Clean energy stored in electric vehicles to power buildings

Stored energy from electric vehicles (EVs) can be used to power large buildings – creating new possibilities for the future of smart, renewable energy - thanks to ground-breaking battery research from WMG at the University of Warwick.

Dr Kotub Uddin, with colleagues from WMG's Energy and Electrical Systems group and Jaguar Land Rover, has demonstrated that vehicle-to-grid (V2G) technology can be intelligently utilised to take enough energy from idle EV batteries to be pumped into the grid and power buildings – without damaging the batteries.

This new research into the potentials of V2G shows that it could actually improve vehicle battery life by around ten percent over a year.

For two years, Dr Uddin's team analysed some of the world's most advanced lithium ion batteries used in commercially available EVs - and created one of the most accurate battery degradation models existing in the public domain - to predict battery capacity and power fade over time, under various ageing acceleration factors - including temperature, state of charge, current and depth of discharge.

Using this validated degradation model, Dr Uddin developed a 'smart grid' algorithm, which intelligently calculates how much energy a vehicle requires to carry out daily journeys, and – crucially – how much energy can be taken from its battery without negatively affecting it, or even improving its longevity.

The researchers used their 'smart grid' algorithm to see if they could power WMG's International Digital Laboratory – a large, busy building which contains a 100-seater auditorium, two electrical laboratories, teaching laboratories, meeting rooms, and houses approximately 360 staff – with energy from EVs parked on the University of Warwick campus.

They worked out that the number of EVs parked on the campus (around 2.1% of cars, in line with the UK market share of EVs) could spare the energy to power this building – and that in doing so, capacity fade in participant EV batteries would be reduced by up to 9.1%, and power fade by up to 12.1% over a year.

It has previously been thought that extracting energy from EVs with V2G technology causes their lithium ion batteries to degrade more rapidly.

Dr Uddin's group (along with collaborators from Jaguar Land Rover) have proved, however, that battery degradation is more complex - and this complexity, in operation, can be exploited to improve a battery's lifetime.

Given that battery degradation is dependent on calendar age, capacity throughput, temperature, state of charge, current and depth of discharge, V2G is an effective tool that can be used to optimise a battery's conditions such that degradation is minimised. Hence, taking excess energy from an idle EV to power the grid actually keeps the battery healthier for longer.

Dr Uddin commented on the research:

"These findings reinforce the attractiveness of vehicle-to-grid technologies to automotive Original Equipment Manufacturers: not only is vehicle-to-grid an effective solution for grid support – and subsequently a tidy revenue stream - but we have shown that there is a real possibility of extending the lifetime of traction batteries in tandem.

"The results are also appealing to policy makers interested in grid decarbonisation."

Explore further: Scientists propose better battery system for smart home use

More information: Kotub Uddin et al. On the possibility of extending the lifetime of lithium-ion batteries through optimal V2G facilitated by an integrated vehicle and smart-grid system, Energy (2017). DOI: 10.1016/j.energy.2017.04.116

skystare

not rated yet Jun 14, 2017

How did the word "clean" get into the headline? If the batteries are charged from the grid in the first place, they cannot be used for "grid decarbonization". Only the elimination of coal-fired generators can do that.

antialias_physorg

5 / 5 (2) Jun 14, 2017

and that in doing so, capacity fade in participant EV batteries would be reduced by up to 9.1%, and power fade by up to 12.1% over a year.

Now *that* is interesting.
I hope some other groups can corroborate this. Would be nice if one could load up one's EV during the day with excess PV/wind from the grid at next to no cost (negative cost?) and then sell part of it back to the grid at night without degrading the batteries. It'd be win-win for everyone.

While it might seem at first this'd mean the energy companies are paying you twice - which they certainly aren't in favor of - it would save them having to build (as much) storage of their own. So in the end it might save them cash overall.

Eikka

5 / 5 (1) Jun 14, 2017

I hope some other groups can corroborate this.

It is a long known fact that lithium batteries keep better at partial charge than at full charge, and the effect is quite dramatic at the top end of the voltage range. Discharging the battery reduces the average state of charge over time and leads to improvements in calendar life.

However, see-sawing energy in and out of the battery still damages it. It's merely a question of which is is more - leaving the car sitting with a full battery, or drawing current out of it. Of course, if you didn't recharge the battery that full in the first place, and then did not discharge it to add more cycles on the battery, it would last even longer.

So it's a bit of a fixed competition.

For the practical reality of it, companies like Tesla guarantee their cells for the equivalent in miles of 600-800 cycles on the battery, which reflects the type of cell they're using (cheap).

