Practical Implementation of Sensorless Torque and Reactive Power Control of Doubly Fed Machines

Comparisons of Vector Control Algorithms for Doubly-Fed Reluctance Wind Generators

“…Various control algorithms 2 have been developed for the BDFRG including scalar control [3,22], primary voltage vector-oriented control (VOC) [3,13,14,22], primary flux (field)-oriented control (FOC) [13,14], direct torque and flux control [22,23], direct torque and reactive power control [24,25], direct power control [26], sliding mode power control [27], and even non-linear Lyapunov control theory [8]. Although a comparative qualitative analysis of some of these control methods has been partly made in [22] (and more detailed for the DFIG in [15,21]), no similar quantitative study, other than that in [22], has been reported specifically on FOC versus VOC of large BDFRG wind turbines, and there has been very little theoretical or applied work published on true FOC for this machine in general.…”

A New Sensorless Speed Control Scheme for Doubly Fed Reluctance Generators

“…Similar studies to those conducted for the BDFIG [24]- [26] or DFIG [6] have also been done on the BDFRG(M) involving: scalar control [2], [23], vector control (VC) [1], [2], [5], [23], [27]- [29], direct torque and secondary flux (λ s ) [23], [30] or primary reactive power (Q) control (DTC) [31], direct power control (DPC) [32] and variable structure control [33]. The preliminary attributes in [2], [23], [33] are intellectually interesting but have been left unproven in practice.…”

“…However, the DTC methodologies in [23], [30] are extremely sensitive to inductance knowledge and λ s estimation inaccuracies so that poor proof of concept results for an unloaded BDFRM have just been reported. These shortcomings have been eliminated and much better response provided by replacing λ s with Q as a control variable in the improved parameter independent DT(P)C schemes [31], [32] albeit at fixed BDFRG(M) loads of no or little interest to the target applications. Although robust and relatively easy to implement in a stator frame without having to know the rotor position or speed, the hysteresis torque (power) controllers in [31], [32] suffer from usual variable switching frequencies and higher flux (torque) ripples, unlike, in this sense, the undoubtedly superior VC.…”

“…These shortcomings have been eliminated and much better response provided by replacing λ s with Q as a control variable in the improved parameter independent DT(P)C schemes [31], [32] albeit at fixed BDFRG(M) loads of no or little interest to the target applications. Although robust and relatively easy to implement in a stator frame without having to know the rotor position or speed, the hysteresis torque (power) controllers in [31], [32] suffer from usual variable switching frequencies and higher flux (torque) ripples, unlike, in this sense, the undoubtedly superior VC. Besides, an encoder is solely required for speed regulation in [31], [32], and its use is under-utilized from this point of view compared to the VC where it additionally serves for torque control.…”

“…Although robust and relatively easy to implement in a stator frame without having to know the rotor position or speed, the hysteresis torque (power) controllers in [31], [32] suffer from usual variable switching frequencies and higher flux (torque) ripples, unlike, in this sense, the undoubtedly superior VC. Besides, an encoder is solely required for speed regulation in [31], [32], and its use is under-utilized from this point of view compared to the VC where it additionally serves for torque control.…”