BEACH: Building Energy Analytics for Cooling and Heating     



BEACH: Building Energy Analytics for Cooling and Heating - overview

The building sector accounts for 40% of today’s global energy consumption, which makes it one of the largest energy-consuming entities on a global scale. This statistic is rising every day. In fact, building energy consumption is projected to increase 50% by 2050 unless we actively implement new strategies to curb the growth.

Currently, large corporations are enduring global pressure to improve their level of environmental sustainability, as well as suffering the increasing costs incurred in operating their buildings. Heating, ventilation and air-conditioning (HVAC) systems dominate energy usage in commercial buildings, accounting for between 40 and 70% of the total building electricity consumption.

Therefore, to reduce overall energy consumption, carbon footprint and rising electricity costs, we must design new techniques for enhancing the energy efficiency of HVAC systems.

Our team of researchers has developed a new approach to operating a building’s heating, ventilation and air-conditioning (HVAC) system using machine learning and IoT data to drive down both energy consumption and costs. This will help decrease the rapid growth in global demand for energy and lower increasingly unaffordable electricity prices.

How have we achieved this?

To reduce energy consumption, our team have used a data-driven model to capture the various temperature dynamics in different zones of a building. We have integrated forecasts of weather conditions, building occupancy and electricity tariff structure, and have taken account the level of flexibility available in the thermal comfort range. By effectively forecasting and evaluating these factors, we are able to minimise the total cost incurred to operate HVAC systems in large-scale buildings.

By analysing this data, our team have developed a model that predicts the optimal sequence of set-point temperatures to be configured in the various zones of a building. This allows HVAC systems to be managed more efficiently and effectively whilst reducing operation costs and overall energy consumption.

This model was tested in a large office building located in Northern Australia, in two sections of the building spanning 1500 m2 and housing 100 people. Overall, the results displayed that cooling energy consumption were reduced by 20% with no thermal discomfort reported by occupants. This also resulted in a substantial reduction in annual electricity bills.

Our work is one of the first demonstrations of the potential benefits of a closed-loop cloud-based HVAC framework. The framework is uncostly to implement, and can be easily adapted to different buildings, which means it can be adopted across the globe today.

What’s more, our approach is easily retrofitted and can be adapted to different buildings, which we expect to make a huge difference.  Moreover, as of 15 August  2019, our solution is commercially available through the IBM Tririga Building Insights software suite.

The architecture of the BEACH system is shown below.

BEACH architecture