Back to Search
Start Over
Direct Torque and Predictive Control Strategies in Nine-Phase Electric Drives Using Virtual Voltage Vectors.
- Source :
-
IEEE Transactions on Power Electronics . Dec2019, Vol. 34 Issue 12, p12106-12119. 14p. - Publication Year :
- 2019
-
Abstract
- One of the main distinctive features of multiphase machines is the appearance of new degrees of freedom (${\boldsymbol{x}}$ – ${\boldsymbol{y}}$ voltages/currents) that do not exist in their three-phase counterparts. As a direct consequence, control approaches that apply a single switching state during the sampling period cannot achieve zero average ${\boldsymbol{x}}$ – ${\boldsymbol{y}}$ voltage production. In direct torque control (DTC) this implies that ${\boldsymbol{x}}$ – ${\boldsymbol{y}}$ currents are not regulated, whereas in finite-control-set model predictive control (FCS-MPC) an enhanced ${\boldsymbol{x}}$ – ${\boldsymbol{y}}$ current regulation is feasible only at the expense of disturbing the flux/torque production. Aiming to avoid these shortcomings, this work makes use of the concept of synthetic/virtual voltage vectors (VVs) to nullify/limit the ${\boldsymbol{x}}$ – ${\boldsymbol{y}}$ voltage production in order to improve the current regulation in the secondary planes. Two strategies using two and four virtual voltage vectors (2-VV and 4-VV, respectively) are proposed and compared with the standard case that applies a single switching state. Since standard MPC has the capability to indirectly regulate ${\boldsymbol{x}}$ – ${\boldsymbol{y}}$ currents, the improvements with the inclusion of VVs are expected to be more significant in DTC strategies. Experimental results validate the proposed VVs and confirm the expectations through a detailed performance comparison of standard, 2-VV and 4-VV approaches for DTC and MPC strategies. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 08858993
- Volume :
- 34
- Issue :
- 12
- Database :
- Academic Search Index
- Journal :
- IEEE Transactions on Power Electronics
- Publication Type :
- Academic Journal
- Accession number :
- 138593163
- Full Text :
- https://doi.org/10.1109/TPEL.2019.2907194