EV charging and hydro generation/distribution challenges

[danbrotherston]

As for load balancing, there is actually a fair bit of research on smart grid stuff. But it really isn't a technical issue, building software for it isn't hard it's kind of an understood task. The big problem is that it is contrary to the self interest society we have built. It requires cooperation of a majority of the population.  So we either need regulations or economic incentives, and I suspect there needs to be considerable policy research into this.

For the power management (it's not really "balancing" as it's a question of managing the load and making sure it's no higher than capacity) we can have incentives to reduce the load -- and we do have that today already, to an extent, with time-of-day pricing.

The problem is that it encourages people to make the right decisions, but it doesn't actually cap or prioritize consumption. Imaging a hot August afternoon, with everyone's air conditioner turned on, plus a few million EVs plugged in. If the demand exceeds capacity, we'll have brown-outs. So, what do we do? Price electricity in August super high so that people won't use air conditioning, or that businesses will reduce/stop electricity consumption in the afternoons?

I think it's not a simple problem. If we had the ability to curtail or disable EV charging in such a situation, it would enable us to stay within capacity, but our electrical distribution system doesn't have that kind of functionality. So, we (1) build more generating capacity, (2) implement super pricing, as above, (3) require all EV chargers to be remote-controllable by the hydro, (4) something else.

And I think we should move this to its own thread ...

This is exactly the trade off between central planning and capitalism, yes, with central planning you have more control, but the self interest argument is possibly superior, if my EV is charging because I am trying to get home for some time critical event, and getting home is important is really important to me, I'm willing to pay the higher price to do that (provided I am able, because inequity is still a thing). Where as you might simply prefer to have a warmer set point on your A/C.

But getting economics right, I think is much harder. For example, how do you charge people the incremental cost of a watt...if our capacity is x and we need x+1 that last 1 watt should pay the 600 million dollars it takes to build a new plant? Except now we've built it and x+1 is cheap now. I suspect the main issue is that there are different levels of responsiveness to parts of the economic system. Building a new plant takes a decade and hundreds of millions of dollars. Turning off one EV takes microseconds. So we need some intentional planning.

Microgrid stuff makes this more interesting actually, if every electrical generator was a rooftop solar installation, it would make the supply responsiveness much closer to the demand responsiveness....

[danbrotherston]

I would start by just making pricing information available in the electrical socket. Then you could tell your dryer a maximum price it should wait for before starting. Same for air conditioners etc.

This is way simpler than trying to design protocols for the power company to turn off your stuff, and also decentralizes the problem: all the power company has to do is provide the correct pricing information (which they should be doing anyway), and every device manufacturer can compete to use that information in the most effective way.

For heating/cooling, one can even imagine a thermostat where you can tell it to cool to 25° for example if power is one price and 23° if it is a cheaper price.

The trick is to harness self interest. Self interest always exists and cannot be entirely eliminated, nor is it clear that one would want to do that. Instead, appeal to a majority’s sense of cooperation to put in place a system that harnesses their natural self interest.

Yes, this would help. But even this would not ensure that we don't have brown-outs unless your pricing goes extreme at the peaks.

And I think that this, too, would already be a big project: how long would it take to get all local hydros to provide pricing info at the socket (or by wifi or something) -- and to get the majority of appliances and air conditioners upgraded to utilize that information? A decade? Not much less, I think.

That depends on your incentives for generators, if they are running near capacity at typical peaks, then yes, but if their capacity is well above typical peaks, and we have a large grid with good interconnections, these things even out enough. It is the situation we have now.

As for upgrading equipment, a decade is the minimum time to roll out a change like this, if tomorrow the rule was that all equipment had to have this ability.

Intrestingly, you mention brownouts, those are actually the current mechanism for managing excessive demand. It is less than ideal because it doesn't discriminate that finely. I get the sense in some places power utilities have agreements with major industrial users who can curb their use in some circumstances, which is useful because presumably they make up a big enough chunk of usage that curbing their usage is meaningful.

It's also useful because they can do so intelligently rather than browning out the local hospital.

