Electronic Thesis and Dissertation Repository

Thesis Format

Monograph

Degree

Master of Engineering Science

Program

Electrical and Computer Engineering

Collaborative Specialization

Artificial Intelligence

Supervisor

Grolinger, Katarina

Abstract

Smart meter popularity has resulted in the ability to collect big energy data and has created opportunities for large-scale energy forecasting. Machine Learning (ML) techniques commonly used for forecasting, such as neural networks, involve computationally intensive training typically with data from a single building/group to predict future consumption for that same building/group. With hundreds of thousands of smart meters, it becomes impractical or even infeasible to individually train a model for each meter. Consequently, this paper proposes Cluster-Based Chained Transfer Learning (CBCTL), an approach for building neural network-based models for many meters by taking advantage of already trained models through transfer learning. CBCTL first clusters the meters based on their load profiles. Next, Similarity-Based Chained Transfer Learning (SBCTL) is applied within each cluster; the first model within each cluster is trained in a traditional way and all other models transfer knowledge from existing models in a chain-like manner according to similarities between energy consumption profiles. A Recurrent Neural Network (RNN) was used as the base forecasting model, two initialization techniques were considered, and different similarity measures were explored. The experiments show that CBCTL and SBCTL achieve accuracy comparable to traditional ML training while taking only a fraction of time.

Summary for Lay Audience

Smart meter popularity has resulted in the ability to collect big energy data and has created opportunities for large-scale energy forecasting. The machine learning techniques commonly used for energy forecasting involve computationally intensive training typically with data from a single building/group to predict future consumption for that same building/group. With hundreds of thousands of smart meters, it becomes impractical or even infeasible to individually train a model for each meter. Consequently, this paper proposes Cluster-Based Chained Transfer Learning (CBCTL), an approach for building neural network-based forecasting models for many meters by taking advantage of already trained models through transfer learning. CBCTL first groups the meters based on their energy usage patterns. Next, Similarity-based Chained Transfer Learning (SBCTL) is applied within each cluster of meters; the first model within each cluster is trained in a traditional way and all other models transfer knowledge from existing models in a chain-like manner according to similarities between energy consumption patterns. A Recurrent Neural Network (RNN) was used as the base forecasting model, two initialization techniques were considered, and different similarity measures were explored. The experiments show that CBCTL and SBCTL achieve accuracy comparable to traditional ML training while taking only a fraction of time.

Download

Share

COinS