I'm trying to control temperature with PID(Kp,Ki, Kd). During the rise time the system use one PID and when target is reached the system switch to another PID with other parameter in order to eliminate the oscillation. I know that I can remove oscillations increasing Kd but in this case it does not work.
Results of real time system
Control temperature with constant PID: see figure onePID.png
Control temperature using PID1 during rise time (blue color) and when temperature reach the target, the controller is changed to PID2 (Steady state error: black color): see figure TWOPID.png
Hi Ruben Hernando ,
You can make a gain schedule so that the transition from Kp = 7.2 to Kp = 5.0 is smoother. In other words, the control action doesn't make a sudden 'jump' when error ≤ 0. You can design the gain schedule, Kp = f(e), as a function of error.
For example, if the error is far away from the origin, Kp = 7.2. If the error is close to the origin, then Kp approaches 5.0.
A classical temperature-heat model is highly nonlinear, therefore it s pretty hard to control with a stable gain PID controller. In simulation, you can use this "switching" method. However, you will have to reset integral gain while switching in real world app.
Another mathematical issue is that; even though you can model all the system in terms of linear parameters, all the system is a "dynamic" model, which means that you can not simple "switch" between controller models, the system is not "static", thus switching between model at any "t" time will add extra computational cost.
If you like to have a deeper understanding of this issue, you can have a look at the topic "convolution".
Dear Ruben,
Try to apply on your system Nonlinear fractional PID control. Here are links and attached file on this subject.
Fractional-order PIλDμ controller design - ScienceDirect
https://www.sciencedirect.com/science/article/pii/S0898122113001247
Design of a Nonlinear Fractional Order PID Neural Controller for ...
https://www.researchgate.net/.../306944317_Design_of_a_Nonline...
Best regards
Nowadays Cascaded PD-PI controllers are being used for better performance than PID controllers. These can be tuned by PSO algorithm. You may try that.
Dear Ruben,
I propose you to try to change the PID controller parameters when the absolute value of the error is less then some value (e.g. 5 or 10 degrees), not when the error is zero!
So, the second PID controller will work around each steady state.
I propose you to use LQR controller, it is better to control multi-variable system.
check the the articles below, they may help.
Article A comparison of PID and LQR controllers for position trackin...
Article Robustness and Disturbance Rejection of PD/H-∞ Integrated Co...
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Control Theory
- Control Systems