A microcosim of this is having backup power generators. Some people have them for their homes, and for most people, I find it very silly (obviously people in locations with highly unreliable power, which can be out for days, may need one). It's an expensive, maintenance intensive piece of equipment, because you can't deal with a few hours of darkness a few times a year. In our building, we have a backup power generator that only supplies emergency lighting and the elevator (although the elevator grounds in case of power failure). That is, we have a generator that manages only the emergency equipment. It still costs us thousands of dollars a year.

[ijmorlan]

Obviously a capitalist sociey such as ours, (contrary to modern conservative views) would prefer an economic model to a regulatory model, I entirely agree. 

But, I think economic models are harder to make work. Ultimately, the best economic model is one with perfect information. Ultimately we don't have that...even as a society, which means we must design the model to try and make it work. That's really hard, and subject to perversions. Regulation too may end up with perversions, but more direct control means adjusting for these issues is faster.

You might as well say “free” rather than “capitalist”. Simple example: on those occasions when I have to wait in my car on a hot day, I idle so as to run the air conditioning. Do I feel bad about that? Not in the slightest; I’m still causing a tiny fractions of the emissions caused by somebody who commutes to Toronto every day (and those people are considered A-okay in our society).

A regulatory approach to cars would probably forbid the “useless” idling of vehicles (in fact, does; all those anti-idling bylaws). But an economic approach says that I can choose how to spend my resources. If one person wants to spend it on commuting every day, that is their choice; if another wants to spend it on occasionally idling and save more of it for other activities, that is their choice. That is why I support the carbon fee and dividend rather than complicated regulations about who may produce carbon dioxide and for what purpose.

Similarly with electricity. As long as I’m fully paying for my electricity use, it’s really nobody’s business how I use it. And if we’re going to enhance the power grid to feed information from the grid to devices, do the simple thing: feed the current price to devices and let their owners determine what to do with that information. People are already used to doing price comparisons at the grocery store and elsewhere; I think they are capable of judging how much it is worth it to them to air condition their house colder than Hoth.

[tomh009]

Similarly with electricity. As long as I’m fully paying for my electricity use, it’s really nobody’s business how I use it. And if we’re going to enhance the power grid to feed information from the grid to devices, do the simple thing: feed the current price to devices and let their owners determine what to do with that information. People are already used to doing price comparisons at the grocery store and elsewhere; I think they are capable of judging how much it is worth it to them to air condition their house colder than Hoth.

As long as there is enough electricity (being generated) for everyone, that makes sense.

But if there is consumption beyond that, you end up situations like the start of the pandemic where the grocery shelves were emptied and some people could not buy the things they were looking for.

[danbrotherston]

Similarly with electricity. As long as I’m fully paying for my electricity use, it’s really nobody’s business how I use it. And if we’re going to enhance the power grid to feed information from the grid to devices, do the simple thing: feed the current price to devices and let their owners determine what to do with that information. People are already used to doing price comparisons at the grocery store and elsewhere; I think they are capable of judging how much it is worth it to them to air condition their house colder than Hoth.

As long as there is enough electricity (being generated) for everyone, that makes sense.

But if there is consumption beyond that, you end up situations like the start of the pandemic where the grocery shelves were emptied and some people could not buy the things they were looking for.

Indeed. Economic models are nice, but rarely capture the full human condition. Even leaving externalities aside, I don't really believe that most people spend their money smartly. And that's before we even touch on inequalities.

When you say something like "your resources", how much of those resources are really yours and how much should really be redistributed. Obviously, it depends on how much resources we each have. I am not arguing for strict equality here, but certainly nobody in our society should be starving because they can't afford food, or to power a fridge to keep that food, and certainly nobody in our society should be a billionaire when such a thing is occuring.

Ultimately, economics and more centralized planning are both tools, neither is inherently good or bad, but each has different strengths or weaknesses. When it comes to something as complex as a hydro grid, there is no clear answer as to what is the right tool, or even that a single tool is sufficient. But it's entirely reasonable to keep all options open.

And ultimately, social priorities are not modeled in economics, there is no value in being a good person beyond what you believe you get from others, or at least I refuse to subscribe to nihilism or whatever you want to call it and believe that economics are the only basis for human behaviour. To contrive an example, if we are forced to choose, we should choose to power hospitals over the charging billionairs EVs not because of economics, but because of humanity.

Anyway, as I pointed out, until our grid is truely decentralized, central planning, whether by the government, or public, or private utility companies will always be playing a role anyway.

