Date of Award

12-2025

Document Type

Thesis

Degree Name

Master of Science (MS)

Degree Statement

Master of Science in Electrical Engineering

Department

Department of Electrical and Computer Engineering

Granting Institution

University of Alaska Fairbanks

Committee Chair

Wies, Richard

Committee Chair

Shirazi, Mariko

Committee Member

Roddewig, Michael

Table of Contents

Chapter 1: Introduction.

1.1 Background

1.2 Problem statement

1.3 Goal statement

1.3.1 Objectives

1.4 Research questions

1.5 Scope and limitation

1.6 Literature review

1.7 Thesis organization.

Chapter 2: Synchronous generator modeling.

2.1 Generator circuit model

2.2 Park transformation

2.3 dq-model

2.4 Classical generator machine model

2.5 Estimation model

2.5.1 State vector

2.5.2 State transition model

2.5.3 Measurement model.

Chapter 3: Particle filter.

3.1 Concept: nonlinear system state estimator

3.2 Relative likelihood and resampling of particles

3.3 Parameter settings

3.3.1 Process noise

3.3.2 Error variance

3.3.3 Measurement noise.

Chapter 4: MATLAB Simulink model testing and results.

4.1 MATLAB Simulink benchmark model

4.1.1 Generator parameters

4.1.2 Transformer parameters

4.1.3 Load banks

4.2 Simulink high-fidelity diesel generator model

4.2.1 Estimation using high-resolution PMU data from high-fidelity diesel model

4.2.2 Estimation using low-resolution SCADA data from high-fidelity diesel model

4.2.3 Estimation using 25 Hz data from high-fidelity diesel model as PMU data

4.3 Estimation results using generator test data from ACEP ETF PSI lab

4.4 Overall discussion of results and error analysis.

Chapter 5: Conclusions and future work.

5.1 Conclusions

5.1.1 Simulink benchmark model

5.1.2 High-fidelity Simulink model of ACEP ETF diesel electric generator

5.1.3 ACEP ETF diesel electric generator scope and SCADA data

5.1.4 Limitations

5.1.5 Future work.

Abstract

The accurate parameterization of machines in power systems is crucial during modeling to study the behavior of the modern power system. However, the parameters of synchronous machines are susceptible to change from the original manufacturer’s specifications due to factors like degradation of the windings and core or for rewound machines. Traditional methods to update these parameters require taking generators offline for various testing methods, which results in substantial revenue losses to the utility and sometimes blackouts in small remote microgrids. To overcome this challenge, this study proposes an approach for online parameter estimation by using available data from the generator terminals. A particle filter algorithm has been designed and implemented to estimate the parameters and states of a synchronous generator during steady-state operation. The precise and robust estimation of generator states and parameters is estimated without taking the generator offline. In this work a dynamic state estimation technique known as a particle filter is applied to estimate the states and steady state impedance parameters of synchronous generators. The results show that the particle filter method provides an effective solution for online state and parameter estimation to enhance the operation and control of generation units.

Download

Share

COinS