Building the digital freak to reshape the energy economy

The figures are stark. To limit the climate to 1.5°C warming, yet still enable living standards in the developing world to continue to rise, the world must remove carbon emissions at a rate of around 10% every year.

Clean energy thought leader Michael Liebreich contends that there are only two ways to achieve it: “On the supply side, taking carbon out of our energy system (clean energy); and on the demand side, taking energy out of our economy (energy productivity). In most discussions, the supply side grabs most attention, sucking all the air out of debate about demand-side efficiency. But the fact is, we will get nowhere near our climate targets without a step change in energy use, as well as production. Indeed, most models of climate-compatible energy pathways see around half of the heavy lifting being done by improved energy productivity.”

This is not an argument for de-growth or managed decline. Bountiful, clean, cheap electricity from low-carbon sources, managed efficiently, can support economic growth and personal liberty. We start by decarbonising electricity, then we electrify everything.

In my previous article I discussed the need for a digital twin model to automate the energy consumption and optimisation of a household or business – a Fully Rational Energy Aware Konsumer, or digital freak. In this article I look at how the coordination of these twins can maximise efficiency on a district or neighbourhood level.

Centralised electricity generation

A century ago, we conceived of and built a centralised electricity generation network, with a small number of large, capital-intensive production facilities. The customers of this network were treated as atomic, non-cooperative and non-interacting entities. Generation capacity was optimised from a central command structure to exceed demand at all times.

Picture a customer drinking from a proverbial fire-hose in this example; plenty to drink (no danger of thirst) but little attempt to optimise efficiency across the network, and no ability for an individual to influence the outcome or stem the flow, beyond closing his mouth for a bit.

The 21st century energy system we are building no longer looks like the linear production line described above. The generation capacity is diverse and distributed, with a very large number of small (renewable) assets. Electricity flows in multidirectional fashion within localised demand and supply loops.

These localised demands can be balanced according to need – communication between customers is key here as energy is traded, stored, utilised and curtailed. Rather than an individual standing in front of the fire-hose, picture a group of friendly strangers cooperating at a bottle party to make sure everyone has enough to drink, and the designated drivers aren’t left out in the cold.

The 21st century certainly sounds like a great time in this context...but how might it work in reality? Imagine a small business that has a high but intermittent requirement for energy. A ceramics manufacturer that needs heat to operate its kilns, for example. As part of its energy productivity initiative this business has invested in a solar roof, and converted the kilns to electric heating, eliminating natural gas from the production line. Their roof may almost cover the average energy demand of the business, but cannot hope to exceed the peak, when a kiln is running. However, they will know their overall average requirement for heat energy, based on their forward-orders and predictions of sales on a monthly or weekly basis. Crucially they will also have a degree of flexibility as to exactly when they fire the kiln.   

Energy networks can gauge with tremendous accuracy how much wind and solar power will be available, three days or more in advance. A moving price for electricity can therefore be calculated, and users can bid for energy capacity within a time window that meets their overall requirements but still achieves the lowest available market price. 

Our ceramics company’s energy officer can coordinate with her sales and production teams to choose the best time to fire the kiln – taking advantage of the lowest market price when electricity is at its most plentiful. The inverse may also be true. On days when the wind drops, or the sun is clouded over, there may be a financial incentive for them NOT to fire the kiln. Curtailment of consumption – provided they have managed to build up sufficient stock in advance – can result in payments greater than the raw energy cost (sometimes up to three times the purchase price per kWh). 

In theory the same benefits should be realisable by individual consumers. Smart meters are an important part of this new energy market – but from a consumer’s point of view they represent only the tip of the iceberg. In representing the total energy consumption of a household in 30 minute intervals they allow for a degree of dynamic pricing, but unlocking the benefits requires more effort and time than even the most committed of consumer freaks could bring to bear. As households become micro power generators with solar rooftops, and micro storage facilities with electric vehicles and house batteries, the imperative to automate the trading and consumption at domestic level will continue to grow. 

Unpeeling the energy onion 

In his article ‘The Energy Onion — a simple conceptual model for a smart system’ David Sykes describes how our new energy system – being demand driven – can be characterised. He uses this model of concentric layers, with the customer at the centre and the Distribution System Operator and Electricity System Operator at the outside: 

It is tempting to re-publish much of the entire article here, but I will concentrate on just the inner layers of the onion to illustrate how our digital freak (otherwise known as the digital twin model of our idealised customer) can be built to optimise the energy productivity of the overall network. 

Sykes’ onion starts with energy efficiency — you don’t need to build generation and networks for energy you don’t use. This is almost the exact opposite of the system we inherited from the 20th Century, which was largely built by nationalised industries incentivised to enable provision to the entire population, and where production was used as a measure of progress or growth. 

Energy efficiency didn’t feature highly when we built our original generation driven system – it was (partially) counter to the business requirement. If people use more energy the system just finds more generation to meet the extra demand and the operators are rewarded and incentivised for doing so. When you take a bottom up approach, energy efficiency falls at the centre of system design. 

The next layer is self-consumption. Electricity that is generated on the rooftop is best used immediately within the household or business itself. We encourage customers to generate their own energy and shift their own demand (or store it) to consume it themselves. Storage can be in the form of heat – in domestic hot water systems, heat batteries or the fabric of the building – or as electrical energy in the batteries of an EV or home energy system, for use later.

Energy can also be ‘stored’ in the form of useful work – the manufacture of energy intensive items such as the ceramic wares in the example above. Other examples in business might be the production of desalinated water or refined aluminium. In the domestic sphere it is in the form of laundered clothes or deep-frozen food, which are the major energy-intensive activities of a household.   

Dynamic pricing and communication of the household requirements require information flows in both directions to be coordinated by a localised freak, which acts to respond to the cost signals from the outer layers of the onion. These cost signals should reflect the carbon intensity and network constraints in real time, to be balanced against the ‘value’ of the energy within the household at that instant. 

Balancing and cooperation 

This approach reduces technical losses in the wider networks and minimises constraint in the low voltage networks – as well as maximising consumption of locally generated green electricity. Moving further out from the customer is local balancing and cooperation. This can take the form of peer-to-peer trading with other households in the same street or district. This means we encourage the balancing of generation and demand at a local level. In practice it might be communities sharing generation or storage assets or suppliers incentivised to develop portfolios of customers in local clusters. 

If each individual household is represented on these sharing networks by their digital freak, they can settle their requirements of demand or ability to supply for optimal productivity at lowest cost and carbon intensity. This requires little more effort than the initial commitment to sign up for a service and trade with other like-minded consumers in a completely automated fashion. 

This means that the localised balancing and cooperation engine becomes one of the most crucial building blocks of our emerging efficient grid. This is a layer of functionality that is almost entirely missing from our current design of system. It can be conceived of as a cloud-based entity acting as a trading and coordinating platform. Indeed, given that it shares many of the features of applications such as MindMyBills, Agile Octopus and NextDoor, there is an argument that suggests that it is the social media giants rather than the energy majors that are best placed to insert themselves as intermediaries and capture this part of the market. 

Could we expect to see Facebook Neighbourhood Energy launching in the near future? Is this the reason that Google thought Nest was a good buy at almost $3bn back in 2014? And might it be the method by which they effect a recovery of its losses? Time will tell, but in the fight to decarbonise industry and win the hearts and minds of uninterested consumers, this could well be a crucial battleground.