[plam]

However, the projection is that the electrical grid will become less and less clean in the next few years as there is a shift to more natural gas away from some nuclear. I think EVs efficiency should be based on the average energy emissions, rather than emissions at the time of charging...I suspect getting that detailed is probably irrelevant and at that level of detail, is going to be missing other side-effects like infrastructure investments to support charging anyway.

Solar is also going to continue to become way cheaper. It's really cheap already. It also isn't the time of day when people are charging their cars though: presumably many people are at work and not charging their cars there.

Most multi-residential buildings have little spare capacity, and I suspect the neighbourhood substations may have the same challenge.

Something that could help is that it would be the neighbourhood substations that I would hope to need more demand for charging at night rather than multi-residential buildings. That depends a lot on urban design. I've read about condos in Montreal where people just leave the condo by car in the morning and come back in the evening, so no better than suburbs. But with our fewer-spaces-than-units condos that are coming up, I can hope that there is also less of that car commuter usage pattern.

If tomorrow, everything we ever built was electric, it would still take 10-15 years to convert the existing fleet. More likely everything going electric means a 30-50 year timeframe.

This is more than enough time to transition.

I'm tired of this defeatism. We need to change our society, we CAN change our society. We need to stop making excuses for NOT changing our society.

Average age of car in the Canadian fleet seems to be somewhere between 9 and 12 years. There are going to be network effects too so I wouldn't be surprised at no gas cars left in 10 years.

I've seen some news about airplanes going renewable too. "Sustainable aviation fuels" might be a thing. I'm somewhat skeptical for various reasons, but we'll see.

I am generally really also skeptical of people saying "oh but we ship things across the world" without actually considering how low-emissions it is to ship. Yes, it would be better to emit less sulfur, but that's not as bad as CO2 (says the person who lives far from a port). There are of course also benefits to having less disposable crap, and of making things in Canada, sure.

Do full EV's charge themselves to a degree as well? Perhaps increasing that capacity should be the focus, to have an EV that rarely needs to be plugged in.

I don't think the physics works out, since perpetual motion isn't a thing and we lose a lot of kinetic energy to air friction and road friction. Certainly cars are more aerodynamic than in the past, and I bet that regenerative braking also helps reduce particulate emissions from braking. But I'd rather redesign cities to require less driving.

Indeed. Economic models are nice, but rarely capture the full human condition. Even leaving externalities aside, I don't really believe that most people spend their money smartly. And that's before we even touch on inequalities.

...

Ultimately, economics and more centralized planning are both tools, neither is inherently good or bad, but each has different strengths or weaknesses. When it comes to something as complex as a hydro grid, there is no clear answer as to what is the right tool, or even that a single tool is sufficient. But it's entirely reasonable to keep all options open.

And ultimately, social priorities are not modeled in economics, there is no value in being a good person beyond what you believe you get from others, or at least I refuse to subscribe to nihilism or whatever you want to call it and believe that economics are the only basis for human behaviour.

It is good to keep all tools open. We sometimes like to reject possibilities out of hand based on ideological reasons that we shouldn't.

I don't really feel like I know enough about this smart grid / economics stuff and it's hard to say. Empirical economics is becoming more of a thing so we get more insight into how people behave in practice. It's somewhat rational and also based on heuristics. There is a lot of irrationality though. Mostly we do what we've done before (which is why nudging works well as an intervention).

[tomh009]

Average age of car in the Canadian fleet seems to be somewhere between 9 and 12 years. There are going to be network effects too so I wouldn't be surprised at no gas cars left in 10 years.

I am generally really also skeptical of people saying "oh but we ship things across the world" without actually considering how low-emissions it is to ship. Yes, it would be better to emit less sulfur, but that's not as bad as CO2 (says the person who lives far from a port). There are of course also benefits to having less disposable crap, and of making things in Canada, sure.

9-12 is the average so many cars are older than that. I will personally be surprised if we reach 80% EV market share (of new cars, not the entire fleet) within 10 years. And in remote areas it will take much longer.

As for ships, bunker fuel is getting much cleaner. It was traditionally 4.5% sulphur, dropped to 3.5% in 2012, and it has been 0.5% since last year. Still higher than automotive fuel, but far cleaner than it used to be.

[ijmorlan]

As long as there is enough electricity (being generated) for everyone, that makes sense.

