Solar City Project: Factors Influencing Energy Consumption and CO2 Emissions

The Solar City Project by University of Ballarat evaluates the drivers of power consumption and CO2 emissions in Loddon Mallee and Grampians regions. Factors such as adoption of solar energy technologies, geographical and physical characteristics of dwellings were assessed to evaluate power consumption. The project found that dark and intermediate colour roofs, older constructions, and higher PV capacity contribute to lower power usage. On the other hand, houses with more lamps, tinted windows with curtains, and dark coloured roofs made of timber and concrete blocks utilize more power. The project recommends using higher PV capacity, windows with double glazed blinds, older constructions, and higher insulation capacity to PV to save power.

Solar City Project
BUSINESS INTELLIGENCE AND BIG DATA
ANALYTICS

Introduction
• The “Solar city project” is an initiative of “University of Ballarat.”
• The project focuses on factors which influence Energy Consumption in the region of “Loddon Mallee” and
“Grampians.”
• The objective of the project was to evaluate the drivers of power consumption and thus the drivers of CO2
• The project wished to unravel features in household which contribute to power consumption
• The feature which are assessed to evaluate power consumption are
 Adoption of Solar Energy Technologies
 Geographical Characteristics
 Physical Characteristics of dwellings

Dark and Intermediate colour roofs contribute to low power usage as compared to Light and Unknown Roofs

Houses with estimated age of Fifteen to nineteen have the highest average power usage.
Houses with estimated age of forty to forty-nine have the lowest average power usage.

The highest average power usage is in the month of July while the lowest average power usage is in December

Heywood has the highest number of houses with PV_Capacity. Tyrendarra has the least number of houses with PV_Capacity.

Houses with estimated age of Thirty to Thirty-nine have the highest number of installed PV_Capacity.

Average Power usage of Other-Tenure houses is the highest. Average power usage of rent-free houses is the least

Average power-usage of Casterton is the highest, while for Portland is the least. Power Usage varies across Suburbs

Average Power Usage increases with number of Lamps in a house.

Power Usage changes with number of Bedrooms, Suburb and Estimated Age of the house. Houses with over sixty and
over from the suburbs of Casterton and Heathmeare having more than 3 bedrooms use the maximum power.

Dark coloured roofs made of Timber and Concrete Blocks utilise the highest power

Tinted windows having Curtains have the maximum power usage. Double Glazed windows with Blinds have the least
power usage.

Inference
Factors influencing Usage of Higher Power
Dark Coloured Roofs
Low PV Capacity
In the Month of July
Others use higher power
Single Storied Houses Use more Power
Dark Coloured Roofs made of Timber Use
More Power
Factors influencing Usage of Lower Power
Light coloured Roofs
High PV Capacity
In the Month of December
Rent Free Houses use lower power
Double Storied Houses Use more Power
Double Glazed blinds

Recommendation
• Higher PV Capacity is better in the long Run. They provide more solar power thus
consuming less power from the Grid.
• Windows with Double glazed Blinds should be used to save power
• Older Constructions are better since they consume less power
• Higher Insulation capacity to PV should be used to get more Solar Power
• More people should live in Rent Free houses

Reflection
It was a new kind of experience to learn BI using a cloud based Analytics tool. The natural
language processing feature of IBM Watson Analytics was really easy to use. Some of the
questions were though not very direct. We could use the Analytics tool very easily.
The analytical tool helped us to create new columns thorough calculations. We could
change shapes at some of the charts to provide a better visualisation.
The colours of the charts could be changed to provide an impact.
We could change the visualisation after the chart was made and thus choose which ever we
thought would be better at providing the information.