SwePub - sökning: id:"swepub:oai:DiVA.org:kth-17...

Numrering	Referens	Omslagsbild	Hitta
1.	Galrinho, Miguel, et al. (författare) On Estimating Initial Conditions in Unstructured Models 2015 Ingår i: 2015 54th IEEE Conference on Decision and Control (CDC). - : IEEE conference proceedings. - 9781479978847 ; , s. 2725-2730 Konferensbidrag (refereegranskat)abstract Estimation of structured models is an importantproblem in system identification. Some methods, as an intermediatestep to obtain the model of interest, estimate theimpulse response parameters of the system. This approachdates back to the beginning of subspace identification and isstill used in recently proposed methods. A limitation of thisprocedure is that, when obtaining these parameters from ahigh-order unstructured model, the initial conditions of thesystem are typically unknown, which imposes a truncation ofthe measured output data for the estimation. For finite samplesizes, discarding part of the data limits the performance ofthe method. To deal with this issue, we propose an approachthat uses all the available data, and estimates also the initialconditions of the system. Then, as examples, we show how thisapproach can be applied to two methods in a beneficial manner.Finally, we use a simulation study to exemplify the potential ofthe approach.

Galrinho, Miguel, et al. (författare)
On Estimating Initial Conditions in Unstructured Models
2015
Ingår i: 2015 54th IEEE Conference on Decision and Control (CDC). - : IEEE conference proceedings. - 9781479978847 ; , s. 2725-2730
Konferensbidrag (refereegranskat)abstract
- Estimation of structured models is an importantproblem in system identification. Some methods, as an intermediatestep to obtain the model of interest, estimate theimpulse response parameters of the system. This approachdates back to the beginning of subspace identification and isstill used in recently proposed methods. A limitation of thisprocedure is that, when obtaining these parameters from ahigh-order unstructured model, the initial conditions of thesystem are typically unknown, which imposes a truncation ofthe measured output data for the estimation. For finite samplesizes, discarding part of the data limits the performance ofthe method. To deal with this issue, we propose an approachthat uses all the available data, and estimates also the initialconditions of the system. Then, as examples, we show how thisapproach can be applied to two methods in a beneficial manner.Finally, we use a simulation study to exemplify the potential ofthe approach.

Träfflista för sökning "id:"swepub:oai:DiVA.org:kth-179576" "