Igor Erceg

The use of programmable automation controller in synchronous generator excitation system

Synchronous generator voltage control system with a programmable automation controller (PAC) is presented in the paper. Such digital control can be used on both separately excited and self-exciter for generators with slide rings and brushless. As a power converter for a generators excitation an IGBT chopper is used. Function of the synchronous generator voltage control system with a PAC was experimentally verified on a separately excited generator.

Fuzzy Logic Control of Synchronous Generator Under the Condition of Transient Three Phase Short Circuit

This article is focused on the implementation of fuzzy logic excitation control of a synchronous generator. A simple fuzzy logic control scheme for voltage control and generator stabilization is tested on the real laboratory model that includes digital system for excitation control (based on four DSPs) and synchronous generator connected to an AC system through transformer and two parallel transmission lines. The experiments were performed for transient three phase short circuit on the one transmission line. The behaviour of the excitation system with fuzzy logic stabilizing controller is compared with excitation system based on the PI voltage controller and conventional power system stabilizer

Laboratory model for design and verification of synchronous generator excitation control algorithms

This paper presents a laboratory model of synchronous generator excitation system based on National Instruments cRIO real-time industrial controller. It was specially designed for development and verification of classical linear and modern nonlinear excitation control algorithms. Real-time Clarke and Park transformations were implemented on the FPGA for measurements of the generator load angle, voltages and currents in the generator dq-frame. Automatic voltage regulator (AVR) and power system stabilizer (PSS2A) were implemented and experimentally verified on an 83kVA synchronous generator. Tests of step changes for generator voltage and mechanical power references, and test of transmission line disconnection were conducted. Experimental results were compared to simulation results of the designed model in Matlab/Simulink.

Estimation of stator resistance and rotor speed for IPMSG using model reference adaptive system

In order to achieve optimal control of an interior permanent magnet synchronous generator (IPMSG) used in wind power plants, parameters of the generator need to be identified or estimated. Furthermore, since mechanically mounted speed sensors can cause plant failures, sensorless control is preferred and generator rotor speed needs to be estimated. Model reference based methods are one of the most used methods for combined stator resistance and rotor speed estimation in permanent magnet synchronous motors. In this paper an estimation of stator resistance and rotor speed using model reference adaptive system (MRAS) is presented. The MRAS consists of the reference and adjustable model where the reference model is used to determine the required states (d and q current components) and the adaptive model provides the estimated values of the states. Simulations were conducted for different rotor speed values and expected stator resistance variation due to temperature increase under load. Simulation results show that the MRAS model gives satisfactory estimates of rotor speed and stator resistance.

MRAS based estimation of stator resistance and rotor flux linkage of permanent magnet generator considering core losses

For maximization of efficiency of interior permanent magnet generators (IPMSG) and optimization of machine control correct parameters of the machine are needed. Parameters used in advanced control of the machine are stator resistance, magnetic flux linkage and inductances of d and q axes. Alongside them ohmic resistance that represents core losses can be employed to further increase efficiency of control. This work will focus on implementation of MRAS estimator that has adaptive models with and without taking into account core losses. Using Matlab Simulink simulations were conducted and comparison of estimators was made under change of current in d and q axis and different values of resistance that represents core losses. Results show that value of resistance that represents core losses influences estimated values of parameters and that it is beneficial to take into account core losses for efficient control of the machine.

Concepts of synchronous generator's digital control

In this paper new concepts of synchronous generator control using up-to-date digital systems are shown. For excitation and speed (active power) controls of synchronous generator same hardware and software support are used. In this concept the difference between excitation control and speed control is the control algorithm. This new concept of generators control is used on the aggregate in hydro power station Vinodol (Croatia). For realization of this generators control a digital system DIRES 21 is used. The CPU contains 4 digital signal processors ADMC 3000 (analog devices). Functions of the generators control system are experimentally verified.

Comparison of torque estimation methods for interior permanent magnet wind power generator

Control reliability of interior permanent magnet generator (IPMG) depends on accurate knowledge of the generator outputs (torque, currents etc.) and model parameters (resistance, inductances etc.). These outputs and parameters can be obtained by measurements or can be estimated. Reducing measurement equipment, cost of the control system is reduced as well. On the other hand, if control system contains accurate estimation algorithms, control reliability tends to increase. In this paper, comparison of the three torque estimation methods is done. First method considers the IPMG torque estimation using cross product of the IPMG flux and current vectors in the αβ reference frame. Second one estimates the IPMG torque also as a cross product of the IPMG flux and current vectors, but in the dq reference frame. Third method is based on generator electrical power and shaft speed. All of the mentioned methods are experimentally compared to the measured IPMG torque.

Nonlinear Observer-based Control of Synchronous Machine Drive System

Starting from a new dynamic system description novel synchronous machine determin istic observers are proposed. Reduced and full order adaptive observer variations are presented. Based on the feedback linearization control law and the use of deterministic observer a novel control system is built. It meets the requirements of high performance tracking system. Adaptivity to stator and rotor resistance and the torque sensorless application is included. The comparison of theproposed novel control with conventional linear and nonlinear control systems is discussed. The given simulational study includes complete drive system integration.

Design of controller parameters according to the transient indices using the dominant poles method

In literature, the dominant poles method is applied for the determination of controller parameters for the processes with the gain coefficient and dominant time constant equal to unity, and with the given values of relative damping coefficient and undamped (natural) system frequency. The new idea and contribution of this paper is determination of the PI controller parameters according to the given peak overshoot and time of peak overshoot by using the dominant poles method. Region in the s plane, determined by dominant poles, was changed by changed of the peak overshoot and the time of peak overshoot. The dominant poles method is applied for PI controller parameters determination of DC motor drive for different values of peak overshoot and the time of peak response. The procedure for obtaining a desired overshoot response and the most favourable controller parameters is elaborated. Systems responses to the step change of the reference signal and obtained by computer simulation are given in the paper for different values of controller parameters.