What does that mean? A wealthier society can afford to generate more electricity than a poor society. In 99% of societies that have existed to date, “enough” electricity was none at all; and even now, “enough” in most places is an amount that probably wouldn’t run my dishwasher.

There is no way around it: you can ration by price, letting each person decide how to use their resources; or you can pretend that a central planner is capable of figuring out what people need and want. Hint: they can’t.

[danbrotherston]

As long as there is enough electricity (being generated) for everyone, that makes sense.

What does that mean? A wealthier society can afford to generate more electricity than a poor society. In 99% of societies that have existed to date, “enough” electricity was none at all; and even now, “enough” in most places is an amount that probably wouldn’t run my dishwasher.

There is no way around it: you can ration by price, letting each person decide how to use their resources; or you can pretend that a central planner is capable of figuring out what people need and want. Hint: they can’t.

There is more than just price which impacts consumption. Economics are not pure. We have different abilities to consume power. For example, a society which outlawed EVs (or never invented them) would see individuals using less power than one which didn't, even if power is cheaper. Cheap power also encourages uses which we wouldn't see in a society with expensive power (like, desalination for example), but it's not true that economics is the only way to affect consumption. Especially something like desalination, is likely to be centrally planned, a society could choose not to do it, even if there was cheap power.

[danbrotherston]

Average age of car in the Canadian fleet seems to be somewhere between 9 and 12 years. There are going to be network effects too so I wouldn't be surprised at no gas cars left in 10 years.

I am generally really also skeptical of people saying "oh but we ship things across the world" without actually considering how low-emissions it is to ship. Yes, it would be better to emit less sulfur, but that's not as bad as CO2 (says the person who lives far from a port). There are of course also benefits to having less disposable crap, and of making things in Canada, sure.

9-12 is the average so many cars are older than that. I will personally be surprised if we reach 80% EV market share (of new cars, not the entire fleet) within 10 years. And in remote areas it will take much longer.

As for ships, bunker fuel is getting much cleaner. It was traditionally 4.5% sulphur, dropped to 3.5% in 2012, and it has been 0.5% since last year. Still higher than automotive fuel, but far cleaner than it used to be.

I agree, even if tomorrow the only vehicles sold were EVs 80% penetration in 10 years would be ambitious. And of course, no new gas cars is not a scenario that is envisioned to happen for another 10-20 years.

I would also expect that in a "no new gas cars" situation, certain people, for rational (live in a rural area, collecting vintage cars) and irrational (self identity is tied to consuming fossil fuels) reasons would collect and maintain a fleet of increasingly older vehicles. While I have no idea how big this group would be, it would absolutely be large in certain geographic areas (say, rural Alberta).

[danbrotherston]

Most multi-residential buildings have little spare capacity, and I suspect the neighbourhood substations may have the same challenge.

Something that could help is that it would be the neighbourhood substations that I would hope to need more demand for charging at night rather than multi-residential buildings. That depends a lot on urban design. I've read about condos in Montreal where people just leave the condo by car in the morning and come back in the evening, so no better than suburbs. But with our fewer-spaces-than-units condos that are coming up, I can hope that there is also less of that car commuter usage pattern.

Yeah, this is true in Montreal, it's also true in other places. There are high density residential apartment buildings in London ON that don't even have so much as a sidewalk to them. Everyone in those buildings is required to leave by car. It's probably a pessimal built form. Dense, but with zero of the usual benefits of density, and with all of the disadvantages of sprawling development in terms of transportation and paved area.

It is possibly the case that there are condos in urban areas with good transit where everyone drives, but that is at least a fairly easily solvable problem with incentives.

If tomorrow, everything we ever built was electric, it would still take 10-15 years to convert the existing fleet. More likely everything going electric means a 30-50 year timeframe.

This is more than enough time to transition.

I'm tired of this defeatism. We need to change our society, we CAN change our society. We need to stop making excuses for NOT changing our society.

Average age of car in the Canadian fleet seems to be somewhere between 9 and 12 years. There are going to be network effects too so I wouldn't be surprised at no gas cars left in 10 years.

I've seen some news about airplanes going renewable too. "Sustainable aviation fuels" might be a thing. I'm somewhat skeptical for various reasons, but we'll see.

