Share

The UK energy sector currently wastes 54 percent of all the energy it produces, according to the Association for Decentralised Energy. Couple this with an increasing need to use renewable energy and a growing demand – the Institution of Mechanical Engineers predicts UK energy demand could outstrip supply within just a decade – and you have an industry in crisis.

Unlike the more modern renewable energy sector, where wind turbines are installed with sensors from day one, the rest of the sector is struggling to harness data from their assets, creating huge inefficiencies that lead to high levels of wastage.

All of these factors open the door to startups like Open Energi, which provides the devices and technology platform required to bring internet of things (IoT) capabilities to those older assets.

The London-based company provides large commercial customers, like water companies or Sainsbury's, with a "gateway device" which collects energy and operational data for assets, from furnaces and fridges to water pumps. Fellow London-based business Origami Energy also offers a similar module called the Origami Energy Router.

Read next: Meet the tech startups working to fix the UK's energy crisis

An engineer fits a small control panel on site that allows it to communicate between Open Energi and system equipment, as well as measure the frequency of the local grid. The company installs smart meters on the equipment to measure consumption second by second.

It then uses proprietary algorithms to define when the assets have some flexibility in terms of consumption and tells it to turn off for a certain time period. This benefits the customer and also helps balance the National Grid better to match supply and demand. The National Grid pays Open Energi for the service and commercial customers like Sainsbury's share this revenue.

Energy sector IoT

The other benefit of the Open Energi platform is that it gives these companies better visibility of their energy consumption in close to real time through an online portal called Dynamic Demand.

Speaking at the DataWorks Summit in Munich last week, Michael Bironneau, technical director at Open Energi told Computerworld UK: "If you consider those industries, they are very data poor. The water sector for instance, some of the pumps we have dealt with are 30 to 40 years old, they did not have the IoT back then and obviously they don't have much data about how their pumps work so we can look at their operations and energy data.

"So being able to bring data to those really data-poor industries has a lot of applications and far-reaching effects we have only just started to grasp."

Bironneau said that customers are starting to value the access to IoT information now more than the revenue they are sharing. "I think it was Welsh Water that said to us the value is having all the data and having people who analyse it for them," he said.

Tech stack

To get the speed of insight required to deliver this real time data, Open Energi first shifted from an SQL Server database to a Hadoop environment in 2015, before embracing more streaming analytics.

"At the time our database was under increasing pressure, so we looked at the Hadoop ecosystem," Bironneau said. "We wasted a lot of time trying to set it up from scratch on Windows, and if you have ever tried that, don't. It is an absolute nightmare. So we downloaded the Hortonworks sandbox on Windows and got up and running with a proof of concept data warehouse in a couple of weeks."

He said that this move brought the infrastructure spend down to a fifth of what it was spending with SQL Server. This move to Hortonworks also opened the company's eyes to streaming analytics capabilities with the Hortonworks Data Flow platform.

So the data science team started to test out some machine learning algorithms and real-time analytics for customer assets.

"We work in an environment which is traditionally very data poor where you have limited control," Bironneau said. "So we thought if we can take the data that we have captured over the last five years and apply it to a data-poor environment fast enough, what kind of savings could you achieve?"

Read next: Why machine learning could be the next frontier for data centre operations

The key is what Bironneau calls "real-time validation".

"Because all of our forecasts happen in near real time for events that will occur within the next five minutes to hour, you get an observable quantity back to see if your original forecast was right or not," he explained. The data science team can then start to implement counters to the errors in the forecast so that the system gets smarter over time.

This has opened up capabilities like "peak price avoidance, tariff optimisation, and imbalance services, which when you bundle these things together, the uplift can be almost a twofold increase in revenue for customers," Bironneau said.

Find your next job with computerworld UK jobs