On-line identification and identifiability analysis of electrical and mechanical parameters of induction machines
| Autor: | Khanh Duy Trieu, Simon Altmannshofer, Christian Endisch |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Work (thermodynamics)
Signal processing Emphasis (telecommunications) 020206 networking & telecommunications 02 engineering and technology Line (electrical engineering) Inductance Identification (information) Control theory 0202 electrical engineering electronic engineering information engineering Identifiability 020201 artificial intelligence & image processing Mathematics Second derivative |
| Zdroj: | AIM |
| DOI: | 10.1109/aim.2017.8014079 |
| Popis: | Many applications of induction machines require a model and its electrical and mechanical parameters. The parameters can be estimated during the machine's operation in closed-loop by recursive algorithms. On-line estimation algorithms can yield suitable results when parameter identifiability is guaranteed. This work analyzes the identifiability of the induction machine by the condition of sufficient excitation. The result is a reduced model order and a condition on the machine operation to yield useful parameter estimates. The practicality of the presented method is shown on data from a real test bed. Additional emphasis is put on signal processing to compute the necessary first and second derivative of noisy measurements. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|