At $150/kWh for the cost of the cells, even if you improve the cycle life by 10% up to 900 cycles - heck - give it a generous 1000 cycles and it still costs 15 cents a kWh to put energy through the battery, because you're wearing it out. That's more than the retail price of electricity.

Would be nice if one could load up one's EV during the day with excess PV/wind from the grid at next to no cost (negative cost?)

That's only possible while the excess PV/Wind is being subsidized by the kWh. This subsidy is the reason why anyone bothers to dump it on the grid, because they have to sell a kWh to get the subsidy.

The power is not free, as you're paying the subsidy through taxes, and through other people's taxes which add to the prices of goods and services you buy.

Also, don't forget to add round-trip losses which increase the price by another 25% because you aren't getting all of the energy back out.

And there it is, in the article itself:

In this work we assumed a particular driver behavior, namely that drivers recharge their vehicles to 100% every night.

What if they don't? Well then the whole study goes out the window.

They tried to fix it by further assuming that the battery is only charged full near depletion, but that's not realistic either. What if the driver puts a limit on the charger and only charges up to whatever they need?

MR166

3 / 5 (2) Jun 14, 2017

This is just more eco-babble from over-funded universities.

So let me get this straight, I park my Prius at work and find out that the building has drained xxx miles of the range from my car because I have some extra to spare. Did I fail to mention that I work for the Socialist Party? But it gets better, I got a call at work and have to drive an additional 50 miles to see my sick mother. Well mom you are just plain out of luck but fear not it is all for the greater good.

greenonions1

2.3 / 5 (3) Jun 14, 2017

MR - this is simulation - to talk about what is possible. It would seem obvious that you will have control of your vehicle. If you don't want to participate in V2G - you will presumably be able to program your car - how you want to. I cannot of course comment for sure on something that has not yet happened. I know that current EVs allow you to program the charge system - to fit your needs. The idea is that it will be beneficial to you - so you will be happy to participate in V2G. It is understood people like you - who are afraid of progress - will resist the new. That is fine - don't buy an electric car - or don't plug it in at work. You will be able to put up solar panels - and batteries at home - and charge off them. Oh no - that is progress - forget that suggestion. Never mind - carry on with your coal burner.

The current warranty on Tesla batteries is 8 year unlimited. https://www.tesla...warranty This does not cover batter fade. Battery fade seems to be very manageable on the Teslas. There are plenty of articles on it - if you learn to use google. Just one - https://electrek....adation/ I think the characterization of Tesla's batteries as 'cheap' - is highly unfair.

1 / 5 (1) Jun 15, 2017

The current warranty on Tesla batteries is 8 year unlimited.

That's because with the larger packs, 85 kWh and up, you're unlikely to reach the mileage necessary to wear them out. The 8 years limit comes up first.

85 kWh * 800 cycles = 68,000 kWh. Divide by 0.38 kWh/mi and you get 179,000 miles. That would be 22,375 miles a year over 8 years, so you'd have to drive 61 miles every single day, including the weekends. You may wear it out if you're using your Tesla as a taxi, but that's probably not covered under the warranty terms.

I think the characterization of Tesla's batteries as 'cheap' - is highly unfair.

If it's not cheap, then why are they claiming sub $150/kWh cost? Tesla is making batteries to the most energy per dollar, and that necessarily comes with compromizes. It's not "cheap" in the China-cheap sense where they skip on manufacturing quality, but simply that they cannot make a battery that is both inexpensive, capacious, and lasts forever.

not rated yet Jun 15, 2017

There are plenty of articles on it - if you learn to use google.

The article you linked makes for a very bad example, as it ignores the basic characteristics of lithium-ion batteries: the wear-out mechanism is not linear. Their data stops before the accelerating drop starts.

They're simply extrapolating from the trend, ignoring the fact that when the cycle limit is up the battery cell starts to degrade exponentially and drops to almost zero capacity within 10-15% more cycles.

With Elon Musk claiming 500,000 mile artifical tests, you have to take it with a spoonful of salt.
With a perfect battery pack with well matched cells, ran under ideal conditions, and charged at the ideal regime - and cycled rapidly so the battery has no time to age - it is possible to reach many more cycles. Then you also exaggerate how many miles you get per watt-hour - like Tesla does (EPA mileage has been consistently lower) - you can get any number you want. It's just marketing.

