Skip to main content

Discovering consumer lifestyles and behavior changes from electricity consumption: Machine learning approach

Primary supervisor

Hao Wang

Thanks to the widespread deployment of smart meters, high volumes of residential load data have been collected and made available to both consumers and utility companies. Smart meter data open up tremendous opportunities, and various analytical techniques have been developed to analyse smart meter data using machine learning. This project will provide a new angle toward energy data analytics and aims to discover the consumption patterns, lifestyle, and behavioural changes of consumers.

This project covers a wide range of research topics in smart meter data analysis toward a better understanding of (electricity and water) consumption behaviours, thus providing insights into the energy programs and policies. For example, previous thesis students have

  • explored the human factors driving energy consumption behaviours;
  • developed machine learning models to analyse the relationship between users' load patterns and their demographic and socioeconomic factor, promoting energy equity and fairness;
  • studied seasonal load patterns and identified essential factors that determine whether users tend to vary their consumption behaviours when the season changes;
  • developed machine learning models (including deep learning models) to detect household activities and disaggregate energy consumption for different activities.

There remain many research gaps in the area, and we very much look forward to advancing this research area with you.

#sustainability

Student cohort

Double Semester

Aim/outline

This project aims to use machine learning to discover when and how consumers use energy for different household activities.

  1. You will be provided with real-world smart meter data.
  2. You will develop deep learning algorithms to learn consumers' behaviors from real data and then provide insights into consumers' consumption patterns, lifestyle, and behavioral changes.
  3. The outcome of the project include insights into developing fair and effective energy programs and policies.

URLs/references

Some recent papers from the team.

1) Hao Wang, G Henri, CW Tan, R Rajagopal, Activity Detection And Modeling Using Smart Meter Data: Concept And Case Studies, 2020 IEEE Power & Energy Society General Meeting, 2020.

2) Zhuo Wei*, Hao Wang, Characterizing Residential Load Patterns by Household Demographic and Socioeconomic Factors, ACM e-Energy 2021 (The Twelfth ACM International Conference on Future Energy Systems), 2021.

3) Zhenyu Wang*, Hao Wang, Analyzing Seasonal Variation in Residential Load Patterns via Two-Stage Clustering and Relative Entropy: Poster, ACM e-Energy 2021 (The Twelfth ACM International Conference on Future Energy Systems), 2021.

4) Zhenyu Wang*, Hao Wang, Identifying the Relationship between Seasonal Variation in Residential Load and Socioeconomic Characteristics, ACM BuildSys 2021 (The 8th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation), 2021.

5) W Tang, Hao Wang, XL Lee, HT Yang, Machine Learning Approach to Uncovering Residential Energy Consumption Patterns Based on Socioeconomic and Smart Meter Data, Energy, Vol. 240, pp. 1-11, 2022.

6) Cameron Martin*, F. Ke, Hao Wang, Non-Intrusive Load Monitoring for Feeder-Level EV Charging Detection: Sliding Window-based Approaches to Offline and Online Detection, IEEE EI2 2023 (The 7th IEEE Conference on Energy Internet and Energy System Integration), 2023.

7) X. Liang, Hao Wang, Hybrid Transformer-RNN Architecture for Household Occupancy Detection Using Low-Resolution Smart Meter Data, The 49th Annual Conference of the IEEE Industrial Electronics Society (IEEE IECON), 2023.

(*Thesis students.)

Required knowledge

Programming skills in Python or R (required); Data Analysis and/or Machine Learning (preferred).