BEEE613 

Module Name Big Data Analytics in Electricity Grids (Elective) 
 Module Code BEEE613 
  Level Six 
 Credits 15 

  

  

Module Description and General Aims 

The objective of this module is to impart to students a detailed knowledge of the use of machine learning and data analytics in different applications related to the electricity grid. The module provides a mathematical background, and a description of the steps required to build and evaluate a machine learning system. Information presented in this module also includes: an introduction to different algorithms used in machine learning, a description of typical applications, an overview of software tools commonly used in machine learning and different case studies. Students will complete a project covering the design of a machine learning system to solve a realistic problem related to the electricity grid. 

  

Learning Outcomes 

On successful completion of this Module, students are expected to be able to: 

  1. Explain the need/importance of machine learning and its applications to data analytics in electricity grid. 
    Bloom’s Level 2 
  1. Apply the basic mathematical concepts used in machine learning. 
    Bloom’s Level 3 
  1. Evaluate and compare different machine learning systems according to an application. 
    Bloom’s Level 5 
  1. Apply machine learning algorithms to solve supervised and unsupervised learning problems using a variety of tools.  
    Bloom’s Level 3 
  1. Design machine learning solutions to efficiently manage various electricity gird data.  
    Bloom’s Level 6 

  

Student Assessment 

Assessment Type When assessed Weighting (% of total module marks) Learning Outcomes Assessed 
Assessment 1 Type: Multi-choice test  Students will complete a quiz to demonstrate a detailed knowledge of the applications of machine learning in electricity grid, building and evaluating a machine learning system. Due after Topic 3 15% 1, 2 
Assessment 2 Type: Short answer questions  Students will complete a test with about 10 questions of numerical problems and short answer questions to demonstrate an understanding of the applications of machine learning in electricity grid, building and evaluating a machine learning system, and the algorithms used in machine learning. Due after Topic 6 20% 1, 2, 3 
Assessment 3 Type: Short answer questions / Practical  Students will complete a test requiring the implementation of solutions to solve specific data analytics problems.  Due After Topic 9 30% 2, 3, 4 
Assessment 4 Type: Project Students will complete a project where they have to analyse data and design a machine learning solution to a problem in electricity grid. Final Week 30% All 
Attendance / Tutorial Participation Example: Presentation, discussion, group work, exercises, self-assessment/reflection, case study analysis, application. Continuous 5%   

  

Overall Requirement: An overall final module score of 50% or above must be achieved to pass the module once all assessment has been completed. 

Prescribed and Recommended Readings 

Required textbook(s) 

Rodolfo Bonnin. (2017). Machine Learning for Developers. Packt Publishing. ISBN: 978-1786469878 

Available on Knovel  
https://app.knovel.com/hotlink/toc/id:kpMLD00001/machine-learning-developers/machine-learning-developers 

Stephen Lucci, Danny Kopec. (2016). Artificial Intelligence in the 21st Century (2nd Edition). Mercury Learning and Information. ISBN: 978-1942270003 

Available on Knovel https://app.knovel.com/hotlink/pdf/id:kt010RUSS1/artificial-intelligence/front-matter 

Reference Materials 

References from authentic websites on the Internet. 

https://waikato.github.io/weka-wiki/documentation
https://cran.r-project.org/doc/manuals/r-release/R-intro.html

IEEE Transaction on Smart Grids journals 

  

Module Content 

  

Topic 1 

Introduction to Data Analytics and Data in Electricity Grids  

  1. Basics of data analytics 
  1. Machine Learning vs Artificial Intelligence 
  1. Supervised, Unsupervised, Reinforcement Learning 
  1.  Types of data in electricity grids: electrical and non-electrical data 
  1. Introduction of data analytics in electricity grids: event analytics, state analytics, customer analytics and operational analytics 

Topic 2 

Python in Data Analytics 

  1. Essential libraries of Python for big data analytics: Pandas, Numpy, Matplotlib, Scikit-learn, Statsmodel, Tensorflow 
  1. Examples of using the essential libraries to analyse electricity grids data 

Topic 3 

Data Flow and Feature Engineering  

  1. Data sources in electricity grids 
  1. Data pre-processing  
  1. Data visualization 
  1. Features and feature vectors 
  1. Dimensionality reduction 
  1. Data mining 
  1. Application of feature engineering to big datasets in electricity grids 

Topic 4 

Commonly Used Supervised Learning Algorithms  

  1. K-Nearest Neighbours 
  1. Decision trees 
  1. Linear regression 
  1. Naïve Bayes 
  1. Implementation of classification algorithms in electricity grids 
  1. Implementation of regression algorithms in electricity grids 

Topic 5 

Unsupervised Machine Learning and Reinforcement Learning Algorithms  

  1. Clustering: K-means, DBENGAN, Hierarchal clustering  
  1. Association: Apriori and other algorithms  
  1. Learning models of reinforcement: Markov decision process and Q learning  
  1. Implementation of clustering algorithms using electricity grids data 
  1. Implementation of association rules using electricity grids data 
  1. Example of reinforcement learning in smart grids, e.g., energy trading 

Topic 6 

Other Algorithms (III) 

  1. Genetic algorithms and other optimization algorithms 
  1. Artificial Neural Networks:  
  • Feedforward Neural Networks 
  • Convolutional Neural Networks 
  • Recurrent Neural Networks 

3. Artificial neural networks in electricity demand forecasting 

Topic 7 

Introduction of other Tools in Data Analytics 

  1. R 
  1. MATLAB 
  1. WEKA  
  1. Could-based Solutions 
  1. Examples of using R and other tools in analysing electricity grids data 

Topic 8 

Event Analytics in Smart Grids 

  1. Fault and failure diagnosis  
  1. Fraud detection 
  1. Predictive outage management 
  1. Statistical Process Control for event/anomaly detection 

Topic 9 

State Analytics in Smart Grids 

  1. Equipment health and condition monitoring  
  1. System and state identification 
  1. Predictive control 
  1. Stability enhancement 
  1. Asset management 

Topic 10 

Customer Analytics in Smart Grids 

  1. Demand response 
  1. Consumer modelling and segmentation 
  1. Customer behaviours 
  1. Peer-to-peer energy trading  

Topic 11 

Operational Analytics in Smart Grids 

  1. Generation and load forecasting 
  1. Real-time energy management 
  1. Price forecasting  
  1. Economic load dispatch 
  1. Network constraint forecasting & dynamic network operating envelopes 
  1. Risk analysis 

Topic 12  

Module Review 

  

Software/Hardware Used 

Software 

  • WEKA 
  • R 
  • MATLAB 
  • Python 

Hardware 

  • N/A