Machine Learning Model for Building Performance Optimisation

Technology Overview

Energy modelling software such as EnergyPlus can be used for developing analysis tools to monitor and analyse the energy efficiency of buildings. However, such software usually require high proficient experts taking considerable amount of time to develop the energy model which is highly customised for buildings with different equipments. In addition, it is often challenging to integrate the energy model with existing building energy management systems (BEMS).

The technology described herein is related to the development of a machine learning model which can optimise building operations based on building data from the building management system and hence, achieve building energy efficiency. The model, which can be installed in BEMS servers, possesses the capability to diagnose areas of inefficiency and provide inputs to control building equipment operations (e.g. pumps, cooling towers and chillers).

Technology Features & Specifications

The machine learning model for optimising building performance consists of the following specifications:

  • Does not require many input variables and has some characteristics of fast calculation and light model
  • Automatically pinpoint and diagnose specific areas for reduction in energy consumption 
  • Provides recommendation on set points for various building equipment (eg. chiller setpoint) to optimise energy consumption without compromising on the building occupants' comfort level
  • Enables energy usage forecasting 


Potential Applications

The technology is applicable for the building energy management industry where it can be integrated with building management system to diagnose areas of energy inefficiency and provide insights to forecast energy consumption in buildings. 

Market Trends and Opportunities

The market size of BEMS is expected to grow almost quadrupled in size, and will be worth over $1 billion by 2020, according to a report by Pike Research.



Customer Benefits

  • Reduction of energy consumption based on optimised building operations 
  • Better maintenance of Smart building implementation by self-diagnosis, automated operation


File Attachments

Make an Enquiry