How to set state space model structure for system identification?

1 次查看（过去 30 天）
Rakshith Badarinath
回答： Rakshith Badarinath 2022-2-10
I am using the system identification toolbox to develop a continuous time state space model for 2 input one output system with disturbance component. The number of states is 1 (order = 1).
dx/dt = A x(t) + B u(t) + K e(t)
y(t) = C x(t) + D u(t) + e(t)
A is 1x1
B is 1x2
u is 2x1
C is 1x1
D is 1x2
When using the system identifaction App, it automatically assumes all matrices (A,B,C,D) are free parameters to be estimated. How can I force the C and D matrices to be the following?
C = 1;
D = [0 0];

请先登录，再回答此问题。

采纳的回答

Rakshith Badarinath
For anyone in the same situation, here is the solution:
opt = ssestOptions;
opt.EnforceStability = 1;
A = 0; % initial value
B = [0 0];
C = 1;
D = [0 0];
Ts = 0; % Continuous time
init_sys = idss(A,B,C,D,'Ts',Ts);
% Set C and D structures as fixed (use values from intialized state space
% system init_sys
init_sys.Structure.A.Free = true;
init_sys.Structure.B.Free = true;
init_sys.Structure.C.Free = false;
init_sys.Structure.D.Free = false;
init_sys.Structure.K.Free = true;
% estimate model based on the data
constrained_stateSpace_model = ssest(your data,init_sys,opt);

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Linear Model Identification 的更多信息

标签

产品


版本

R2020b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by