Connect with us

Hi, what are you looking for?

Tech & Science

Digital imaging helps to map a city’s energy usage

Creating an energy usage map for a major city, with a population of millions and a complex infrastructure, is very difficult. A new digital solution is here.

London, as seen from The Shard. Image by Tim Sandle.
London, as seen from The Shard. Image by Tim Sandle.

For many city planners and businesses the amount of energy used by a business of civic amenity is of economic importance, as well as having a considerable environmental impact. While some organizations may understand and be able to map their own premises, what does a modern city look like?

Providing an energy usage map for a major city, with a population of millions and a complex infrastructure, has proved very difficult to construct. A new research project has provided the means to make this task simpler and with it the tools to address cost and pollution matters.

The new model comes from the University of Pittsburgh, as a result of a problem statement designed to reduce building emissions and acknowledging that in the typical U.S. city, the building sector accounts for around 40 percent of the energy use. The problem is that detailed analysis on a city-wide scale has been lacking.

The new model to achieve the required in-depth analysis is called the ‘Urban Building Energy Model’ and it  uses street-level images to categorize and estimate commercial buildings’ energy use.

This is based on image processing, where the researchers used publicly available Geographic Information System (GIS) data and street-level images to develop their UBEM. From this, they next created 20 archetypes of buildings that comprised eight commercial use types.

The buildings were next sorted into the groups based on categories including use type and construction period. For each category, street-level images were processed to determine the building material, window-to-wall ratio, and number of floors, and LiDAR data (a light based type of RADAR) was used to determine building height.

Testing this out, it was possible to simulate and map the annual energy use intensity of many structures in Pittsburgh, with just a 7 percent error rate.

Building on the model, the researchers hope to introduce machine learning to rapidly analyze and categorize building images. This will enable data to be used to enable cities to reach their energy goals and efficiency regulations.

This is part of the process that many countries are realizing. To meet or exceed their climate and energy goals aggressive action and solid planning is needed. To make this process easier and to boost accuracy, specialist, digital models are required.

The research appears in the journal Energy and Buildings. The research paper is titled “Urban building energy model: Database development, validation, and application for commercial building stock.”

Written By

Dr. Tim Sandle is Digital Journal's Editor-at-Large for science news. Tim specializes in science, technology, environmental, and health journalism. He is additionally a practising microbiologist; and an author. He is also interested in history, politics and current affairs.

You may also like:


Businesses should beware the insider threat, says a cybersecurity analyst.


FIFA Women's World Cup 2019 Final - The US team receiving their champion medals. Image - Howcheng, CC SA 4.0. The US Soccer Federation...


Anxious investors, employees and suppliers describe a scramble inside teetering Chinese property giant Evergrande.


Students fled from university buildings when the gunman opened fire - Copyright AFP DESIREE MARTINOla CICHOWLASA gunman killed six people on a university campus...