Matlab: Concatenating single-channel models to a multi channel model with System Identification Toolbox -
i have following problem: have simo system , want apply subspace identification algorithm (moesp in specific case) using matlab function n4sid (which performs subspace identification of state space models) of system identification toolbox. due large amount of data , outputs not possible feed algorithm data @ once more gradually more suitable.
consider have 2 outputs 1 single input: perform subspace identification of 2 siso systems independently , concatenate them obtain final , complete model having same order of previous models: possible default system identification toolbox of matlab? have reference? have asked on matlab forums no answer plus read user's guide of system identification toolbox.
// example: in = excitation signal. out1, out2 = output signals // ts = sampling interval signals 0 mean // putting data iddata format dat1 = iddata(out1,in,ts); dat2 = iddata(out2,in,ts); //considering model order of 10 m1 = n4sid(dat1,10); // m1 = siso model output out1 , input in m2 = n4sid(dat2,10); // m2 = siso model output out2 , input in // estimating model 2 outputs , 1 input: dat = iddata([out1 out2], in,ts); m = n4sid(dat,10); // m = simo model outputs out1 , out2 , input in
looking @ example: how m m1 , m2 without performing estimation again?
i have related question: subspace algorithms based on l-q decomposition of matrix. suppose have 2 matrices decomposed example a1, a2. there way in matlab decomposition of matrix obtained concatenating a1 a2
a =[a1, a2]
from single decompositions?(especially interested in l matrix) in advance
Comments
Post a Comment