A collaborative organisation for digital development in the energy sector
The idea behind Smart Energi is simple: by sharing costs and knowledge, you can achieve more and help drive faster and more efficient digital development in the energy industry. Smart Energi and Solita have jointly created new data and AI-based services that accelerate digitalisation through smart data analysis.
Smart Energi is a collaborative organisation for digital development in the energy sector. Today it has 10 member companies, including E.ON, Göteborg Energi, Tekniska Verken in Linköping and Sundsvall Energi. Solita coordinates the collaboration and leads specific development work.
“It started with an algorithm, which became an application, which then became an industry initiative, which gave rise to an innovation project with the support of the Swedish Energy Agency, Energiforsk, and researchers from academia. We develop methods and frameworks, and support companies in their use and operational development. And next year, an industry association will be launched”, says Anders Tholen, SVP Design & Strategy at Solita, and one of the driving forces behind Smart Energi.
An established collaborative organisation with governance, member meetings, knowledge sharing, joint analysis and solution design for advanced analytics and AI methods in energy distribution optimisation
K2 – A smart tool to analyse anomalies in energy consumption and energy distribution
Data Science Community – a community and partnership between Smart Energi, Energiforsk: Futureheat, the Swedish Energy Agency and universities and colleges
Data Science Innovation Portal – a portal with collected reasoning and innovation results in smart anomaly detection for the energy industry – for all companies to share and use in their own or collaborative development projects, such as in Smart Energi
The algorithm identifies anomalies in the district heating system
The algorithm underpinning the entire project is used to identify anomalies through data analysis. It is the result of work that we carried out together with Göteborg Energi.
“We achieved good results with our algorithm and soon realised that many more people would benefit from it. Specifically, this can involve detecting broken components or incorrect settings in the district heating system by looking at the data, rather than waiting for them to be labelled – either strictly physically or in the form of unexpectedly high bills for the customer. These methods enable you to carry out proactive rather than just reactive maintenance”, says Anders Tholen.
Data-driven and AI-based anomaly detection in the customer’s plant is good for everyone involved. This may mean, for example, faults occurring during the summer being detected immediately. Typically, they are discovered much later – often in the winter when heat is most needed. Using data showing anomalies makes it easier to fix faults in time and to optimise the fixing process in cooperation with the customer.
“Finding anomalies isn’t the hard part; all algorithms can do that. The problem is that they sound the alarm far too often. The challenge was to find an algorithm with high precision and accuracy. Energy companies need to know exactly which anomalies are worth acting on – both in terms of value for the customer and for investing maintenance resources”, says Maria Hansson, Business Designer at Solita.
We see that there are big gains to be made by developing algorithms for self-monitoring systems, but it’s something we wouldn’t have been able to do on our own.
Peter LindströmHead of Distribution & Operations, Sundsvall Energi
Uniform data paves the way for AI-driven innovation
Solita and Göteborg Energi decided to make the algorithm available to other energy companies, and started a collaborative initiative and developed a joint application: K2. The initiative quickly grew in membership. Since all member companies use K2, data is organised in the same form, and a taxonomy has been developed to further structure data along the entire chain from anomaly to cause and action. Having data structured in the same form not only facilitates joint solutions, it is also highly sought after in innovation and research. Lots of data, in uniform form, is paving the way for innovation driven by data and AI.
“There are so many benefits to working together in this way, as many energy companies have similar needs and challenges. And for those of us developing products and services, the better and more standardised data we have to work with, the better”, says Maria Hansson.
“Many people think that the energy industry will soon be totally connected, but this isn’t the case. We are quite far from a comprehensive solution for smart optimisation of the overall energy system, which is why this initiative, among many other collaborative initiatives, is so important. While many companies have digital and smart solutions for customer communication, production and consumption monitoring, they haven’t made much progress in optimising and automating the entire chain between production, distribution, maintenance and consumption. The full ecosystem isn’t there yet”, says Maria Hansson.
There are so many benefits to working together in this way, as many energy companies have similar needs and challenges. And for those of us developing products and services, the better and more standardised data we have to work with, the better.
Maria HanssonBusiness & Ecosystem Designer, Sustainable Value Creation, Solita
Joint Open Source-based development
Many member companies see great benefits in an industry-wide initiative like Smart Energi.
“For us at Borås Energi, it is important to be involved and contribute to effective digitalisation in the energy industry. By sharing resources, knowledge and ideas, we can drive digital development much faster – and at considerably lower costs than if everyone works individually”, says Andreas Carlsson at Borås Energi och Miljö.
“A major advantage is that everything is based on Open Source. In Smart Energy, we own everything we develop collectively – the idea being that the resources we invest will go towards solutions and work on developing the use of these methods and systems. We don’t wish to be locked into a single software vendor at the risk of becoming too dependent on that vendor’s software upgrades. Within Smart Energi, we, as a company, can control what we prioritise and roll out”, says Ulf Hagman at Göteborg Energi.
“We want to work proactively, not just act when something has already happened. We see that there are big gains to be made by developing algorithms for self-monitoring systems, but it’s something we wouldn’t have been able to do on our own”, says Peter Lindström, Head of Distribution & Operations at Sundsvall Energi.
By sharing resources, knowledge and ideas, we can drive digital development much faster – and at considerably lower costs than if everyone works individually.