Operational data from onsite energy systems (like heat networks) is extremely hard to come by. Very few people manage to get hold of it, and those who have it rarely share.
What are the typical loads in dwellings? What are the network losses? Do customers all demand heating at the same time or are demand events spread out?
Who knows? Engineers don’t stick around and find out how their designs work in real life; ESCOs hold their cards close to their chests; and many landlords fail to extract or make use of their own data.
This dearth of data has hamstrung the industry at a time when it should be racing ahead. It’s one of the biggest reasons why, when it comes to energy performance, we’re just not getting better fast enough.
In late 2014, when DECC put out a call for proposals to improve heat networks, we saw a chance to unlock some of these data silos and accelerate development of the heat market.
Our proposal was a web-based system for visualisation and analysis of network performance data. In addition, we wanted it to be a platform for stakeholders to learn from each other in a secure and anonymised way. Happily for us, DECC decided to back our project.
Trust would be crucial: without it we wouldn’t get people using the system. Our approach to data handling had to be comprehensive, transparent and digestible, so that all users of the system have a clear understanding of how it works.
For this reason, we enlisted the help of the Open Data Institute (ODI). Founded by Sir Tim Berners-Lee and Sir Nigel Shadbolt, the ODI provides training and consultancy on the best ways to share data appropriately, while keeping personal and confidential data secure.
So what is Open Data? The ODI says:
Open data is data that anyone can access, use or share. Simple as that. When big companies or governments release non-personal data, it enables small businesses, citizens and medical researchers to develop resources which make crucial improvements to their communities.
From the beginning, the ODI team proved hugely capable but highly partisan. All our meetings featured a variant of the following conversation:
ODI: Have you considered making any data from the system available as Open Data?
Guru: Not a chance – anonymised data will only be shared between paying users, not with the public. We’re an SME after all, not a big corporate, and we have to focus on revenue!
ODI: Hmmm.
But the ODI never gave up. Until one day something clicked and we began to think about what might happen if we did make some anonymous data freely available for others to use. Which data would have the biggest impact? And how would we share it without killing the commercial model of this new visualisation and analysis platform?
After a lot of mulling, we had an idea.
Arguably, the biggest and most common design problem for heat networks is oversizing. This wastes huge amounts of capex and dooms the resulting networks to suboptimal efficiency for their entire service lives.
Why is oversizing so common? Because designers lack confidence that their design will meet peak demand and, to avoid liability risk, they compensate by oversizing. As a result we routinely encounter systems designed for loads four or five times larger than they’ll ever experience.
Equipped with real-world data, designers would be able to confidently specify smaller pipes and plant without worrying that they are endangering their professional indemnity insurance. But what specific information is needed?
There are two pieces of information that underpin engineers’ sizing calculations: peak demand per flat and an allowance for the fact that residents probably won’t call for heat at the same time (called the diversity factor).
Just two pieces of information, shared as Open Data, could have a big effect on the market. Actually, when we did the numbers, we realised it might have a huge effect – potentially saving hundreds of millions of pounds in capex and running cost over ten years, while giving the sector a much needed kick up the learning curve.
We had to weigh this up against the potential commercial value of the information to Guru. In the end we were able to justify the move for both noble reasons and mercenary ones (e.g. it would drive potential customers to the new platform, which contains a trove of more detailed data). And we let the ODI know that we’d finally given in.
So this Thursday, at the launch event for the visualisation and analysis platform, we’ll also present the Open Data that will be freely available via a Creative Commons ShareAlike license.*
Initially, this Open Data will take the form of diversity curves, combining peak demand and diversity factor, based on more than 40m data points across 2000 dwellings. Over time, the curves will be updated as more data flows into the platform. But from day 1, the data will provide a justification for proper sizing and a stark wake-up call for the wildly oversized designs that are so common now.
*Any derivative works using the data have to be attributed to Guru and shared publicly under the same license.
Excellent. 😀
We’ll follow your lead on this one when the format is unveiled on Thursday. 🙂
—
Marko
[…] kinds understand that they can open up data and work together to solve common problems while still being competitive. Whether it be in the banking sector, the agriculture sector or anywhere else, businesses support […]
[…] understand that they can open up data and work together to solve common problems while still being competitive. Whether it be in the banking sector, the agriculture sector or anywhere else, […]