I am generally really also skeptical of people saying "oh but we ship things across the world" without actually considering how low-emissions it is to ship. Yes, it would be better to emit less sulfur, but that's not as bad as CO2 (says the person who lives far from a port). There are of course also benefits to having less disposable crap, and of making things in Canada, sure.

I suppose there is the possibility of sustainable fuels, but I'd be surprised if that was viable before electric planes. There are already a couple models of electric planes, but they are generally small general aviation and training aircraft, the technology currently does not permit a large commercial passenger plane to be electric.

Ironically, the main climate change (not environmental) benefit of manufacturing in Canada might be that the cost is much higher so consumption would be lower. Of course, this is where it is a cultural shift, rather than a technology one. We as a society are used to the idea that consuming more is a good thing in all cases and circumstances. Hell, obesity rates if nothing else shows that is not true, and that quality should matter. [1]. Of course, Canadian manufacturing is also likely to use cleaner power, so it's hard to say if that would be a great improvement in CO2 emissions or not.

Do full EV's charge themselves to a degree as well?  Perhaps increasing that capacity should be the focus, to have an EV that rarely needs to be plugged in.

I don't think the physics works out, since perpetual motion isn't a thing and we lose a lot of kinetic energy to air friction and road friction. Certainly cars are more aerodynamic than in the past, and I bet that regenerative braking also helps reduce particulate emissions from braking. But I'd rather redesign cities to require less driving.

I'm not sure what is meant by charging themselves. A few production EVs have solar panels, but AFAIK none of those solar panels are used to charge the main propulsion battery, and are only used to run auxillary equipement.

In terms of running a larger capacity, with less frequent charging, this to me, is suboptimal. A main benefit in the user experience of an EV is that one does not need to monitor and periodically refuel the vehicle, you just plug in on arrival home and you never need to think about it or make a trip to a fueling station. (This is of course, not including long trips which require planning and charging). If some background elecronics want to monitor charge levels and optimize power consumption by not charging all the time, while maintaining sufficient range for my daily trips, I suppose that would be fine--that is the smart grid idea after all. Tesla's already moderate charging, by not fully charging the batteries unless requested, but this is to increase battery longevity rather than for any power consumption goals.

[1] It did occur to me yesterday to ask myself a question about the current belief that "old was better" seen in "they don't make them like they used to". While many modern appliances do have many improvements, certainly in safety but also sometimes in usability and efficiency, it is the case that most of the R&D effort has gone into making them cheaper and easier to manufacture and not to improving longevity (in some cases, equipment has been made intentionally less durable). So I was wondering if that view is more prevalent now than it used to be, and if that was driven by an actual decrease in the durability and reliability of modern consumer (and some industrial) equipment.

In any case, I don't have a good answer for what economic policies would help drive manufacturerers to build more reliable and repairable equipment. Certainly it is contrary to the interests of manufacturers right now.

(And yes, I am absolutely creating a new tangent to this original tangent).

[tomh009]

As long as there is enough electricity (being generated) for everyone, that makes sense.

What does that mean? A wealthier society can afford to generate more electricity than a poor society. In 99% of societies that have existed to date, “enough” electricity was none at all; and even now, “enough” in most places is an amount that probably wouldn’t run my dishwasher.

There is no way around it: you can ration by price, letting each person decide how to use their resources; or you can pretend that a central planner is capable of figuring out what people need and want. Hint: they can’t.

This goes back to my original point: if we add millions of EVs, their power consumption may exceed the current spare capacity. And, if that's the case, we'll need to build additional generating capacity, whether zero-carbon or not. (And I do believe building additional generating capacity would need to be centrally planned.)

[plam]

9-12 is the average so many cars are older than that. I will personally be surprised if we reach 80% EV market share (of new cars, not the entire fleet) within 10 years. And in remote areas it will take much longer.

As for ships, bunker fuel is getting much cleaner. It was traditionally 4.5% sulphur, dropped to 3.5% in 2012, and it has been 0.5% since last year. Still higher than automotive fuel, but far cleaner than it used to be.

I agree, even if tomorrow the only vehicles sold were EVs 80% penetration in 10 years would be ambitious. And of course, no new gas cars is not a scenario that is envisioned to happen for another 10-20 years.

