Nowadays, an engineer’s work consists more and more of obtaining mathematical models of the studied processes. Great part of the literature referring to system identification deals with how to find polynomial models as Prediction Error Methods (PEM) and Instrumental Variable Methods (IVM). In case of complex systems, the state space model appears as an alternative to PEM and IVM models. For multivariable systems, these methods provide reliable state space models directly from input and output data. As systems of large dimensions are usually found in industry, the application of subspace identification algorithms in this field is very promising. Currently the subspaceidentification models Multivariable Output Error State sPace (MOESP) and Numerical algorithms for Subspace State Space System IDentification (N4SID), are topic of study. The objective of this work is to implement th...

Nowadays, an engineer’s work consists more and more of obtaining mathematical models of the studied processes. Great part of the literature referring to system identification deals with how to find polynomial models as Prediction Error Methods (PEM) and Instrumental Variable Methods (IVM). In case of complex systems, the state space model appears as an alternative to PEM and IVM models. For multivariable systems, these methods provide reliable state space models directly from input and output data. As systems of large dimensions are usually found in industry, the application of subspace identification algorithms in this field is very promising. Currently the subspaceidentification models Multivariable Output Error State sPace (MOESP) and Numerical algorithms for Subspace State Space System IDentification (N4SID), are topic of study. The objective of this work is to implement th...