Here's a typical curve of a laptop grade lithium cell:

http://batteryblo...les2.jpg

It goes on for about 1000 cycles to the knee point where the accelerating wear really picks up (~63% capacity) but it reaches the 80% capacity point at around 800 cycles. These are the kind of cells that Tesla uses - they push the maximum capacity out of the material, and as a compromize they don't get a very long lifespan.

In the article given by greenonions, there's actually no telling what the real percentage of wear on each individual datapoint is, because the cells actually do vary in capacity and each battery ends up with somewhat more than the nominal amount, so the one car that went to 125k miles (200k km) with 93% left can easily have started with more capacity in the first place. These outliers are driving their trend as there are few cars that have been driven that far, which makes the conclusions very shaky and unreliable.

3 / 5 (2) Jun 15, 2017

Eikka - all your hand waving is defied by reality. The data on current Model S batteries is coming in - and it defies your 80% capacity be 800 cycles. https://electrek....adation/ but what do you care about reality? Next gen batteries look like 1200 cycles - still above 95%. But hey - why don't you and Willie Ward go work for Tesla - you can show all those stupid engineers how wrong they are. Keep driving you coal burners...

https://electrek....ifcycle/

Eikka - all your hand waving is defied by reality. The data on current Model S batteries is coming in - and it defies your 80% capacity be 800 cycles. https://electrek....adation/

That is exactly the study I was criticizing: they haven't got good enough data to make the claim. The data has only a handful of cars that have been driven past 100,000 miles, and the true initial capacity of their batteries is essentially unkown, so the present state of decay is also unknown.

The conclusions are bunk because the data is not strong enough to support them: If you eliminate one or two of those outliers, the trend will change dramatically, so instead of hundreds of cars their result actually depends on only a couple, and even they haven't reached anywhere near 800 cycles. The P85 model at 200,000 km has actually done only 550 cycles or less.

Repeating a lie won't make it true.

Next gen batteries look like 1200 cycles - still above 95%.

You're simply pulling it out of your hat. There are batteries that achieve that already, but they're not cheap or compact, and hence not used in EVs where everyone's trying to minimize weight and cost.

Electric car batteries do not need to perform over that many cycles, because a large enough battery cannot be driven to exhaustion - as I pointed out even 800 cycles is enough for a car that goes 200+ miles per cycle, because it's more than most cars are driven, or can be driven in the first place - the calendar life of the cells is the limiting factor that comes up first.

Unless of course you decide to hook your car up to a V2G scheme and spend those cycles, but what would be the point? The cycle life and calendar life are connected, so that the older the battery is, the more the cycles wear it out, so you'd just make your battery die faster.

The original warranty on the 85 kWh battery pack before Tesla changed it to unlimited, was for 160,000 miles (240,000 km), about 700 cycles, and none of the cars in the electrek site graph have reached even that. Over 90% of the cars in the study haven't even reached half that.

You can't tell a trend from that. Or you can, but it will be whatever you want it to be. Here's why:

http://www.askana...esh.html

Q2. Why, at all, do we need to bother about outliers ?
Outliers might mislead analysts to altogether different insight regarding a data.

From a statistician point of view, it distorts the range, mean and standard deviation of the data. It also gives a wrong trend in the data as demonstrated below :

Greenonions: the article links to its source data on google docs, and they have added more high-mileage vehicles to it since. In the data graphs section you will find the estimate on cycle count, which I screenshot for you here:

http://www.dumpt....niu9.png

The new trendline shows that the capacity starts to collapse around 900 cycles. They are apparently using an actual battery fade model to come up with the trendline.

However, they calculate one cycle relative to the average battery capacity which is already subject to wear, whereas I count one cycle as relative to the initial capacity, so their cycle numbers are about 10% higher than mine. Why? Because calculating hypothetical wear using the average capacity would mean that the load changes as the battery ages, which does not happen.

So where their chart says 900 cycles, that's about 820 cycles relative to initial capacity. Close enough. Their data supports my assertion.

You can also clearly see how some of the cars have much more battery capacity than nominal. Some still have 102% capacity at 200 cycles, and the spread is large. That's why the handful of high-mileage cars can change the results dramatically - if were (un)lucky these few cars have started with extra miles on the bank and drive the results up for the rest of them.

And it seems to be so, because the very next chart shows capacity remaining vs. age, and the trendline actually curves up: the older cars seem to have better batteries - as if Tesla initially built them better and then the quality dropped.

Also, editors' notes:

11 Feb 2016: The trendline for the first chart (Mileage vs Remaining range chart) has been updated. The old trendline was a third order polynomial trendline. It worked fine for 0-120,000 km where there are lots of entries but after there were no more entries it showed a sharp drop. The new trendline is polynomial until the data ends but linear afterwards.

