There are consumers of energy who study their bills every quarter, who shop around for the lowest tariffs and the greenest suppliers. They insulate their homes, improve their efficiency and invest endless hours in research. The industry has a collective term for this vanishingly small proportion of customers, which typically represents a single-digit percentage of the total. They’re known as ‘FREAKs’.
Building the digital freak to reshape the energy economy
I imagine that this term stands for Fully Rational Energy Aware Konsumers. These are the people who can be relied upon to make sensible purchasing and usage decisions. They are committed to driving down emissions and prepared to make behavioural changes.
The unfortunate fact though, is that there simply aren’t enough FREAKs around. For all the worries expressed – over half of all adults are now ‘concerned’ or ‘alarmed’ about global warming – when it comes to how we consume energy the vast majority of citizens simply do not care enough. Or know enough, or have enough time, capital or mental energy to make an impact.
In my view, it’s unfair to put the responsibility for our broken energy systems onto the consumer. And it’s not sufficient to let the market dictate policy. We must reimagine and reshape our energy economy and we should use the digital twin method to put an endlessly empowered and informed customer – a digital FREAK, if you will – at the heart of this revolution.
Decarbonising domestic heating
We partnered with the Energy Systems Catapult to produce their farsighted ‘heat as a service’ platform to look at long-term behavioural shifts in the bid to decarbonise domestic heating. This system uses a digital twin of the household to enable efficiency optimisation. The approach has enormous promise in many other parts of the energy landscape, in particular in electricity production and use, where displacing fossil fuels is the aim.
We must remove carbon emissions from our global economy at a rate of around 10% per annum in order to limit the climate to 1.5°C warming, while still enabling living standards in the developing world to continue to improve. No other course of action can be considered sufficient, and this fact must dominate political, economic and commercial discourse.
The principal challenges in reshaping our energy economy to reward energy productivity stem from our outdated model of an energy system and how it should be designed and operated, coupled with a failure to consider customer behaviour as a function of this productivity. In tackling this last point, it is clear that building a better insight-driven model of the customer is key. Once built, this model – our digital FREAK – can be programmed to optimise the system for the benefit of the customer that it models, whilst improving the energy productivity of the system as a whole.
Separating the current buzz around digital twins from the practical aspects of implementation is the subject of much of our work at Cambridge Consultants. My colleague Harsha Kudoor has recently described how the digital twin approach can be applied to urban or indoor farming. In his fascinating article, he details how the basic principles of an interaction model can be applied to optimise the growing conditions for each and every instance (plant) in a crop and so increase food yields with reduced inputs.
Dynamic virtual representation
A digital twin, put simply, is a dynamic virtual representation of a product, its environment and its interactions. Once built it can be used as a design tool, particularly in markets where physical prototypes are terrifically complex and expensive to produce (automotive and aerospace dominated early uses of the tool). This allows the possibility of modelling and recognising undesirable behaviours or unexpected interactions in complex systems.
Once a product is launched into the marketplace, the digital twin can continue to be modelled and updated with real-world inputs acquired during product usage. This extended system modelling allows for monitoring, maintenance, and data gathering to be used for future improvements.
Producing a digital twin of a household user of energy and enabling it to interact within the wider network – as a Fully Rational Energy Aware Konsumer – makes it possible to increase the energy productivity of the network as a whole, while driving costs down for the customer.
The diagram below is a simplified representation of our digital freak – the components of a digital twin system for a household within the context of a localised balancing and cooperation network:
At its most fundamental level, the MVP for a household digital twin will need a degree of hardware and software. The implementation will vary, but may include:
- Household appliances (heating, cooling, etc) connected to a smart home controller
- Means of generation and/or storage
- A cloud-based computation and storage platform to run the interaction model on instances of environment and plant data models
- Sensors and embedded system for reporting data from the physical instance to the digital twin
- A user application to run and view analytics
- Actuators and system integration to make changes to the physical instance based on feedback from the digital twin. These could be networked smart switches but would preferably be smart appliances coupled via APIs to the controller
Learned behaviour
Each household digital twin will update based on learned behaviour. For instance, if it is observed that a household’s energy consumption always peaks on a Sunday evening, when the laundry’s done, then the internal ‘value’ of home-produced electricity is then at its highest. The export price would have to be exceedingly high in order to overcome the utility of doing the laundry at this time.
Conversely, it may be observed that the heat pump which maintains the temperature of the dwelling can shift its hours of operation to take advantage of supply when it is at its cheapest and lowest carbon intensity.
This operation would therefore take place not only based on the signals from the room thermostats, but also taking in information from the localised balancing and collaboration engine and from external environmental and weather sensors. This time shift could be performed automatically by the digital freak, provided that the acceptable bounds of comfort were not exceeded.
Want to find out more?
In conclusion, I believe that the digital twin approach to customer-centric energy management has a significant part to play in the battle to reduce the carbon intensity of the grid. Two key elements of this system are the digital FREAK, for the intelligent monitoring and control of an energy customer at atomic (household or business) level and the localised balancing and coordination engine.
These two systems can interact with the wider distribution network and ultimately the electricity system operator through a simplified interface based on spot price and carbon intensity. All the complexity of controlling individual loads and coordinating the energy usage is handled as close as possible to the customer.
