Intelligent heating à la Aarhus

Nick Harrison from the Telegraph couldn’t have put it better when describing Denmark’s second largest city, Aarhus, as a place that finds “the perfect balance between being an innovative university city while retaining its historic legacy”. The city has a long standing tradition of exploring innovative approaches when it comes to energy sources, and has spent the last 50 years refining its energy roadmap all while preserving its heritage.

As populations grow and cities expand, so does the need for energy and sustainable solutions. For the Danish city, the path to a greener future includes 2000 km of pre-insulated pipework, Raspberry Pis and the latest digital technology.

With a clear green vision of becoming carbon neutral by 2030, Aarhus is investing smartly in its district heating network and is digitally paving the way for a sustainable future. Through cloud, IoT and Machine Learning technologies, the city is bringing District Heating into the 21st century.

AffaldVarme Aarhus (Department of Waste and District Heating) is the brain behind the city’s modern waste management and district heating network, and is responsible for supplying heat to 95% of the municipality’s 300,000 inhabitants. Since the 1980s, Affaldvarme Aarhus and consumer-owned Combined Heat and Power (CHP) plants have cooperated on covering the municipality’s needs in terms of heating.

Considering the volume they handle, and the complexity of the network, the city-owned company relies heavily on accurate and timely data in order to operate efficiently and monitor their operations. Quick and easy access to their data is therefore crucial, and in 2013, they made the leap to cloud technology. “The cloud has proven to be an intuitive tool that allows our customers to easily view and read the data without special technical skills,” explains Adam Brun, Development manager at Affaldvarme Aarhus. Recently, and in partnership with Microsoft partner Inspari, they moved their operations to Azure.

We used to work with a highly closed environment with very limited options. When we piped all of our measurements into Azure, all of the data became suddenly available from all of our management and planning tools.

Adam Brun, Development manager at Affaldvarme Aarhus

Quick access to precise measurements is absolutely key to the company, both from the customer’s side and from the company’s side. “We are now able to measure everything from consumption, temperature and flow going in and out, and we can quickly shape extra services out of that new and easily available data,“ continues Brun.

By handling data from more than 43,000 measurements points through Azure and Microsoft SQL, the company is able to monitor and operate the network faster, with more flexibility and much more efficiently than ever before. Moving to the cloud has also enabled the energy company to reap the benefits of IoT technology and reduce the measurement costs dramatically. “Currently we have Raspberry Pi devices set up in our distribution net that transmit data directly to Azure. That alone helped us reduce our measurement costs from 100,000 DKK down to 1,000 DKK. That’s 1 % of the cost we normally pay,“ explains Brun. 

From a financial perspective, moving to the cloud costs us a third of what we used to pay.

The scheduling and production planning software is at the core of Affaldvarme Aarhus’ district heating system. By moving their platform to Azure, the company is now able to rapidly scale their operations depending on demand and the time of the year. Furthermore, the flexible aspect of the cloud makes it easier for AffaldVarme Aarhus’ partners and suppliers to integrate their solutions across the company’s systems both from a back-end and front-end perspective. “Besides all the known advantages of using Azure, one of the things that was new to me is that it made it a lot easier for our software suppliers specialising in production planning tools to integrate their solutions across our systems. It made all of our IT processes more efficient and faster – and it saved us months of work,” explains Brun.

The next big step for the energy supplier is to start using Machine Learning in order to capture vast amounts of fine-tuned data, based on frequency and vibration analysis. The idea here is to use Rasperry PI devices to harness around 40,000 samples/sec, crunch them, and identify patterns. “We will be able to notice changes in patterns in frequencies emitted by specific parts of the network. It will help us prevent pumps failing and we will be able to track it down to the exact part that is failing based on the frequencies that it’s emitting.”

When a pump is new it sounds like Queen. When it’s failing it sounds like Metallica. Machine Learning will help us capture the different frequency and vibration patterns and turn that precious data into actions.

By integrating the latest digital technology under one roof, Aarhus Affaldvarme is rapidly gearing up to meet the 21st century energy and climate requirements in a smart and sustainable manner. In the near future, the company will be able to draw behavioural insights from the data they are capturing, and fully leverage on the power of ICT to benefit their customers well-being and needs. 

*   *   *   *   *

There’s still a lot of work to be done before the European Union reaches its Europe 2020 target of 20% of renewable sources. At the time this story was written, approximately 50% of the total energy consumed in Europe is being used for the generation of heat for domestic or industrial purpose, and there is more heat being wasted than needed to heat all of the buildings. With smart District Heating, the excess heat could easily be repurposed, recycled and reused, and as seen with Aarhus, when combined with the right technology, results can reach encouraging new heights.