40220782 (Information theory and power system)

Course Name: Information theory and power system

Course Number: 40220782

Program: Undergraduate program

Type: Elective

Credits: 2

Term Offered: Fall

Prerequisite(s): Power system analysis, Probability theory

Instructor(s): Sun Hongbin

Reference(s):

Thomas M. Cover, Joy A. Thomas, Elements of Information Theory, John Wiley & Sons, Inc., 1991

Xuelong Zhu, Fundamentals of applied information theory, Tsinghua University Press

Papers about information theory applied in power systems

Course Description:

The course is about the fundamentals of Shannon information theory with its application in power system. Basic concepts in information theory, such as information, discrete entropy, continuous differential entropy, mutual information, discrimination information and information channel model, are briefly introduced. Decision-making principles based on information theory, such as maximum entropy, minimum discrimination information and minimum information loss, are covered. It then focuses on applications of information theory in power system, such as power system state estimation, fault diagnose, load forecasting, optimal power flow and knowledge discovery.

Course Objectives and Outcomes:

  Numbers in brackets are linked to department educational outcomes.

1.an ability to apply knowledge of probability and information theory in power system [1]

2.an ability to design and conduct simulation experiments, as well as to analyze and interpret data [2]

3.an ability to identify, formulate, and solve power system problems based on information theory[5]

4.a knowledge of information society issue[10]

5.an ability to use the computer for power system practice.[11]

 

Project(s):

Study on minimum information loss based topology error identification.

Study on surrogate constraint optimal power flow

Study on minimum information loss based load forecasting

Study on feature selection based on mutual information

 

Course Topics:

1 Introduction

Information systems applied in power system

Brief view of information theory

History of information theory

2 Basic concepts in information theory

Uncertainty and self information

Shannon discrete entropy

Union entropy and conditional entropy

Discrete mutual information

Continuous differential entropy

Continuous mutual information

Discrimination information

Relationship among discrimination information, mutual information and entropy

3 Information theory based decision-making principle

Information resource and channel model

maximum entropy principle

minimum discrimination information principle

minimum information loss principle

4 Minimum information loss based state estimation for power system

information channel model and information loss for analog measurement

information channel model and information loss for digital measurement

Minimum information loss based analog-digital state estimation

Relationship between Minimum information loss and traditional state estimation

Minimum information loss based topology error identification

5 Minimum information loss based fault diagnose for power system

Uncertainty in fault process

information model and information loss in primary protection

information model and information loss in backup protection

information model and information loss in breaker

minimum information loss based fault diagnose

6 Applications of information theory in load forecasting

Maximum entropy based load forecasting

Minimum information loss based load forecasting

7 Applications of information theory in optimal power flow

Mathematical model of optimal power flow

Information theory based surrogate constraint

Mechanism analysis based on statistics

8 Information theory in power system knowledge discovery

Feature selection based on information theory

Refined rule discovery for power system operation

 

Course Assessment:

 Project report measures, 100 points.