New direct torque neuro-fuzzy control based SVM-three level inverter-fed induction motor

Abstract

In this paper, a novel direct torque neuro-fuzzy control (DTCNF) scheme combining with space voltage modulation (SVM) technique of a three levels inverter is presented. Using neuro-fuzzy technique, the reference space voltage vector can be obtained dynamically in terms of torque error, stator flux error and the angle of stator flux. Compared with conventional direct torque control (C_DTC), in this new technique, the ripples of both torque and flux are reduced remarkable, and switching frequency is maintained constant. Simulation results verify the validity of the proposed method.

Review of recent advancements of direct torque control in induction motor drives – a decade of progress

“…Intelligent controllers such as fuzzy, artificial neural networks (ANN) and neuro-fuzzy are applied to DTC to extract better performance. Like other previously discussed schemes, torque ripple reduction is the main objective of using intelligent controllers [17,47,[53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69].…”

“…Torque ripple and flux ripple reductions of 73.68 and 38.66% were observed compared with conventional DTC (Table 8) by using this algorithm. Similar optimised voltage vector sector scheme has been given using SVM for three-level inverter [67]. Like fuzzy controllers [58], NFCs were also used to tune torque hysteresis band online [68].…”

Fuzzy logic speed controller optimization approach for induction motor drive using backtracking search algorithm

“…It is seen that the settling time of the BSAF controller is 0.055 sec which is very small compare to other controller as shown in Table 5. [3] PID controller vector control DSP-TMS320F2812 0.85 [8] FLC-PI controller vector control Matlab/Simulink 1.8 [14] ANFIS controller Direct torque control Matlab/Simulink 0.2 [35] PID-FLC controller scalar control Matlab/Simulink 0.6 [36] GA-ANFIS controller scalar control DSP-dsPIC30F6010 0.7 [37] PI controller vector control Matlab/Simulink 0.5 [38] FLC controller vector control Matlab/Simulink 2 [39] sliding-mode controller vector control Matlab/Simulink 0.3 [40] PI …”

“…In addition, ANN is utilized to identify TIM parameters [13]. Adaptive neuro-fuzzy inference systems are merged between ANN with fuzzy logic controller (FLC) and applied to a new vector control in TIM [14]. In [15], ANN is used to estimate rotor speed of a DC motor for closed-loop control.…”

“…Each output signal cycle is divided into six sections in the shape of a hexagon(Fig. 5b)[7,14]. The reference voltage and the in each sector are products of two adjacent non-zero vectors and zero vectors, respectively.…”