Reliable and cost effective storage is key to keeping the lights on in the renewable world, so E&T visited a Cambridge company that thinks they have found a solution.
In the drive to reduce carbon emissions the pressure is on to generate more and more electricity from renewable sources. The EU has set an ambitious target of generating 20 per cent of electricity from renewables by the end of the decade, with further increases over the following years.
If renewable energy is to increase much beyond its current share of the electricity generation market however, there will need to be some significant advances of reliable and cost-effective storage systems.
Aside from easing the intermittent nature of renewable energy, storage can provide 'ride-through' for momentary outages, and extended protection from longer outages. Coupled with advanced power electronics, storage systems can reduce harmonic distortions, and eliminate voltage sags and surges.
In combination with renewable resources, energy storage can increase the value of photovoltaic (PV) and wind-generated electricity, making supply coincident with periods of peak consumer demand.
But a new storage device from a team of engineers in Cambridge could be set to boost the prospects of renewable energy with a solution to the on-off power of wind and solar energy.
The concept of the Isentropic Pumped Heat Electricity Storage system (PHES) grew from work in the late 1990s by Isentropic's chief technical officer, Jonathan Howes, who, like many engineers, was interested in methods of clean power generation.
'As an engineer it has been a little frustrating over the years, because until two years ago, nobody seemed to have really taken on how serious energy storage actually is,' he explains. 'I think that it has now begun to sink in, insofar as none of this actually works unless you can store the stuff. There is talk about super grids, but super grids are super expensive.
'In the 1990s, I spent my spare time trying to dream up ways of generating energy, and it just dawned on me in the late 1990s that it was all solving the wrong problem because with any renewable you were going from generation on demand to generation when it is available, and that implies a massive problem. Unless you solve it, you really have to redesign society, and redesigning society is not an attractive option so it is better to find a way of storing energy.'
Howes explains that perfect machines tend to work at one power level. 'You can intuitively see that,' he says. 'You don't need to be an engineer to see that if you design a machine to do a certain job, it is best off when it is doing exactly that job rather than doing half that job or twice that job.
'Every machine has an ideal speed at which to run. If you run it at the ideal speed in terms of efficiency you will get the best possible use of fuel. If you generate electricity constantly at that power level 24 hours a day, you need to be able to take that power 24 hours a day. But society doesn't need that power 24 hours a day, so if you can store the stuff, you can get away with smaller generating equipment.
For an energy storage method to be successful, cost is of paramount importance, since electricity is only made more expensive by performing additional processes. 'The Isentropic solution is of exceptionally low cost and high efficiency and in this respect can match or outperform the only viable current large-scale technology, pumped hydro,' Howes added. And the facts bear out his assertion with PHES coming in cheaper than the only other current viable option in pumped hydro-storage.
The Isentropic PHES system utilises a highly reversible heat engine/heat pump to pump heat between two storage vessels containing particulate mineral. A gas circulates through the machine and, to store energy, is first compressed, which raises its temperature to 500C. It is then passed through one of the stores in which it heats the mineral by direct contact, which also cools the gas to close to the original temperature.
It is then expanded back to its original pressure, which cools it to around -160C and it is then passed through the other store where it cools the mineral by direct contact and the gas is then warmed back to close to its original temperature. This process requires energy, which can be supplied electrically via a motor and is thus an energy storage process. Discharge is the opposite of the charge process and releases energy, in which situation the machine drives an electrical generator.
The innovations that have allowed high efficiency to be realised are primarily concerned with ensuring that each process within the system is performed with minimum loss. This results in an electricity-in to electricity-out (round trip efficiency) in the range of 72 to 85 per cent. This compares very favourably with typical pumped-hydro values of 74 per cent.
'Isentropic is also able to demonstrate a storage build cost of $10 per kWh for a utility-scale application, which is the lowest cost of any storage technology, including battery and pumped hydro,' Howes added.
Proving the concept
Two prototype machines have validated performance predictions, and design of a small utility scale demonstrator is underway. The first prototype was based on a lot of mathematical modelling.
'With good engineering you do a detailed simulation or mathematical model of what you want to build,' Howes says. 'You go and build it and then you do some testing which tells you what is wrong with your modelling.'
The results from that first prototype didn't deliver quite what Howes wanted, so he improved the modelling and designed a second prototype that concluded testing about a year ago. 'The purpose of the second prototype was to get the thermodynamics right,' Howes adds. 'Many heat engines look elegant on paper, but thermodynamically they are not too clever so the efficiency is not that great. So we thought we would go all out on generating the best thermodynamic machine we possibly can.
'We won't worry too much about mechanical design or mechanical quality because if we can't get the thermodynamics, we will draw a line under it and walk away. We finished testing that about a year ago, the results exceeded expectations which was a good thing and a bad thing. It was a good thing because it meant we had really got something, but a bad thing because it meant that my mathematical model still wasn't that great.'
That prototype was very heavily instrumented and so Howes has refined the analysis further since then and redesigned that machine to turn it into prototype number three. 'We are just concluding that process and that will be assembled in the next month.'
Helping with wind
There are various locations at which these devices could sit, such as the site of a wind farm. 'If you imagine a wind farm with no storage trying to satisfy with 100 per cent reliability on a certain load demand, that wind farm would have to be almost infinitely big, because you have got to have it absolutely everywhere to make sure that you are always taking energy.
'That is not viable. If you imagine that same money being spent on storage equipment with no generation, that is also pretty useless because there is no energy to store. If you imagine a wind farm with a certain amount of storage, as you increase the amount of storage, you can give reliability to a low profile with a smaller wind farm. If you add storage you can use fewer turbines to satisfy the same demands.'
The minimum effective storage is thought to be about 18 hours of the farm capacity and between two and three days the optimum. 'What that allows you to do is that the wind farm operator could then go to a network supplier and say 'I will guarantee to provide this energy profile every single day' and they will get a better price for their energy.'
With storage installed, it gets quite interesting because if the wind doesn't blow for three weeks, there is a store on site that allows the operator to buy electricity in from the cheapest supplier. 'It turns a wind farm from a niche renewable energy supplier into a power station. When there is no wind blowing, it buys in the cheapest power to satisfy the contract. It is actually quite an interesting model I think.'