So it's actually not a relevant extrapolation, because we know the battery fade is not linear. It should show a sharp drop at some point, but of course a linear trendline will not.

You're simply pulling it out of your hat.

Sod off Eikka. I even provided a link to that claim. Again for you - https://electrek....ifcycle/
From the link

The improved cells that they created from their research have performed exceptionally well after over 1,200 cycles

I am sorry you hate progress. I am sorry you hate that reality knows better than you do. I just don't understand the need of you, and MR, and Willie etc. - to need to piss on the hope the human race has for a better way. I think we will keep moving forward - and some day we will live for ever, reach out beyond our planet, do amazing things. You can still drive your coal burner. I just don't understand why you think it cool to stand in the way of that progress. I am so tired.

Sod off Eikka. I even provided a link to that claim.

Well I'm sorry it looked exactly the link you already repeated twice.

And the second article isn't relevant to your claim, because it doesn't apply to the EV batteries. The new cell is of the NMC type, which Tesla uses for their Powerwall products, while the cars get NCA type cells. Why? Because NCA offers higher energy density than NMC - less weight to carry around. It says so in the article that you obviously didn't read once again:

Tesla currently uses nickel cobalt aluminum (NCA) battery cells for its vehicles. For its stationary energy storage products, like the Powerwall and Powerpack, the company uses nickel manganese cobalt oxide (NMC) cells, which typically have a longer cycle life, but less energy density.

As for

I just don't understand why you think it cool to stand in the way of that progress. I am so tired.

Again. I don't. I stand against bullshit pretending to be progress.

5 / 5 (1) Jun 15, 2017

The reason why fake progress is dangerous is because people don't check up on it after they've decided that they did good by paying into yet another scheme.

They set up the projects, pay and congratulate themselves about how bright the future is going to be now that they're "part of the team", and they tell stories about how well it's all going with all the indicators going up and up - nevermind that they're just exaggerating our making up the numbers as they go along and ignoring contradicting data - and then years and years later: nothing.

Meanwhile actual realistic alternatives suffer by not getting the necessary funding and attention, so the whole thing is retardation, not progress. Progress happens despite rather than because of whatever hype you bought into.

You should never lie about e.g. how many homes your wind turbine can power, but unfortunately the liar gets the funding and you don't, so the people are left with the illusion that everything is going fine.

to need to piss on the hope the human race has for a better way.

To have hope, and I mean actual hope instead of wishful thinking, you have to have accurate and realistic information about your options, so you can choose the best.

Hype isn't hope - it's self-deception.

It imperative for anyone who claims to strive for progress to expose false information, hype and exaggeration, and any sort of "sales pitch" that tries to slip you aspartame for sugar to sweeten up the deal.

I stated that next gen batteries "look like 1200 cycles - still above 95%" and I supplied a source. You then declare I am "pulling it out of my hat." Here is what the article said - and it totally supports my assertion.

If made into a car battery pack, 1,200 cycles would translate to roughly 300,000 miles (480,000 km) – meaning that a battery pack could still retain about 95% of its original energy capacity after ~300,000 miles – or 25 years at the average 12,000 miles per year

It imperative for anyone who claims to strive for progress to expose false information

Perhaps you should look hard at yourself about that one. When that includes lying, as you have done in the past to disparage renewables - it may tell you something about yourself. None of the information I gave today was false. The Tesla model S is real - and the data we presented was real. I really despair - watching you small minded trolls - never missing an opportunity to be negative.

Eikka - to be clear - let's give just one example of you lying. You said.

companies like Tesla guarantee their cells for the equivalent in miles of 600-800 cycles on the battery, which reflects the type of cell they're using (cheap).

That is a lie - and it took me a few seconds to google the fact that Tesla's warranty is unlimited mileage. If you are such a crusader - fighting for the noble cause of truth - why do you spread misinformation like this?

"The Battery and Drive Unit Limited Warranty covers the repair or replacement of any malfunctioning or defective Model S or X lithium-ion battery for a period of 8 years or unlimited miles/km, with the exception of the original 60 kWh battery (manufactured before 2015) that is covered for a period of 8 years or 125,000 miles (200,000 km), whichever occurs first. "

Onions you are only partially correct. There is a mileage limit on the earlier cars.

1 / 5 (1) Jun 16, 2017

There is a mileage limit on the earlier cars.

But not on current cars - which is of course the salient point. If you want to talk about what Tesla used to do - you of course should specify such - especially if you are the crusader - defending the world against misinformation and false hope.