Alternate current, intelligent current
29th April, 2009
Instead of spikes in demand and coal-fired solutions, fridges and washing machines may soon be available that can regulate their own energy usage. Mark Anslow reports on a new generation of electrical appliances that ‘listen’ and learn
'What do you think was happening here?’ Jon Fenn, electricity operations manager for National Grid, is standing pointing at a jagged graph projected on to the wall of his ofice in the grid’s electricity control centre in Berkshire.
The graph shows the nation’s total electricity demand during the first round 2006 World Cup match between England and Sweden. Demand steadily falls throughout the first half, followed by a sudden spike at half-time, followed by another steady fall, then another spike and a plateau at the end of the match. Fenn is pointing to the lowest point of demand, right at the end of the game’s first half.
‘What do you think we were looking for?’ he asks?
I look blank. ‘We were looking to see if there was going to be any extra time played,’ he explains patiently.
For the controllers who staff the National Grid’s control room 24 hours a day, extra time in a national football match means something very significant. They are waiting for hundreds of thousands of kettles to be boiled at half-time, countless fridge doors to be opened and a multitude of kitchen lights to be flicked on. At half-time in 2006, electricity demand soared by almost two gigawatts in a matter of minutes – equivalent to suddenly needing the combined output of nearly two Dungeness B nuclear power plants. Twenty minutes later, as everyone sat back down in front of the TV, the demand had disappeared.
To anticipate this sort of spike in demand, the grid engineers quickly need to bring extra power plants online. Extra time played in the game could mean they bring the power on too early, risking tripping fuses on the grid. Bring it online too late, however, and blackouts could result.
It’s a fine balancing act, and also highly expensive and polluting. A key component of being able to match these sudden spikes in demand is what is known as ‘spinning reserve’ – essentially keeping a power station running but only using a part of its output, ready to ramp up to full power at a moment’s notice.
Because this kind of use is unsuited to nuclear power plants, which take days to come on- or ofline, and can potentially damage the more modern and sensitive natural gas plants, it tends to be the workhorse coal power stations that fulfil this ‘balancing’ role. One estimate suggests that more than 2.1 million tonnes of CO2are produced every year simply keeping these power stations ‘ticking over’, waiting for us to flip the kettle on.
It’s not just football matches that generate these spikes in demand, however. In fact, every winter’s day our demand for electricity soars from a night-time low of around 35 to almost 60 gigawatts during the evening rush hour. In the summer, that profile is different still – flatter, but with different peaks when air conditioning equipment is switched on in the heat of the day. Fenn can point to little spikes in our electricity demand during the middle of the night when night storage heaters suddenly trip into life.
‘We’re students of collective public behaviour,’ he admits.
A cool innovation
Endearing as it may be, however, our electrical behaviour is becoming increasingly problematic. The current reliance on coal power plants to balance out demand will come to end as a result of the 2007 Large Combustion Plant Directive – legislation that regulates the non-CO2emissions from coal and oil plants, and which will force several to close by 2015. And on the flip-side, as more wind energy is brought online the variability of the grid will increase. Although wind energy is not ‘unpredictable’ as some critics suggest – it can be accurately forecast hours in advance – it does prove a problem for the current grid setup, where supply has to be matched to demand at all costs. Wind may blow strongly in the middle of the night, when the demand is low, but slacken off during the evening rush hour.
What’s clearly needed is some way of matching not only supply to demand, but also demand to supply – some sort of what the industry likes to call ‘demand management’.
‘Don’t say demand management!’ hisses David Hirst when we first meet. ‘Say demand preperation. Only the electricity companies would be so arrogant as to talk about “managing” their customers.’
A former IT expert and the inventor of a technology that might help even out the variability on the grid, Hirst has developed a small, cheap piece of electronics that could be built into all new home fridges and freezers. Currently being marketed by British company RLtec, the device would constantly ‘listen’ to the frequency of the grid – a direct indication of whether the grid is over- or underpowered. If the grid frequency drops then the fridge would know that lots of consumers had suddenly increased their electricity demand – perhaps for the half-time cuppa – and that operators in the grid control room would be about to open the throttle on a series of coal-fired power plants. In response, the fridge could switch its cooling unit off until the grid frequency had returned to a normal level – in effect reducing ‘non-essential’ demand until the grid operators had managed to balance the system again, hopefully without resorting to too much coal.
‘The fridges would only remain off for between 15 and 30 minutes,’ Hirst says. ‘Any longer than half an hour, and the fridges would say to themselves, “stuff this for a laugh”, and start working again. Food preservation is paramount.’
Some good scientific modelling work has been done on this, which suggests that if each of the three million domestic fridges sold in the UK every year were fitted with this technology (known as ‘dynamic demand’), then the equivalent electrical response of all these units would be 35 megawatts – the size of a small wind farm. If, however, all of the UK’s 40 million fridges were eventually replaced with dynamic demand units, then the response level would rise to between 728 and 1,174 megawatts – a level that RLtec claims would make an entire spinning reserve power plant obsolete.
It’s an attractive idea, and one that National Grid welcomes.
