Ramon A. Felix

Avoiding controller singularities in adaptive recurrent neural control

Abstract In this paper, to overcome the controller singularity problems, a novel neural parameters adaptive law for on-line identification is proposed, such strategy avoid specific adaptive weights zero-crossing. Using a priori knowledge about the real plant, a recurrent neural network is proposed as identifier. Based on the neural identifier model, a discontinuous control law is derived, which combines Block Control and Sliding Modes. The proposed scheme is tested in a induction motor via simulations.

Neural block control for synchronous generators

In this paper, a novel approach to control a single generator, connected to an infinite bus, is presented. Modifying published results for nonlinear identification using recurrent neural networks, a block controllable neural identifier is proposed, based on this neural model a control law is derived, which combines sliding modes and block control. The neural identifier and the proposed control law allows to reject external disturbances caused by generator terminal short circuits and mechanical power variations. Applicability of the approach is tested via simulations.

Real-Time Direct Field-Oriented Control of induction motor for electric vehicles applications

In this paper, a real-time Direct Field-Oriented Control (DFOC) is proposed for torque and flux control of electric vehicles powered by induction motors. DFOC is one of the most popular control strategies in the industry for induction motor drives and has the advantage to produce a continuous time control signal, helping to extend life of electric vehicle batteries; further, the Space Vector Modulation (SVM) improves torque, flux, and current steady-state performance by reducing the ripple. To estimate the rotor flux, a sliding mode observer is used. The experimental implementation in real-time of this control has been successfully done, used a 1/4 hp induction motor based on a dSPACE board. Results are presented to illustrate the advantages of this control scheme.

Real-time direct field-oriented and second order sliding mode controllers of induction motor for electric vehicles applications

This paper presents a comparison between two control techniques: Direct Field-Oriented Control (DFOC) and Second Order Sliding Mode Control (SOSMC), both on real-time implementations. DFOC is one of the most popular control strategies in the industry for induction motor drives and has the advantage to produce a continuous control signal, helping to extend life of electric vehicle batteries. On the other hand SOSMC is based on the super-twisting algorithm, which reduces chattering, rejects disturbances and produces a continuous control signal. To modulate the inverter pulses, Space Vector Modulation (SVM) is used and to estimate the rotor flux, a sliding mode observer is applied. Real-time implementation of these controllers have been successfully done, using a 1/4 hp induction motor prototype. Results are presented to illustrate the respective advantages and drawbacks. The proposed schemes allow easy integration of these kind of vehicles into a system of systems configuration.

Second-order sliding mode speed controller with anti-windup for BLDC motors

In this paper, a second-order sliding mode speed controller with Anti-windup for Brushless DC (BLDC) motors is proposed. The speed control is developed on the mathematical model based on the well known (d; q) field oriented frame. The super-twisting algorithm is a second order sliding mode controller that allows the rejection of the chattering phenomena while maintaining the robustness of the control. The anti-windup is a technique to handle with the pure-integrator saturation phenomenon, which may lead to instability and low performance in real-time control implementation. By using a Lyapunov function, the closed-loop stability of the system is demonstrated. The control law is implemented on an Digital Signal Processor (DSP) TMS320F2812. Simulations and experimental results are presented and compared to the same control strategy.

Induction motor torque control via discrete-time sliding mode

This paper presents a discrete-time sliding mode controller combined with the block-control technique for induction motors. The sliding mode controller is designed to force the system to track a torque reference and a flux magnitude reference, on the basis of a nonlinear discrete-time model. Then, a reduced order observer is designed for rotor fluxes. Simulations illustrate robustness with respect to external load torque. The proposed control scheme is compared with a direct field oriented controller.

Induction motor torque control via discrete-time neural sliding mode

This paper presents a discrete-time sliding mode control design based on a neural model for induction motors. A Recurrent High Order Neural Network (RHONN) trained with an Extended Kalman Filter (EKF) is used to identify the model. The sliding mode controller is designed to force the system to track a torque reference and a flux magnitude. Then, a reduced order observer is designed for rotor fluxes. The simulations illustrate robustness with respect to external load torque and parameter variations.