Lab4_jaleti

pdf

School

University of Notre Dame *

*We aren’t endorsed by this school

Course

EKMA4416

Subject

Physics

Date

Apr 3, 2024

Type

pdf

Pages

12

Uploaded by SuperIronDragonfly3078

Report
Lab 4: The swing up pendulum. (Answer Sheet) Name: Part of this write-up and figures were taken directly from Quanser’s lab manuals, Quanser Inc. 1 Joshita Alexi
1 Total energy 1.1 Experiment: the controller using E . Figure 1: Simulink model used with QUARC to run swing-up controller In this experiment, we will design a controller of the form u = sat u max k e ( E - E r ) ˙ cos( ) . (1.1) 1. Run the setup swingup.m file in Matlab. 2. Open and run the QUARC Swingup.slx in Matlab. This should look like Figure 1. 3. Be ready to hit the stop button at any time in case the pendulum goes wild. Place your critical value (minimum value to make the pendulum stand up) E r with k e = 1 here: Place your best k e and E r here: 2 Er = 3500 m5 ~ 34 seconds for pendulum to stand up We = 60 M5s Er = 50 mJ
For your best k e and E r place the graph of the pendulum angle here: For your best k e and E r place the graph of the pendulum energy E here: 3
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
1.2 Experiment: the controller E with the sign function. In the final experiment, we will design a controller of the form u = sat u max k e ( E - E r )sign( ˙ cos( )) . (1.1) 1. Run the setup swingup.m file in Matlab. 2. Run the QUARC file Swingup sign.slx in Matlab. This should look like Figure 1. 3. Be ready to hit the stop button at any time in case the pendulum goes wild. Place your best k e and E r here: For your best k e and E r place the graph of the pendulum angle here: 4 ke = 30 M5/5 Er = 85M]
2 Kinetic energy 2.1 Experiment: the controller using E c . Figure 2: Simulink model used with QUARC to run swing-up controller In this experiment, we will design a controller of the form u c = sat u max k e ( E c - E r ) ˙ cos( ) (2.1) 1. Run the setup swingup.m file in Matlab. 2. Open the Swingup center.slx in Matlab. This should look like QUARC model presented in Figure 2. Hit monitor and tune to run this file. 3. Be ready to hit the stop button at any time in case the pendulum goes wild. Place your critical value (minimum value to make the pendulum stand up) E r with k e = 1 here: Place your best k e and E r here: 5 Er = 50 ~ 13 . 922 seconds for pendulum to stand up Ke = 10 M5 , s Er = 20 m5
For your best k e and E r place the graph of the pendulum angle here: For your best k e and E r place the graph of the pendulum energy E c here: 6
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
2.2 Experiment: the controller using E c and sign function. In this experiment, we will design a controller of the form u = sat u max k e ( E c - E r )sign( ˙ cos( )) . (2.1) 1. Run the setup swingup.m file in Matlab. 2. Run the QUARC file Swingup center sign.slx in Matlab. This should look like Figure 2. 3. Be ready to hit the stop button at any time in case the pendulum goes wild. Place your best k e and E r here: For your best k e and E r place the graph of the pendulum angle here: 7 We = 45 M5/5 Er = 41MI
3 Potential energy 3.1 Experiment: the controller using E p . Figure 3: Simulink model used with QUARC to run swing-up controller In this experiment, we will design a controller of the form u p = sat u max k e ( E p - E r ) ˙ cos( ) (3.1) 1. Run the setup swingup.m file in Matlab. 2. Open the Swingup PE.slx in Matlab. This should look like QUARC model presented in Figure 3. Hit monitor and tune to run this file. 3. Be ready to hit the stop button at any time in case the pendulum goes wild. Place your critical value (minimum value to make the pendulum stand up) E r with k e = 1 here: Place your best k e and E r here: 8 Er = SOmJ ~ 2 . 8 seconds for pendulum to stand up ke = 3 M5/s Er = 305
For your best k e and E r place the graph of the pendulum angle here: For your best k e and E r place the graph of the potential energy E p here: 9
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
3.2 Experiment: the controller using E p and sign function. In this experiment, we will design a controller of the form u = sat u max k e ( E p - E r )sign( ˙ cos( )) . (3.1) 1. Run the setup swingup.m file in Matlab. 2. Run the QUARC file Swingup PE sign.slx in Matlab. This should look like Figure 3. 3. Be ready to hit the stop button at any time in case the pendulum goes wild. Place your best k e and E r here: For your best k e and E r place the graph of the pendulum angle here: 10 Pre = 20 MT/s N2 . b seconds Er = 80 m5
Present a short discussion on which one of the controllers you think performed the best. Explain what di erences you observed (if any) when using the sign function version of the controllers. 11 I think the controller using kinetic energy and sign function is the best. This is because the pendulum took the shortest time of approximately 2.2 seconds to stand up compared to the other controllers. The pendulum was also more stable and in the graph you can see that there is minimal noise and even oscillation to portray that the controller was stable before standing up. One of the differences I observed when using the sign function version of the controllers is that the time it took the pendulum to stand up was much shorter than the base controller of that particular energy. Also I noticed the Er value for the sign version of the controller is greater than its base controller.
Explain, from a physics standpoint, why po- tential energy results in a better lower bound than kinetic energy for the Lyapunov func- tion in this experiment. Could you find an even lower bound? Why can’t you find ever- decreasing bounds that result in better con- trollers? 12 Potential energy represents stored energy and intuitively maps to the "distance" from a desired stable state, making it a conservative measure for stability. Kinetic energy, on the other hand, can dissipate in many systems, possibly giving misleading stability indications. When using Lyapunov's Direct Method, potential energy fits the criteria of a decreasing, positive function, signaling movement toward equilibrium. Thus, potential energy provides a more robust lower bound for the Lyapunov function than kinetic energy in assessing system stability. Yes, you can find an even lower bound by optimizing the controller so that the pendulum swings up right immediately. But that will not guarantee a stable controller. Increasing bounds complexity doesn't guarantee improved controller outcomes. Over-optimization can lead to controllers failing in untested conditions. Inherent physical limits of systems cannot be surpassed by tighter bounds. Identifying suitable Lyapunov functions presents challenges that limit optimization depth. Balancing performance with robustness is essential for controller reliability across diverse scenarios.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help