‘Evening-out demand would make things very much simpler for us,’ Fenn says. ‘There’s a great opportunity for technology here.’
There is, however, also a great shrugging of shoulders when it comes to deciding who will pay for installing the dynamic demand equipment in the fridges. In theory, the rapid-response service provided by the fridges is worth a fair amount of money to National Grid – between £4.40 and £34.10 per fridge, in fact, according to Governmentcommissioned research. This is because it offsets expensive charges made by coal power plant operators for their usual balancing services. National Grid, though, is reluctant to commit to funding the fridges without knowing exactly how effective they will be in aggregate. There is also talk of using money from the Carbon Emissions Reduction Target (CERT) energy eficiency levy on the power companies – the tax that explains why your utility is throwing energy-saving light bulbs at you and offering to lag your loft. As yet, everyone is waiting for the result of a larger smart fridge trial, due to report back in 2010.
Fridges are only the tip of dynamic demand iceberg, however. For a start, Hirst says, they essentially only allow the grid to ‘borrow’ power for half an hour – after that, they all need cooling back down again. Where things get more exciting, though, is when you look at the possibility of shifting the ‘on’ times of other appliances, such as dishwashers.
Hirst calculates that if the UK’s 10 million dishwasher owners were to load up the machines with crockery and then, rather than switch them on immediately after dinner, set the machines simply to have the load washed between 11pm and 7am, then the grid would effectively have 10 gigawatt hours (Gwh) of flexible storage. This vast amount is equivalent to the capacity of the Dinorwig pumped storage plant in Wales, which pumps water up into a huge reservoir when electricity is cheap and then lets it roar through turbines when demand surges in the evening. It could also save a considerable amount of CO2 often produced during the evening peak by gas power stations.
The list goes on. Immersion heaters in our hot water tanks are, like fridges, currently a law unto themselves, tripping on when their thermostats tell them to. As long as water is hot for morning showers or evening baths, however, the exact time at which these devices run is not especially important to us – but very important in terms of running a low-carbon electricity system. Similarly, certain spaceheating systems, such as storage heaters or underfloor heating could become more flexible. What’s needed is a way of bringing these appliances together so that they know when is the most eficient time to power up.
Enter the smart meter – the only part of the much-vaunted ‘smart grid’ that the householder will ever see. A smart meter is essentially a meter that allows two-way communication: the electricity company can read the meter remotely and the householder can see both how much power they are using, and how much it costs. Unfortunately, that’s about where consensus on smart meters stops. Some would like the meters simply to give customers information on energy usage; others see the display part of the meters – sited in the house – as a tool to discourage householders from using energy at peak times; still others would like to see the meter communicate remotely with appliances in the house, allowing them to operate at the most eficient times of day.
Joe Short, an expert in the field and founder of the charity Dynamic Demand, warns that a mistake with smart meters at this early stage could spell disaster for developing truly energy eficient ways of running our homes.
‘We need to be very careful that the smart meter agenda is not driven by the agenda of the large energy suppliers,’ he says. ‘The big players are interested in smart metering because of the automatic meter-reading element, but we need a signal to get into the house – either a carbon or price signal. We need not to miss the opportunity of all these smart meters.’
David Hirst is worried about the influence of the energy companies on smart metering for a different reason.
‘Do you really want someone to be able to control the appliances in your house? Least of all the electricity companies?’ he asks, voicing a concern already raised by consumer groups.
Hirst has a different model for smart meters, one that explains why he insists on referring to demand ‘participation’ rather than ‘management’. He wants to see smart meters tune into future price broadcasts from the electricity companies, sent out every few minutes. A high electricity price would indicate high demand on the grid (and hence, high CO2 emissions), while a low price would indicate either low demand or an abundance of wind or solar energy. Appliances – laundry machines, dishwashers and even electric cars – would calculate when would be cheapest to run and plan to wait until the best time to switch themselves on.
‘It would mean you, and your appliances, would have a choice,’ Hirst says. ‘Occasionally, you may simply say, “I need it urgently – I’ll pay the extra”. That’s participation.’
The smarter way forward
The future of smart meters – and smart appliances – is yet to be written. A recent trial in the District of Columbia, US, saw 1,400 customers fitted with smart meters coupled to their air conditioning units. With the householder’s day-to-day permission, the electricity company was able to deactivate the home’s air conditioning system at times of peak electricity demand, and give customers a rebate on their bill as a result. The UK’s own trials, conducted by Ofgem, have been disappointing, dogged by equipment problems. But elsewhere in Europe – notably in Italy, where almost 30 million meters have been installed – the response has been positive.
‘Smart’ appliances and demand-responsive fridges are, of course, only bits of kit. If householders and tenants fail to engage with the new devices – and in one of the UK trials a quarter of those using energy monitors didn’t even bother to replace the batteries when they ran out – then no amount of technical wizardry will help. The purpose of all these devices is simple: in the words of Jessica Strömbäck of the VaasaETT Global Energy Think Tank at a ‘demand response’ conference in January, ‘It is important in the long run that customers change their view of electricity from a natural human right to the costly resource that it is’.
Mark Anslow is the Ecologist's News Editor.
This article first appeared in the Ecologist May 2009