Zbigniew Krzeminski

Predictive current control of voltage-source inverters

A new predictive current controller for a voltage-source inverter is presented in this paper. Practical aspects of realizing the new controller in a system with a digital signal processor (DSP) are considered. Delays introduced by measurements are considered and an improved algorithm with one-period prediction of current is presented. The controller was realized in an experimental system with DSP and field-programmable gate array circuits. Results of the simulations and experiments are presented.

Artificial-neural-network-based sensorless nonlinear control of induction motors

In this paper, two architectures of artificial neural networks (ANNs) are developed and used to correct the performance of sensorless nonlinear control of induction motor systems. Feedforward multilayer perception, an Elman recurrent ANN, and a two-layer feedforward ANN is used in the control process. The method is based on the use of ANN to get an appropriate correction for improving the estimated speed. Simulation and experimental results were carried out for the proposed control system. An induction motor fed by voltage source inverter was used in the experimental system. A digital signal processor and field-programmable gate arrays were used to implement the control algorithm.

Space-Vector Pulsewidth Modulation for Three-Level NPC Converter With the Neutral Point Voltage Control

This paper proposes a new space-vector pulsewidth modulation (SV-PWM) strategy for a three-level neutral point clamped inverter. The presented SV-PWM strategy makes it possible to control the neutral point voltage by optimum choice of switch sequence for any position and length of output voltage vector. The proposed solution takes into consideration the unbalance of the dc-link voltages. It also analyzes the influence of vector sequences on the predicted unbalance of the dc-link voltage. The solution allows selecting optimum vectors and their on-time durations in order to reduce quickly the dc-link voltage unbalance. The calculation of the space vector area proposed in this paper takes into consideration voltage unbalance and its influence on the length and position of vectors. The proposed approach assures properly generating the voltage output vectors, even in the case of the existing large voltage unbalance in the dc-link. The results of the experimental investigation of the proposed modulation strategy are presented in this paper.

Speed and Load Torque Observer Application in High-Speed Train Electric Drive

This paper presents an application of induction motor mechanical speed and load torque observers in high-speed train drives. The observers are applied for a 1.2-MW electric drive with an induction motor. The goal of using such observers is to utilize computed variables for diagnostic purposes of speed sensors and torque transmission system. The concept of diagnostic system is presented in this paper, and proper criteria are proposed. The suggested system is designed to work without a speed sensor in the case of existing sensor faults. Monitored motor load torque is used to limit the maximum motor torque in the case of existing problems in the gearbox. The results of simulation and experimental investigations for a 1.2-MW induction motor drive are presented.

Application of Speed and Load Torque Observers in High-Speed Train Drive for Diagnostic Purposes

In this paper, an application of speed and load torque observers is presented. The aim of the observer implementation is to be used in the diagnostic system for a motor speed sensor and a torque transmission system. The diagnostic system is dedicated to the high-speed train propelled by an induction motor. An analysis of the observed variables allows the identification of problems in the mechanical part of the main traction drive. For example, in the case of speed sensor faults, the motor control system could be switched into speed-sensorless control. The proposed observers were verified by simulation and on a test bench for 1.2-MW rail vehicle motor used in high-speed trains.

Observer of induction motor speed based on exact disturbance model

A simplified speed observer was proposed previously by the author to estimate the rotor speed needed in control system of a drive with induction motor. Small error of estimated rotor speed was observed but errors of the stator current and rotor flux vector components appear in dependence of rotor speed. A new structure of the speed observer is proposed now with exact disturbance model and gain coefficients depending on stator frequency. The new structure was derived from the previous one by analysis of dependencies appearing in differential equations. Explanation of observer stability is presented. Properties of the proposed observer are presented using simulations of full control system.

Advanced control of induction motor based on load angle estimation

An advanced control system with load angle adjustment is introduced. The method is based on the action of a phase-locked loop, in which a position synchronization of two vectors to obtain a constant command angle between them is realized. In the system presented, the vectors are stator current and rotor flux. The load angle is kept constant by changing the position of stator current vector as a result of tuning its pulsation. Proportional-integral and fuzzy logic controllers are used to control the load angle. Because of using the load angle controller and simple relations for state variables, the proposed idea does not require exact speed measurement. The discussed control system is not sensitive to motor resistance variations. This idea is realized on a fixed-point digital signal processor and field-programmable gate arrays. Experimental results for the control system fed by a voltage-source inverter and controlled using a predictive current controller are presented.

Speed observer system for advanced sensorless control of induction motor

This paper presents a sensorless control system for induction motors, which is realized on a fixed-point digital signal processor (DSP) and field programmable gate arrays (FPGAs). An observer system has been developed for estimation of speed and the other state variables. The proposed observer system is verified for different conditions of motor operation. Experimental results for the control system fed by voltage source inverter controlled using predictive current controller are presented.

Performance comparison of SiC Schottky diodes and silicon ultra fast recovery diodes

Advanced control systems combined with high speed gate driver circuits enable extremely high rate of change of power devices voltages, up to hundreds of kV/us. Short rise times of power devices could cause significant EMC problems, which are unacceptable in majority of power electronics applications. It is known that voltage variations during diode switch-off depend on how long it takes for the charge stored near the p-n junction to be recovered during voltage reversing. In fast switching applications good forward recovery characteristics are needed. The silicon carbide (SiC) diodes characterize almost zero reverse recovery charge. However the lossless operation in connection with extremely high dv/dt could cause the SiC diodes less effective in damping the voltage ringing. The compromise between high efficiency and low EMI emission is therefore the actual aim of the research. The paper compares the static and dynamic characteristics of ultra fast silicon (Si) and SiC Schottky diodes and presents the study of the mechanism of parasitic high frequency oscillations during turn-off transient.

Properties of sensorless control systems based on multiscalar models of the induction motor

Purpose – To develop general forms of multiscalar models of the induction motor and to present properties of the sensorless control systems based on such models.Design/methodology/approach – Previously presented multiscalar model of the induction motor based on a stator current and rotor flux vector is generalized as a model of type 1. New model of type 2 is defined for stator current and the vector which is directly controlled by a voltage vector. The above models are applied in a sensorless control system with speed observer. Dynamical properties of the sensorless control systems are investigated by simulations and experiments.Findings – Application of the multiscalar model of type 2 results in higher exactness of sensorless control system than application of the multiscalar model of type 1. Controlled variables are more smooth in transients.Research limitations/implications – This is not an analytical proof of stability of the control systems.Practical implications – Provides very useful information fo...