I would also expect that in a "no new gas cars" situation, certain people, for rational (live in a rural area, collecting vintage cars) and irrational (self identity is tied to consuming fossil fuels) reasons would collect and maintain a fleet of increasingly older vehicles. While I have no idea how big this group would be, it would absolutely be large in certain geographic areas (say, rural Alberta).

Sounds if we had no new gas cars being sold tomorrow we'd expect 50% EVs in 10 years.

People used to go to the hardware store to buy gas. It will probably be harder to buy gas generally. Might be like buying diesel today.

This goes back to my original point: if we add millions of EVs, their power consumption may exceed the current spare capacity. And, if that's the case, we'll need to build additional generating capacity, whether zero-carbon or not. (And I do believe building additional generating capacity would need to be centrally planned.)

Because our economy relies on relentless growth there is some planned growth in generating capacity anyway, but I guess the point here is that the growth rate will be higher than planned. On the other hand, solar is getting cheap really fast.

Good to hear that bunker fuel is also getting cleaner.

Yeah, this is true in Montreal, it's also true in other places. There are high density residential apartment buildings in London ON that don't even have so much as a sidewalk to them. Everyone in those buildings is required to leave by car. It's probably a pessimal built form. Dense, but with zero of the usual benefits of density, and with all of the disadvantages of sprawling development in terms of transportation and paved area.

It is possibly the case that there are condos in urban areas with good transit where everyone drives, but that is at least a fairly easily solvable problem with incentives.

One problem with Western society is techno-solutionism. EVs don't solve everything. Urban planning doesn't solve everything. Vaccines don't solve everything. If the problem is big enough (eg climate change) it has to be all the bullets, not the silver bullet.

I understand that it's not just incentives but dis-incentives. Parking and driving actually have to be inconvenient. They are definitely convenient in cities now.

EV capacity: for short trips you don't need huge capacity, but it may help with range anxiety (which isn't necessarily rational).

"They don't make them like they used to": there's always selection bias here. We only see the ones they made like they used to and which didn't explode or catch on fire or otherwise fail. And planned obsolescence was not at all new; it was defeated by imports which didn't buy in to that. I think they absolutely make cars better than they used to for instance. Washing machines is a mixed bag: they're less durable but they work better. And I think it's got to be right-to-repair legislation.

[Coke6pk]

I don't think the physics works out, since perpetual motion isn't a thing and we lose a lot of kinetic energy to air friction and road friction. Certainly cars are more aerodynamic than in the past, and I bet that regenerative braking also helps reduce particulate emissions from braking. But I'd rather redesign cities to require less driving.

I'm not sure what is meant by charging themselves. A few production EVs have solar panels, but AFAIK none of those solar panels are used to charge the main propulsion battery, and are only used to run auxillary equipement.

In terms of running a larger capacity, with less frequent charging, this to me, is suboptimal. A main benefit in the user experience of an EV is that one does not need to monitor and periodically refuel the vehicle, you just plug in on arrival home and you never need to think about it or make a trip to a fueling station. (This is of course, not including long trips which require planning and charging). If some background elecronics want to monitor charge levels and optimize power consumption by not charging all the time, while maintaining sufficient range for my daily trips, I suppose that would be fine--that is the smart grid idea after all. Tesla's already moderate charging, by not fully charging the batteries unless requested, but this is to increase battery longevity rather than for any power consumption goals.

(And yes, I am absolutely creating a new tangent to this original tangent).

My comment was do non-hybrid EV's "recharge themselves" to supplement  the plug in, not replacing it... I realize perpetual motion isn't a thing.

My thoughts were more of a plug in and charge for 1/2 hr as opposed to a 3 hr charge if the battery wasn't drained over the course of a trip because it was regenerated by braking, wheel turning etc.

Coke

[tomh009]

My comment was do non-hybrid EV's "recharge themselves" to supplement  the plug in, not replacing it... I realize perpetual motion isn't a thing.

My thoughts were more of a plug in and charge for 1/2 hr as opposed to a 3 hr charge if the battery wasn't drained over the course of a trip because it was regenerated by braking, wheel turning etc.

Regenerative braking is how non-plugins charge. And, yes, plug-ins and BEVs do the same, it's just less significant given the much larger batteries.

Charging from normal wheel motion would be counterproductive as it would increase the energy required for keeping the car in motion.