LQR regulator doesn't regulate the system
3 views (last 30 days)
Show older comments
Hello,
I used the function lqr to get a regulator that helps me to regulate my system.
My function is:
r = lqr(A,B,Q,R)
With this function i get a regulator that gives me two conjugate complex poles. Thus the system is not regulated, but continues to oscillate. I don't get any errors or warnings.
How can I fix this problem? .
2 Comments
Sam Chak
on 22 Mar 2023
Hi, in order to investigate, could you show the matrices for A, B, Q, R?
Also provide the desired settling time if you do not have any specific constraint for the control input u.
If (A, B) is controllable, then the desired settling time is sufficient for the eigenvalue design.
Answers (1)
Sam Chak
on 22 Mar 2023
Hi @DoctorCrow
Your selection of and results in the closed-loop eigenvalues having negative real parts very close to zero. Consequently, although the system states converge to steady-state, they also oscillate for a much longer period of time.
A = [0 1 0 0; 0 0 14.72 0; 0 0 0 1; 0 0 -3.066 0];
B = [0; 0.001; 0; -0.000125];
Q = [1 0 0 0; 0 1 0 0; 0 0 1 0; 0 0 0 1];
R = 1;
[K1, P, e] = lqr(A, B, Q, R)
sys1 = ss(A-B*K1, B, eye(4), 0);
step(sys1)
Increasing Q should improve the system response, as shown in the following plot.
Q = 1e7*eye(4);
[K2, P, e] = lqr(A, B, Q, R)
sys2 = ss(A-B*K2, B, eye(4), 0);
step(sys2)
2 Comments
Sam Chak
on 22 Mar 2023
You are welcome, @DoctorCrow. If you find the solution helpful, please consider accepting ✔ and voting 👍 on the answer. Thanks a bunch! 🙏
See Also
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!