Dear colleagues in System Engineering and Automatic Control:
I used to apply classic controller on industrial processes such as PID controllers
Do you have an idea about what are the recent and modern techniques that scientists and researchers are using ?
Dear Basim Touqan,
We can mention in a non-exhaustive way such PID controller based on self-adaptive technique, fractional PID controller, PID and fuzzy set controller, optimization of PID parameters using heuristic or meta-heuristic algorithms...
For more détails and information about this subject , I suggest you see the links on the topic.
https://www.sciencedirect.com/science/article/abs/pii/S0016003218300437
https://warwick.ac.uk/fac/cross_fac/iatl/reinvention/archive/volume5issue2/hornsey/
Conference Paper Intelligent PID controllers
Article General type industrial temperature system control based on ...
Article Fractional-Order PID Controllers for Temperature Control: A Review
Article A Robust Fractional-Order PID Controller Based Load Frequenc...
Article A Review of Recent Developments in Autotuning Methods for Fr...
Best regards
Hi,
In addition to the classic PID controllers, in modern control theory there are other complex control techniques such as adaptive/robust control, neural networks, fuzzy control, and methods for optimization such as genetic algorithms, heuristics/metaheuristics algorithms. Actually in many cases, the PID control is combined with modern techniques.
Dear Basim Touqan,
I will suggest you to choose a controller according to the complexity of your system and power calculation of your embedded computer.
PID still a great controller if you have to control a SISO(Single Input Single Output) system. It can achieve precisely a target point and is easy to implement.
If your system is a linear MIMO(Multiple Input Multiple Output) system I'll suggest you to use a LQR( Linear Quadratic Regulator) controller. It's a easy to use and powerful controller.
Most of the time these two controller are sufficient in industrial process control.
For more complex system and powerful embedded computer there is other complex theories that you can use such as :
robust control(for more guarantee in stability of the system);
impedance control(powerful for humanoid robots);
model predictive control(for linear or non linear system. It's able to track evolution of system dynamics but with high need of power calculation);
linear quadratic gaussian(for observable linear system if sensor are noisy or there is not sensor for all system's states) and his non linear variants;
...
My english is not perfect because I'm a french speaker but I hope that it's understandable and helpful for you.
Best regards,
Jonathan
...
I suggest to Follow #relaxc and #relaxctuner http://relaxc.appedge.com/
free tool to tune #relaxc and to make a PID well-balanced.
Hello,
you can try this technique:
https://arxiv.org/pdf/1305.7085.pdf
Best,
Loïc
- Badges
- Science topic