基于可再生能源和负荷不确定性的微电网能源管理系统研究

摘要

A Microgrid is an active distribution system consist of distributed energy resources(DER),energy storage units(ESU)and controllable loads(CLs).It is colmected to the main utility grid through the point of common coupling(POC).The main objective of this research study is to provide forecast and economic dispatch related solutions for grid connected and isolated microgrid in order to control and optimize the flow of energy from microgrid to the main utility grid.A practical test bed system is considered to carried out test consist of a PMSG type Wind turbine,Solar Panels,Vanadium redox flow battery,Lithium-Ion battery,Microturbine unit and Diesel generator with a power rating of2.5MW,450kW,200kW,150kW,130kW,500kW respectively.A critical load of2MW and a non-critical load of500kW are considered in this study.A critical load is supposed to receive power in both modes of operations of microgrid while the non-crRical load can be curtail based on the availability of energy.
　　Robust Fuzzy prediction interval modeling is used for generation forecasting of the wind and solar resources.A dilute,distributed and intermitted nature of power sources create a huge challenge in front of scientists to overcome,the forecast of these sources even get harder with constraint in the isolated mode of microgrid operation,where a little error creates a large instability in demand and supply of energy.The overall problem is analyzed from the basics and modifications are made in existing algorithms to create more intelligent approach.The cost function is minimized based on hourly and minute based predictions.The minimum error is obtained based on real power values from the site via comparison through different benchmarks.
　　In the second part of this paper,swarm optimization is used to address economic dispatch problem in grid connected microgrid.Matlab and Python languages are used to solve total cost function of microgrid including investment,fuel,and operation,maintenance,and replacement costs of different components of the microgrid.The main objective of this part of the study is to minimize the transaction cost function while satisfying the load demand of the site under uncertain behavior of renewable generation and load.The minimum total cost function is compared to the various situation through scenario based analysis in order to check under different load and generation uncertainties.
　　At last,the total cost function is minimized and integrated into forecasting module which then works together to provide the solution of economic dispatch problem.The proposed method is valid for grid connected and isolated microgrids.

目录

声明

List Of Acronyms

Abstract

Table of Contents

List of Figures

List of Tables

Chapter 1 Introduction

1．1 Microgrid Architecture

1．2 Types of Microgrid Control

1．3 Advantages of Microgrid

1．4 Challenges in Development

1．5 Dissertation layout

Chapter 2 Literature

2．1 Smart Energy Management

2．2 Forecasting Techniques

2．2．1 Generation Forecast

2．2．2 Load Forecast

2．3 Economic Dispatch and Unit Commitment

2．4 Research Statement

2．4．1 Plan for research study

2．5 Summary

Chapter 3 Forecast

3．1 Introduction

3．2 Prediction formulation for Microgrid

3．2．1 Takagi and Sugeno Model

3．2．2 Coverage Grade

3．2．3 Particle Swarm Optimization for FPI

3．2．4 Training Inputs for FPI

3．3 Summary

Chapter 4 Energy Management System

4．1 Introduction

4．2 Economic Dispatch

4．2．1 Cost Functions and Constraints

4．2．2 Scenario based EMS

4．2．3 FUzzy Battery Model

4．2．4 EMS project PT2016-GS01

4．3 Summary

Chapter 5 Results and Discussion

5．1 Simulation Results

5．1．1 Wind Forecast

5．1．2 Solar Forecast

5．1．3 Load Forecast

5．1．4 EMS and test cases

5．2 Summary

Chapter 6 Conclusion and Future Direction

6．1 Summary

6．2 Contribution

6．3 Future Work

Appendix

References

Publications

Acknowledgement

Biography