Predicting Building Energy Consumption Using Machine Learning Methods: a Comparison

The aim of this project is to predict building energy consumption for next days. Firstly, we will collect several data such as weather data, time data, and historical energy consumption of the building. We will analyze the collected data to recognize the usage patterns of the building, for instance, the low and the high electricity consumption of building. After data analysis, we will apply several machine learning models to predict energy consumption of building, and finally all models will be compared to choose the outperform method. This can assist buildings owners and maintenance staff to optimize the operation of building equipment as well as to identify further saving opportunities.

Faculty Supervisor:

Jiacheng Wang

Student:

Azim Keshtkar

Partner:

SES Consulting Inc

Discipline:

Engineering - mechanical

Sector:

Energy

University:

Program: