Systems
Systems can eventually be simplified so we can analyze them easily.
Where we can analyze with inverse Laplace transforms.
Signals
Input
- Reference (generally a step or ramp):
Perturbations
- Disturbance:
- Sensor Noise:
Output
- Output response:
We always want .
Analysis
- Is the plant stable?
- Where are the poles/zeros? where the poles are roots of and zeros are roots of
- What does the impulse response, step response, and ramp response look like?
- Is the time-domain behavior acceptable?
Reviewid1((Some text))
Open Loop Topology
In theory, if we don’t like the behavior of the Plant, we can design the controller to replace the poles:
This is an open loop with no feedback:
The overall transfer function becomes:
If , we can cancel the poles of the Plant and replace them with . The plant must be stable for us to do this. Pole cancellation does not work on unstable plants.
Closed Loop Topology
flowchart LR id1((Σ)) --> id2("Dc(s)") id2 --> id3("G(s)")
If the output
Example of Closed Loop Inspection
This must be unstable because there is a pole in the Ride-hand plane/
Can the plant be made stable?
- Move the unstable pole from the RHP to the LHP. How do we do this? feedback
In a PID controller, , where is a constant gain.
Note that is forward gain, which is equivalent to
The is the feedback gain.
When , we call it unity gain.