AI and IoT in building energy management


As carbon emission reduction targets drive utility adoption of clean and renewable energy resources, fluctuation in energy supply and demand present a huge challenge for utilities in ensuring reliability of grid networks.

The variations in energy supply and demand, and failures by emerging distributed energy resources models to provide base load capacity for utilities, are driving an increase in the rollout of demand side management programmes.

This article was originally published in Smart Energy International 4-2018

Energy providers have over the past years topped investments in increasing consumer awareness and adoption of home energy management solutions, and government parastatals continue to partner with solution providers to reduce energy use in buildings.

The research firm predicts the number of smart thermostats connected to smart grid control systems will reach 80 million by 2022, amongst the 300 million smart homes to have been developed by that year.

A senior research analyst with Navigant Research said: “With the development of the smart home and connected devices, energy management has become a critical part of the digitisation of the home.”

Continued research and development of technologies within the energy management segment is resulting in the introduction of new embedded solutions, artificial intelligence, IoT enabled devices and cloud-based services that drive traditionally disconnected, on-site (premise) systems.

This applies to central air conditioning and heating systems in homes, or a commercial building’s HVAC building management system.

Metering & Smart Energy International spoke with Steve Nguyen, vice president of product and marketing at BuildingIQ, to learn more about the latest trends within the AI, IoT and smart home energy management space, their integration and impact in the smart energy market.

Integrating AI and IoT in building energy management

The best way to demonstrate how AI and IoT can best work together is through the creation of a “digital twin” – a digital representation – of the building. Occupancy details, occupant comfort, thermal and weather models are used to create a digital twin of a building and can be enhanced further through machine learning and cloud computing.

The core of a digital twin is the building management system (BMS), which provides a stream of data that feeds this advanced model. Data from various IoT devices can refine the model.

These can include IoT-enabled electro-pneumatic thermostats for analogue buildings; IP-enabled lighting systems, with built-in presence and temperature sensors; blind controllers, and more.

With the creation of a digital twin, stakeholders can see how a building would respond to numerous variable changes in real time and help create a responsive building design.

Factors hindering adoption of IoT enabled building energy management solutions

Adoption can be hindered due to a number of factors, but mostly it comes down to a lack of knowledge and experience using high-tech solutions in buildings, as well as the assumption that AI and machine learning are meant to replace the role of the facilities personnel.

They aren’t! Thus, we need to educate the market. It’s still a very new concept for a building management team to work with an organisation’s IT team to allow remote access of a building’s control systems – even if only for monitoring.

Another big factor limiting the IoT building market is the big name BMS companies that monopolize the market. They often have competitive services that restrict access to vendors, effectively vendor locking buildings. If you want to control peak energy usage, become more energy efficient, and positively impact productivity and comfort, you need to deal with the BMS directly.

Regional penetration of smart energy management solutions

Adoption rates of various services are dependent on region. In North America, data visualisation and ticketing services are popular. It is worth noting that older buildings have the option to implement simple/basic solutions and transfer to more sophisticated services, as their needs change.

Many buildings in the region are using advanced predictive energy optimisation (PEO) software, which offers closed-loop control of a building’s HVAC system based on predictive analytics. This is particularly useful in regions with peak demand tariffs and efficiency incentives.

In Australia, there is a growing opportunity for commissioning, due to the fast rate at which new buildings are being constructed within the country. Recently, we’ve been more involved in implementing energy management in buildings from the ground up. We’ve seen a lot of success with this approach in the Australian market where we have a partnership with BUILDINGSENSE.

Infrastructure requirements

Most commercial buildings have a BMS in place, which platforms easily integrate with.
The BMS is then used as a proxy for control and a source for gathering data about the building’s performance.

However, success can be reached without a BMS. The BMS can be bypassed to gather data directly from various pieces of equipment.

Top 3 considerations to be taken when implementing a BMS

1. Open protocol: BACnet/IP is the internationally accepted standard from data in modern buildings. When implementing a BMS, it’s imperative that owners/operators have their BMS utilise BACnet/IP and that they have control over the access to the BACnet/IP data and points. Having such control over the data ensures that should they wish it, owners/operators may work with the BMS controls provider or 3rd party vendors to implement cloud-based systems that leverage the BMS.

2. IP connectivity: It’s surprising given where we are as consumers that many buildings


Granted, doing so is a security risk; however, their IP departments should understand those risks and how to mitigate them.

Therefore, it is very important that owners/ operators work with their information infrastructure teams to ensure that providing IP access to the BMS is a) expected, and b) designed in.

This would include any necessary DMZ, specialised routing, reserved IP addresses and other measures. Here’s a fact – there’s no such thing as an on-premise cloud solution. If you want the power and benefits of cloud computing and solutions – things like instant deployment of updated algorithms, unlimited storage, limitless compute power – make sure you are OK with an IP-connected BMS.

3. Documentation: Perhaps the Achilles heel of many a system is the lack of good documentation. Creating and maintaining good system documentation (control sequences, asset registries/records, software licensing, etc.) is a key to protecting the owner/operator from unplanned expenses down the road.


At the end of the day, the building environment is much more complex than the home, yet many of us in the industry want many of the same services. I come from a home energy disaggregation business, yet that is not available to the commercial sector. We can solve the question of what equipment consumes what amount of energy, but it takes a lot of math.

The influence of AI is similar. AI can do a lot. For sure it can steer both home and building owners in the right direction with greater and greater accuracy. However, unless you are willing to spend a lot of money replacing assets, we have to be smarter and more inventive about how we use data.

Concepts like the digital twin can be taken very far both in the home and in the commercial environment. Still, the basic tenants are the same – get connected, get standardised, get on the cloud. SEI

Success story

One great example is that of St John of God Murdoch Hospital in Australia, which houses 500 beds and serves nearly 50,000 patients per year.

For the project, projected energy consumption was significantly reduced via data visualisation, fault detection and closed-loop control of specific HVAC zones. Despite only managing half of the 413,872-sq. ft. (34,450-sq. m.) floor space of the hospital, HVAC energy usage was cut by 10% and overall energy consumption for the entire hospital by 5%.

By utilising the software on a subscription basis, St John of God Murdoch Hospital has essentially been able to sign up for continuous energy savings. The initial energy reductions have translated to roughly $4,500 in net savings per month.

About the author

Steve Nguyen leads product and marketing at BuildingIQ to bring a suite of cloud-based energy intelligence service to the commercial, healthcare, education, and other building markets. Steve joined BuildingIQ from Bidgely, where he ran marketing and helped utilities to better engage consumers with their energy use through the power of disaggregation. He holds a B.S. from Brandeis University and M.B.A. from Boston College.

About BuildingIQ

BuildingIQ (BIQ.AX) provides technology-enabled services to help building owners and operators worldwide save energy, increase operational efficiency, and improve tenant comfort. Through its 5i cloud-based platform, BuildingIQ delivers on the promise of the Internet of Things (IoT) to help visualize, analyze, control, and optimize energy usage within a single facility, campus or portfolio of buildings. More than 100M square feet of global building space is currently being serviced by BuildingIQ to reduce operating costs.


Previous articleThe way towards a decarbonised Europe
Next articleThis is not a drill!
Nicholas Nhede is an experienced energy sector writer based in Clarion Event's Cape Town office. He has been writing for Smart Energy International’s print and online media platforms since 2015, on topics including metering, smart grids, renewable energy, the Internet of Things, distributed energy resources and smart cities. Originally from Zimbabwe, Nicholas holds a diploma in Journalism and Communication Studies. Nicholas has a passion for how technology can be used to accelerate the energy transition and combat climate change.