Liu, Tao (2021)
Parameter Identification of PMSM with Considering
Nonlinearity of the Inverter.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00019740
Dissertation, Erstveröffentlichung, Verlagsversion
Kurzbeschreibung (Abstract)
Electrical parameters of a Permanent Magnet Synchronous Machine (PMSM) vary with the load condition, magnetic saturation, and temperature. When the parameters change greatly from their initial values, the performance of the PMSM drive system can be significantly degraded or even failed. Because of this phenomenon, precise on-line parameter identification of PMSMs is essential to ensure a high-performance drive system. However, the accuracy of the parameter identification methods is affected by the nonlinearity of the inverter, because the inverter nonlinearity results in error between the actual output voltages measured at the terminals of the machine and the reference values. The main objective of this research work is to develop suitable online identification algorithms adopted for machine parameters, where the inverter nonlinearity effect is analyzed and compensated. The inverter nonlinearity curves dependent on stator currents are measured offline and then used as a look-up table for the online identification process. The proposed identification algorithms based upon recursive least squares method (RLS) and model reference adaptive control (MRAC) are presented and compared. Experimental results show that both the methods ensure fast convergence and can be implemented in required real-time performance systems. Based on the steady-state equations in the d-and q-axis, the algorithms are capable of simultaneously estimating stator resistance and inductance. For the implementation of the proposed algorithms, an AC drive system consisting of control and power boards is designed. On the control board, a structure of DSP (digital signal processor)+FPGA (field-programmable gate array) is adopted, in which the DSP conducts the control algorithm while the FPGA undertakes tasks involving signal processing, including current/voltage sensing and encoding of rotor position. On the power board, a conventional two-level voltage source inverter is integrated, where the space vector pulse-width modulation (SVPWM) is applied for generating the switching signals. The proposed algorithms are implemented on the designed drive system, which is verified to be effective and reliable via experimental results.
Typ des Eintrags: | Dissertation | ||||
---|---|---|---|---|---|
Erschienen: | 2021 | ||||
Autor(en): | Liu, Tao | ||||
Art des Eintrags: | Erstveröffentlichung | ||||
Titel: | Parameter Identification of PMSM with Considering Nonlinearity of the Inverter | ||||
Sprache: | Englisch | ||||
Referenten: | Griepentrog, Prof. Dr. Gerd ; Leidhold, Prof. Dr. Roberto | ||||
Publikationsjahr: | 2021 | ||||
Ort: | Darmstadt | ||||
Kollation: | XIX, 148 Seiten | ||||
Datum der mündlichen Prüfung: | 28 Juni 2021 | ||||
DOI: | 10.26083/tuprints-00019740 | ||||
URL / URN: | https://tuprints.ulb.tu-darmstadt.de/19740 | ||||
Kurzbeschreibung (Abstract): | Electrical parameters of a Permanent Magnet Synchronous Machine (PMSM) vary with the load condition, magnetic saturation, and temperature. When the parameters change greatly from their initial values, the performance of the PMSM drive system can be significantly degraded or even failed. Because of this phenomenon, precise on-line parameter identification of PMSMs is essential to ensure a high-performance drive system. However, the accuracy of the parameter identification methods is affected by the nonlinearity of the inverter, because the inverter nonlinearity results in error between the actual output voltages measured at the terminals of the machine and the reference values. The main objective of this research work is to develop suitable online identification algorithms adopted for machine parameters, where the inverter nonlinearity effect is analyzed and compensated. The inverter nonlinearity curves dependent on stator currents are measured offline and then used as a look-up table for the online identification process. The proposed identification algorithms based upon recursive least squares method (RLS) and model reference adaptive control (MRAC) are presented and compared. Experimental results show that both the methods ensure fast convergence and can be implemented in required real-time performance systems. Based on the steady-state equations in the d-and q-axis, the algorithms are capable of simultaneously estimating stator resistance and inductance. For the implementation of the proposed algorithms, an AC drive system consisting of control and power boards is designed. On the control board, a structure of DSP (digital signal processor)+FPGA (field-programmable gate array) is adopted, in which the DSP conducts the control algorithm while the FPGA undertakes tasks involving signal processing, including current/voltage sensing and encoding of rotor position. On the power board, a conventional two-level voltage source inverter is integrated, where the space vector pulse-width modulation (SVPWM) is applied for generating the switching signals. The proposed algorithms are implemented on the designed drive system, which is verified to be effective and reliable via experimental results. |
||||
Alternatives oder übersetztes Abstract: |
|
||||
Status: | Verlagsversion | ||||
URN: | urn:nbn:de:tuda-tuprints-197405 | ||||
Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften und Maschinenbau | ||||
Fachbereich(e)/-gebiet(e): | 18 Fachbereich Elektrotechnik und Informationstechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Stromrichtertechnik und Antriebsregelung |
||||
Hinterlegungsdatum: | 21 Dez 2021 07:59 | ||||
Letzte Änderung: | 22 Dez 2021 11:05 | ||||
PPN: | |||||
Referenten: | Griepentrog, Prof. Dr. Gerd ; Leidhold, Prof. Dr. Roberto | ||||
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: | 28 Juni 2021 | ||||
Export: | |||||
Suche nach Titel in: | TUfind oder in Google |
Frage zum Eintrag |
Optionen (nur für Redakteure)
Redaktionelle Details anzeigen |