effects, etc and still get the optimal solution (up to discretisation error, and i guess local minima, but probably these won't be an issue for this system).
THE ROCKET EQUATION REALLY SUCKS & SOME VERY CLEVER PERSON(S) MUST & WILL FIND A WAY OF BYPASSING OR SIDESTEPPING IT ! ! ! ? ? ?🤔🤔🤔🤔🤔🤔🤔
'THE ROCKET EQUATION' IS WHAT IS STOPPING US FROM REACHING THE STARS & THIS IS WHY IT REALLY SUCKS ! ! !😭😭😭😭😭😭😭
FOR YEARS I BELIEVED THAT THE ROCKET EQUATION IS NON NEGOTIABLE BUT IS THIS REALLY TRUE ? ? ?🤔🤔🤔🤔🤔🤔🤔
Great video man ! You bring together my two favorite things in the world, Physics & Python. Your content quality is literally awesome.
Now I'll have to binge watch all your videos. Keep it up!
When you say "standardized", do you mean "normalized"? I'm familiar with normalizations in ML applications.
Hi, Mr. Problem Solver, I understood that this is a good way to model how to use fuel to make rocket fly far , however what about changing weather, not stable atmosphere, changing friction coefficient b in real time and real mass of fuel by doing real physical experiment on the street ?
What about max q (max dynamic pressure)? I think you want max q to occur where atmospheric pressure is low. It would be neat if your plots showed max q.
Great problem. But I had a little trouble following it because thrust is not mentioned. To make the formula easier to understand, can dm/dt can be replaced with thrust?
The other fascinating problem: what is the most efficient way to burn fuel in order to land a rocket on Earth?
Itd be interesting to see how you would chamge the equations if you modeled friction as a function of position (consider it proportionaly to air density, which decreases as position increases).
It would be great to see a gradient of friction (in function of the altitude) instead of setting it constant... Maybe, it has been left to the reader XD
Great content, first time on your channel !! +1 sub
I was hoping there would be a variable air friction scenario, where it was dependent on position in the atmosphere. Great video as always!
Great video! Can you highlight more on the standardization of equations and how to get an intuitive idea of it? Also, a bit more explanation on the system of equations S will be much appreciated in any future videos. Keep up the good work.
Of course dude! I was thinking on making a video on just this topic in the future. If you're curious about the S thing, I think I explain it a little better in my SciPy Tutorial video where I spend time talking about solving ODEs in python!
Fantastic video for a deeper intro to computational physics
excelente video lo disfrute mucho :D
Can you kindly share any of ur social platform to get in touch for a chat
There is a discord server link in the description of the video
Yes come join the discord!!
I think this is a demonstration of "gravity drag." The idea is that the longer you burn (turn the rocket on), the less efficient the ascent.
With n=5, the rocket approximately doesn't start burning at the beginning, and then burn all of its fuel at t=T_0. Which means less burn time, less gravity drag, more efficient.
I have to clarify, gravity drag has an effect only when the rocket is pointing vertically. The less time it burns vertically, less gravity drag. So to minimize gravity drag, you need to point horizontally ASAP or deplete your fuel ASAP when pointing upwards.
(You need to horizontal speed to achieve orbit anyway)
very interesting process
Sorry but you can't use this formula.
en.m.wikipedia.org/wiki/Tsiolkovsky_rocket_equation
Is there an issue here? Those results correspond to the same problem but without gravity or friction.
@@MrPSolver you can't apply the newton second law to mass variable systems. Its violate the Galilei relativity principle.
@@tiagovportela You are correct in the statement that F=ma is not applicable in mass variant systems, but the equation F=dp/dt, which I'm using here, is indeed valid in mass variant systems.
@@MrPSolver en.m.wikipedia.org/wiki/Tsiolkovsky_rocket_equation#cite_note-Halliday-9
@@MrPSolver im referring to dp/dt
im glad i found your channel
have you read the road to reality?
Excellent video!
Absolutely splendid!
Thank you very very much for making those videos🙏.
Unique and amazing content. Very very Helpful.
You’ll not believe how much I love your videos and how much I learn from you!!
You’re the best!!
Honest criticism.. drop the (impromptu?) music videos. They “remove” more from the videos quality than they add.
DROP ALL THE MUSIC pls
Hello..I am a PhD student in physics from Iraq..I hope you can help me find codes in the Python program to study the Fe(II)particle (ising model 2D)to determine the spin crossover of the electrons and find the energy..with many thanks to you.