Dear Dhruvin,
This is a very broad question. There are many controllers and there is no philsophical restriction on the kind of controller you may use. This choice may be restricted by the amount of information you have (do you have a model or not, availability of sensors, ...). The structure of the controller may be imposed or guided by your system. For instance if your temperature set-point depends on another controlled quantity, then you could go for a cascade controller ...
Generally speaking, simple linear controllers like PI or PID controllers may be well suited and sufficiently performant. There are relatively easy to design and tune, and cheap to implement if you have a "real" system. If you have a model of your system you could also end up with an Model-Predictive Controller (MPC controller).
Personally, I believe that there is no need in choosing an advanced controller if you don't have strong specifications (like strict following of temperature profiles, fast response, robustness, ...), i.e. if you just want to close the loop. But for sure, "no free lunch", with simple controllers you will certainly end up with good but perfectible performances. More advanced controllers will probably lead you to better tracking performances, but are more challenging to design and tune.
You should read one (or more) of these books
Process Dynamics and Control, Dale E. Seborg, Duncan A. Mellichamp, Thomas F. Edgar, Francis J. Doyle, III, John Wiley & Sons
Process Dynamics, Modeling, and Control, Babatunde A. Ogunnaike, W. Harmon Ray, Oxford University Press
Process Control Theory and Applications, J.-P. Corriou, Springer Science & Business Media
One thing to consider when looking at the control system is knowing the total thermal mass or capacitance in electrical parlance. this is going to define the responsiveness of your system and essential to choose a good control loop.
- Badges
- Science method