Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Regenerative Braking Project

  1. Oct 17, 2014 #1
    I'm doing a university project which looks at regenerative braking in aircraft when coming down to land. Around 3MW of power is available when a large plane lands and we want to try and capture some of this by using generators on the wheels to charge supercapacitors, which will feed into batteries which will then be used to run electric motors for taxiing the plane.

    It would be very interesting and helpful for inspiration if anyone could share their thoughts on this project


    The landing of the plane will be simulated with a motor generator as in the picture. So I would like to simulate this by using the motor to have a high initial torque and then gradually decrease over time. Would the best way to do this be to have a variable current source going into the DC shunt motor? I also assume that the output voltage would then come from the field winding of the generator? I am a bit rusty on motor/generator concepts as I have spent more time studying electronics so please forgive me
    Last edited: Oct 17, 2014
  2. jcsd
  3. Oct 17, 2014 #2


    User Avatar
    2017 Award

    Staff: Mentor

    MW is a unit of power, not of energy.

    Just to check the overall plausibility, with an A330 as example (other airplanes should give similar values):
    Typical landing mass 160 tonnes, typical landing speed about 250 km/h or 70m/s.
    You can calculate the maximal energy you get out of this.

    Airlines try to save mass wherever possible. For a typical flight length, every kg of additional mass needs about 150g of additional fuel (source). Burning kerosene releases about 40MJ/kg, let's assume a turbine can use this with an efficiency of about 1/3. If your system is too heavy, just carrying it around in flight costs more energy than the braking process gives. How much mass do you get as maximum? Does that look realistic?
  4. Oct 17, 2014 #3
    I will do the calculations for for the A380 which I'm going to use as my model aeroplane.

    Yes like you said the extra weight is going to be very important so I will be sure to look into the extra fuel required from the weight of the system to determine if it is feasible or not.

    Also though, electric taxiing is a great improvement over the traditional method of moving aircraft on the runway which requires thrust from the engines. It is much more efficient, safe and doesn't have some of the other drawbacks that jet engine thrust does while on the ground. There is already a working electric taxiing system, however it is powered by 1 or 2 (presumbly kerosine or diesel) generators rather than a self sufficient source such as regenerative braking
  5. Oct 17, 2014 #4
    While I cannot offer much technical insight, I think that you are on the right track with this. According to this article, "A Boeing 747 can consume a tonne of fuel and emit several tonnes of carbon dioxide during an average 17-minute taxi to take-off." It would be interesting to investigate how much fuel this really is as a fraction of the total the aircraft carries on a typical flight. I assume that larger aircraft such as these are scheduled on longer-hauls.
  6. Oct 17, 2014 #5


    User Avatar
    Science Advisor

    Cool project.
    Some things that come to mind:
    What proportion of the braking force does the wheel system provide?
    I'd guess for large planes that most braking force is provided by reversing engine thrust reversal.
    the energy capture from wheel braking may be limited by the braking force that the landing gear is designed to apply (regularly, not emergency conditions).
  7. Oct 18, 2014 #6


    User Avatar
    2017 Award

    Staff: Mentor

    Well, I guess that fraction would be increased if it can be used.
    The same system could also be used to taxi from the gate to the runway, which would increase the overall effect by a factor of 2.
    On the other hand, you could use those pushback vehicles for both ways. The weight of their energy storage would not matter as they have to be heavy anyway and they stay on the ground. No one does that.

    Back to the original question: airplanes land at a high speed and then slow down. It would be interesting to simulate this very sudden start - maybe with a connection that gets established while the motor is running?
  8. Oct 20, 2014 #7
    I have clients working on this and can confirm the real motivator is to have "wheel power" allowing electric taxiing of the plane, typically off the off of the APU - without starting the main engines. This is where the weight justification comes from - THEN the landing REGEN becomes viable / valuable.....

    One other limiting factor will be the time you have to work with, the plane needs to slow down quickly, so the simulation needs to factor that ( 10 to 15 seconds ??) as well as managing wheel slip - different conditions, ensuring the wheels keep traction while braking - as I am writing this I am realizing -- you need an ABS type sensor to manage torque and eliminate wheel slip...so the wheel needs a minimum turn-on speed - then field current is managed in a feedback loop. The regen energy probably also needs to be converted / managed electronically so that the storage medium does not get abused - this type of very high short term current can be managed with thermal mass... and we are back to calculating the payback vs mass (fuel burn ) again....

    Good project --

    Oh for the simulation - if you can -- you may want to just use a flywheel, a rotating mass energy calculation is easier than measuring the power input from the motor ... esp for the lower energy, trial and error - learning phase.
  9. Oct 20, 2014 #8
    I'm curious about this and find it interesting. Just clarifying that your clients hope to draw energy from the Auxilary Power Unit to taxi, not that they want to be able to taxi with the APU shut off I assume?
  10. Oct 21, 2014 #9
    Yes, this is one of reasons why most of the kinetic energy of an aircraft can't captured in this way.
  11. Oct 21, 2014 #10
    Yes - I meant running from the APU - running off is an ambiguous figure of speech... FYI - originally they wanted to use the nose wheel, did't work. As a M.E. -- can you figure out why?
  12. Oct 21, 2014 #11
    No idea. Can I guess?

    My guess is that a certain amount of torque would be needed to overcome rolling and static friction. Making the blind assumption that larger motors can produce a greater maximum torque, perhaps the motor needed proved too large for the nose wheel. They seem to look a bit lighter than the rear pair. It may be more practical to divide this load up between two mid-size motors rather than one big one.
  13. Oct 21, 2014 #12
    Nope, think of a model of a modern airliner, apply force at the front wheel...a lot of money was spent on this before the weakness was recognized.
  14. Oct 24, 2014 #13


    User Avatar

    Staff: Mentor

    Intriguing question...

    Since the OP has not responded, I'd guess that it has to do with not enough traction at the nosewheel to be able to move the plane (especially forward). But I'm not getting your comment about a lot of money being spent before the weakness was recognized. Can you elaborate?

    BTW, Wikipedia has a pretty good article on "Pushback" procedures for airliners: http://en.wikipedia.org/wiki/Pushback

    Last edited by a moderator: May 7, 2017
  15. Oct 24, 2014 #14
    That tractor has much bigger wheels than the airliner
  16. Oct 25, 2014 #15
    The plane is pretty balanced on the middle tires, when they tried to use the nose wheels, they would not keep traction, and if the plane was on the an incline they really had problems.
  17. Oct 26, 2014 #16
    I would guess that the initial touchdown will be very unstable like you say especially for the front wheels. I'd also imagine that the huge amount of initial torque generated would be problematic to the generator and I have no idea how a conventional generator would cope with such a high and unstable force. We done a few tests on flight simulator (not sure how accurate) but after the initial touch down there is around 20-25 seconds of where the plane slows down from 140 knots to standstill. So ideally we have say 20 seconds usable time to capture the kinetic energy.

    Also since we are looking at the Airbus A380, it has 22 wheels so potentially more space for experimentation, however it will be difficult to know exactly how each wheel would behave

    Last edited: Oct 26, 2014
  18. Oct 27, 2014 #17
    The generators used on the main engines on the A380 are variable frequency at 360-600Hz so they can operate at different speeds. Maybe something like that will be able to cope with the landing although they will be much larger.

    I'll look at the current technology for regenerative braking in electric cars and trains as those generates must operate on the same principal
  19. Oct 27, 2014 #18

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    there's already huge disc brakes present. It seems to me plausible they could be magnetized to give poles, then an armature added to the wheel assembly, approximating a dynamo resembling those in homemade flat disc wind turbine alternators.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook