Ahmed Cheriti

Field-oriented control of induction motors using neural-network decouplers

This paper presents a novel approach to the field-oriented control (FOC) of induction motor drives. It discusses the introduction of artificial neural networks (ANNs) for decoupling control of induction motors using FOC principles. Two ANNs are presented for direct and indirect FOC applications. The first performs an estimation of the stator flux for direct field orientation, and the second is trained to map the nonlinear behavior of a rotor-flux decoupling controller. A decoupling controller and flux estimator were implemented upon these ANNs using the MATLAB/SIMULINK neural-network toolbox. The data for training are obtained from a computer simulation of the system and experimental measurements. The methodology used to train the networks with the backpropagation learning process is presented. Simulation results reveal some very interesting features and show that the networks have good potential for use as an alternative to the conventional field-oriented decoupling control of induction motors.

Implementation of a DSP based real-time estimator of induction motors rotor time constant

Implementation of ac drives insensitive to parameter variations is an important need in the field of high performance drives. For drives controlled by the indirect rotor flux oriented control method (IRFOC), the rotor time constant (/spl tau//sub r/ = L/sub r//R/sub r/) exerts a dominant role in the loss of dynamic performance and its variation results in an undesirable coupling between flux and torque of the machine. This paper presents a new scheme for on-line estimation of rotor time constant using dq representation of the model in the stationary reference frame and measurements of accessible motor variables only (voltages, currents and speed). The estimator is tested by simulation in the MATLAB/SIMULINK environment and validated experimentally on a 1/4 hp squirrel cage motor and a 1/4 hp wound rotor motor with implementation on a TMS320C31 digital signal processor.

Power factor improvement of an AC/AC converter by association of PDM control and passive filtering

PDM control and passive filtering are combined to improve, to a significant degree and at lower costs, the power factor of an AC/AC power converter consisting of a single-phase diode bridge rectifier and a series inverter without the smoothing filter of the DC voltage. A low-pass filter absorbs the harmonics of the rectifier input current, whereas PDM control, applied to the inverter, reduces the size of it filters. An optimization routine is used to solve for the position of the pulses or equivalently the holes in the PDM pattern to cancel low order harmonics. The size of the filter decreases with an Increasing number of holes created by PDM control in the current drawn by the converter. Simulation results validate the theoretical results.

A real time energy management for electrical vehicle using combination of rule-based and ECMS

The studies on energy management and optimization are still at an early stage. The energy management of a fuel cell (FC) hybrid vehicle supported with a storing device such as batteries or supercapacitors, is a challenge. Indeed, regardless of the followed route, the available power must be distributed among the various components, in order to minimize hydrogen consumption and increase their life expectancy. This article present a real-time energy management strategy for a fuel cell vehicle equipped with a battery. It is based on the rule-based method combined with the equivalent consumption minimization strategy ECMS. This strategy is developed and simulated by using a dynamic model of the vehicle developed in the Matlab/Simulink environment. Eventually, simulation results are provided to validate the strategy under various vehicle masses.

Application of the current-injected modeling approach to quasi-resonant converters

This paper describes how the current-injected (CI) method, which has been applied only to pulse-width modulation (PWM) DC-DC power converters, can be extended to quasi-resonant (QR) power converters. The methodology for extending this small-signal modeling approach is described in detail. It is also shown that QR dynamic models are easy to obtain since they are derived directly from PWM power converter models. These new models result in a unified block diagram from which zero-voltage-switching (ZVS) or zero-current-switching (ZCS) transfer functions of the basic topologies, such as buck, boost, and buck-boost operated in half-wave (HW) or full-wave (FW) modes, are found. As an application of this method, a ZVS boost power converter and ZCS boost power converter were fabricated and tested. In addition, small-signal models of these power converters were derived with the help of the state-space averaging (SSA) method. The agreement of the CI method simulations with the experimental results for the two QR power converters is comparable or better than that of the SSA method.

LVRT control strategy for PV grid connected system based on P-V optimal slope MPPT technique

The work presented in this paper describes an improvement of a photovoltaic grid connected system based on P-V optimal slope MPPT technique. This system has a LVRT capability with reactive power injection. To achieve this improvement, no new component is added. Only the flexibility of the used MPPT technique is exploited to establish a control law which gives the system a suitable behavior when voltage sags occur. Modeling of generation system under voltage sag is performed and control law is presented. The results are validated by simulation in the Matlab-Simulink environment.

Multi-drive management system to mitigate voltage sags

Multi-drive systems are frequently used in many industry applications such as in paper, textile, and plastic winding process. Adjustable speed drives are very sensitive to power quality problems in particular to voltage sags, short-duration reductions in the rms voltage. This paper presents a multi-drive management strategy to mitigate voltage sags based on the self-kinetic process energy recovery. Simulation and experimental results are shown to demonstrate the efficiency of the proposition.

New adapted forms of P-V optimal slope MPPT for a better grid connected PV system integration

The work presented in this paper describes an improvement of a P-V optimal slope MPPT technique. To achieve this improvement, all the degrees of freedom of the inverter control are exploited. Two forms are then established. Modeling and design of a PV power plant using these improved forms are performed. The results are validated by simulation in the Matlab-Simulink environment.

Application of the current-injected modelling approach to quasi-resonant converters

Until now, the injected current method has been applied only to DC-DC power converters. This paper describes how this method can be extended to quasi-resonant power converters. It is shown that quasi-resonant models are easy to obtain since they are derived directly from the PWM power converter models. These new models result in a unified block diagram from which zero voltage switching or zero current switching transfer functions of the basic topologies such as buck, boost and buck-boost operated in half-wave or full-wave are found. As an application of this method, a zero voltage switching boost power converter and a zero current switching boost power converter are fabricated and tested. Experimental results are compared to computer simulations obtained with this method and with the state-space averaging method.<<ETX>>