Using Google Trends as a proxy for occupant behavior to predict building energy consumption

Objectives:

In research related to building performance simulation or prediction, how to quantify the occupant behaviors in different building types, countries and regions is a very difficult topic. Google Trends can provide search volume of various keywords in specific country/zone, and can probably infer the occupant behavior from these time-series trends.

Dataset:

Google Trends data:

• Daily trend data
• Download-Google-trends-data.ipynb

Building Data Genome Project 2 (BDG2):

• BDG2 Dataset

ASHRAE - Great Energy Predictor III (GEPIII):

• GEPIII Dataset

Modeling:

• Baseline model (a public notebook on kaggle)
• Proposed method with google trends

Analysis:

Derive occupant behavior from the time series of electricity meters via PCA, and caculate correlations between meter readings (daily values) and google trends:

• Correlation-between-energy-and-topics.ipynb

Vizualization and analysis of results:

• Result.ipynb