Purpose Direct Torque Control (DTC) of induction motor drives is a well-established technique owing to features such as fast dynamic and insensibility to motor parameters. However, conventional DTC scheme, based on comparators and the switching table, suffers from large torque and flux ripples. To improve DTC performance, this study aims to propose and implement a sensorless finite-state predictive torque control using extended Kalman Filter in dSPACE environment. Design/methodology/approach This paper deals with the design of an extended Kalman filter for estimating the state of an induction motor model and for sensorless control of systems using this type of motor as an actuator. A complex-valued model is adopted that simultaneously allows a simpler observability analysis of the system and a more effective state estimation. Findings Simulation and experimental results reveal that the drive system, associated with this technique, can effectively reduce flux and torque ripples with better dynamic and steady state performance. Further, the proposed approach maintains a constant switching frequency. Originality/value The proposed speed observer have been developed and implemented experimentally under different operating conditions such as parameter variation, no-load/load disturbances and speed variations in different speed operation regions.

No abstract

This paper presents the use of the predictive strategy concept to improve the Active Power Filter (APF) performance, by compensation of the reactive power and elimination of the harmonic currents generated by non-linear loads. Predictive control is generating considerable interest when it comes to implementing current control strategies in active power filter. The proposed strategy provides a simple controller incorporating Phase Locked Loop (PLL) independency. The prediction is evaluated using a cost function that quantifies the desired system behavior. The cost function used in this work evaluates the filtered error of the currents. This strategy minimized the number of sensors, ease of practical implementation and reduced system size and cost. The effectiveness of the proposed controller is confirmed through simulation and experimental validation using a hardware prototype based on dSPACE-1104

The present paper proposes a detection method of the broken rotor bar fault in an induction motor at low-speed operation. The diagnosis method is based at the first place, on Hilbert Transform (HT) that is used to extract the stator current envelope; then on Discrete Wavelet Transform (DWT) which processes the previously produced signal. As such, the calculation of the stored energy on envelope levels allows to determine the severity of the fault. In this work, the induction motor is controlled at the very low-speed range and rated load via using sensorless Backstepping control. This nonlinear control is executed to preserve a satisfactory performance speed control during the presence of broken rotor bars to ensure operational continuity. Moreover, Model Reference Adaptive System (MRAS) is used for speed reconstruction to improve the control's system reliability and to reduce its cost. Considerably, through the use of simulation and real-time implementation using MATLAB/Simulink with the dSpace 1104 control board, the effectiveness of the diagnosis and control techniques is evaluated.

No abstract