Scribd
Upload
13 views3 pages

CSD PPdoc

This presentation summarizes the modeling and control of a ball and beam system. It begins with an introduction to the plant and goals of position control. Modeling is then discussed, using both whitebox and blackbox approaches to derive transfer functions for the angular velocity, angle, and position subsystems. A PD controller is chosen and implemented in simulation. Steps are taken to test the system, and real-time reference input is also demonstrated, though more testing was prevented by plant issues. Overall the project achieved satisfactory position control through modeling and controller design.

Uploaded by

hung kung
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
13 views3 pages

CSD PPdoc

This presentation summarizes the modeling and control of a ball and beam system. It begins with an introduction to the plant and goals of position control. Modeling is then discussed, using both whitebox and blackbox approaches to derive transfer functions for the angular velocity, angle, and position subsystems. A PD controller is chosen and implemented in simulation. Steps are taken to test the system, and real-time reference input is also demonstrated, though more testing was prevented by plant issues. Overall the project achieved satisfactory position control through modeling and controller design.

Uploaded by

hung kung
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Martin

Kassem
Julien

Introduction
Slide 1: Quick introduction, Quick presentation of the lab, of the team…

Slide 2: Overview, quick slide where we are just planning the presentation

Slide 3: Introduction of the plant, explanation of ‘ball and beam’,(How it’s working, the input,
instrumentation, the sensors, the actuator, goal is to maintain the ball etc…) explain the
specification and the goal ( or maybe in another slide?)

Slide 4: Static characteristic, explain the difference between the graph and the real
measure, deadzone, etc.. [-1.05;2.05],explain that we had to linearize this

Slide 5: Cascade control was chosen (If anyone has a better quality image of the cascade
control than mine please put it in place of mine)

Modelling
Slide from 6 to 10: Blackbox and whitebox
So first we began with a white box approach in order to compute theoretically the
mathematical relationship that describes the interactions within the system, white box offers
valuable insights by revealing the order of the system, including the number of zeros and
poles which are important parameters.
So we made 3 equations for each subsystem, (Angular velocity, angle and position). We
made the equations with our knowledge in physics and mathematics.

Once we had the “theoretical” equation, we have to verify them with the black box approach,
that offers the advantage of speed and reliability, furthermore we had already compute the
equations with the white box methods so we can compare them with the blackbox’s result so
it can allows us to verify the white box equations. particularly regarding the system order.
The concept in the whitebox approach is to introduce a step at the input of the system and
analyzing the step response, with MATLAB the transfer function of the system can be
determined. So like the white box approach, we’re doing a black box approach for each of
the 3 subsystems.

DC motor and angular velocity


Input : current Output : angular velocity
First system thats need to be characterized was the dc motor
Reference is a step who start slightly below 2.05v because of the deadzone
We send 3 differents steps and we comparing the simulated output with the real output of the
plant for each step, of couse there are differences between the simulated and the real one.
The white box show that DC motor can be approximated by a first order, therefore the first
order approximation was chosen.
Moreover, design choice does not matter that much due to the fact that inner loops are faster
that outer loop.

Angle
Input : angular velocity Output : Angle
Reference is a step for PD controller.
Simple integrator as expected in the white box part. theoretical and practical are really close
(cf graphics)

Position
Input : Angle Output : position
In this case, it would be optimal to start the experiment with the ball in the middle of the
beam, but it wasn’t possible as it required a controlled system. It was decided that the ball
would start furthest from the position sensor, so that for a positive step, the ball would have
to travel the whole beam before colliding and bouncing back. The amplitude of the step has
to not be too low, as the ball might not even move and friction would not be negligible.
Double integrator as expected in the white box part. theoretical and practical are really close
(cf graphics)
There is delay due to friction

Controller
Slide 11: Talk about the controller, why the choice of PD controller, phase lead controller…

How we went from G3 to te CL. (cf report)

Simulation
Slide 12: Simulink, talk about, discretization,zoh, tustin, of course this is the simulation and
the real results will be different but close

Final result
Slide 13: Various step
Slide 14: consecutive step,Talk about the error when the reference is above 35cm due
to the instrumentation
Slide 15:step lower noise

Bonus
Slide 16 - 17 : We were able to try a real time regulation, which means that a human is the
reference with a signal of a potentiometer which can be scaled by another potentiometer in
the range 0-10V, This signal is acquired by the computer, just like the other signals.
It means that we could choose the position of the ball in real time with that potentiometer
instead of choosing it in the code, but the plant stopped working before new test could be
conducted.

Then, we were able to put a sinusoidal input to see how the plan react at this, (Cf video after
this slide).
Demonstration
Slide 18: i put the entire video (1min12)( maybe we should accelerate?)

Conclusion
Slide 19: After deep analysis and tests, we were able to handle the real plant, we managed
to control the position of the ball on the beam.
These lab sessions were really useful because we didn’t know the results beforehands so
we had to think and to make wise choices like the choice of the controller, the white/black
box approaches, the cascade control approach… discussions were held for theses subjects
and a detailed examination was made.
Testing with step references using matlab in order to compare with white box revealed
satisfactory performance and show that we made the good choices and the good
approaches.
The only regrettable aspect is the loss of data for real-time regulation, which would have
been an added value in the realization of this project.

Last slide: Thanks for listening to us if you have question blablabla

You might also like