This dissertation has investigated both the “framework” level and certain “dedicated control techniques” for grid-connected two- and three-level neutral point clamped back-to-back power converters and permanent magnet synchronous generator wind turbine systems. Deeply investigated “framework” level control techniques include the classical vector oriented control, direct control with both modulator and switching table, deadbeat control and the modern predictive control classes, with computationally efficient and performance enhanced direct model predictive control techniques as the major emphasized ones. The closely investigated “dedicated control techniques” presented in this work include performance enhanced DC-link control, grid side voltage sensorless control with fast state estimation, generator side encoderless control using fundamental model, etc. The real-time verification of the electric control unit, particularly its functionality at abnormal situations (e.g., the low-voltage ride through (LVRT) capabilities of the underlying LVRT control unit), for grid-tied back-to-back power converter wind turbine systems, through the signal level hardware-in-the-loop technique, has also been closely discussed and investigated. The performances of the aforementioned control techniques are experimentally assessed at three self-designed test-benches. Digital controller realization via field programmable gate array of the aforementioned techniques has been comprehensively studied as well.
«
This dissertation has investigated both the “framework” level and certain “dedicated control techniques” for grid-connected two- and three-level neutral point clamped back-to-back power converters and permanent magnet synchronous generator wind turbine systems. Deeply investigated “framework” level control techniques include the classical vector oriented control, direct control with both modulator and switching table, deadbeat control and the modern predictive control classes, with computationa...
»