On power markets, snow storms, and $16k power bills • Hacker News Board

Re: On power markets, snow storms, and $16k power bills

#131

Post by masklinn » Mon, Feb 22, 2021, 9:07 PM UTC

techsupporter wrote: post #65

masklinn wrote: post #52

Earlier quoted context omitted.

> Granted, if the cost savings were enough They probably are enough over long periods, or retailers wouldn't be profitable. However that assumes you hedge properly. If you treat your savings from "good times" as extra cash to spend you're going to get burned to the ground by the bad times. It also means you have to be extremely reactive, now you need a setup to quickly cutoff electricity if wholesale prices skyrocket…

> If you treat your savings from "good times" as extra cash to spend you're going to get burned to the ground by the bad times. Exactly, though to give a lot of customers of these plans credit, they don't have the extra cash. Griddy's own marketing says that "96% of the time," the wholesale rate is well under the average retail per-kWh rate. This is yet another one of those hidden costs of how being lower-income is v…

I would agree with that. Griddy specifically marketing towards poorer populations by touting savings is really objectionable at the best of times.

Re: On power markets, snow storms, and $16k power bills

#132

Post by fennecfoxen » Mon, Feb 22, 2021, 9:09 PM UTC

ahepp wrote: post #72

techsupporter wrote: post #46

Earlier quoted context omitted.

Exactly. In my experience, there are exactly two types of customers on the kinds of plans like Griddy offers: a) people who are heavily into cost-optimization and have spent the cash to automate their living quarters to take advantage of wholesale price swings; and b) people who are trying to stretch a dollar because a single dollar is all they have left. For people in that second category, the dollar broke and smack…

I used to have these guys (EDIT: energy dereg companies, not griddy specifically) knocking on my door regularly with a pen, asking me to sign a contract for real time pricing literally on the spot, because I could save a bunch of money on my bill I would be shocked if there is any energy deregulation company that doesn't resort to this, because it's the basic result of hiring contractors and paying them on commission

> knocking on my door regularly with a pen

How delightful! Here in New York we just have CleanChoice Power doing it via mail fraud.

(By "mail fraud" I mean that they do their best to dress it up as a something like power bill, except demanding a signature, to trick the unwary into signing it without reading.)

Re: On power markets, snow storms, and $16k power bills

#133

Post by filoleg » Mon, Feb 22, 2021, 9:09 PM UTC

hn_throwaway_99 wrote: post #56

filoleg wrote: post #35

Earlier quoted context omitted.

Which is why the parent comment says "can have a limit", not "must require a limit". That way, those who would rather get through a blackout without paying insane bills can have an option to do so, while those who care about their pipes freezing can pay to keep the electricity.

I still think this is totally unfeasible. Then all the news stories would just be "Fine print in electric contract shut off power for many, now facing thousands in water damage bills."

Totally agreed with you, it is probably completely unfeasible. I was just assuming we were talking about this in terms of ideal wants.

Re: On power markets, snow storms, and $16k power bills

#134

Post by thesteamboat » Mon, Feb 22, 2021, 9:16 PM UTC

thesuitonym wrote: post #128

jandrese wrote: post #26

It strikes me that Texas is suffering the same sort of problems that companies that switch to just-in-time delivery do. When everything is working they're more efficient and can out-compete their neighbors, but since they're running with little to no safety margin it only takes one incident to cause immediate disruption to the line. Of course when the consequence of a disruption is you miss your production targets fo…

The problem wasn't really the just-in-time nature of power generation--nearly all power grids operate this way, usually with little more than a flywheel on the generator to smooth out fluctuations. Sometimes you get a stored water battery, but that doesn't last more than a few hours at peak consumption. We don't really have the ability to store power in a way that's efficient enough to not rely on JIT generation. No,…

I don't think the parent was talking about JIT power generation but rather using JIT (in other industries) as a comparison to to talk about the efficiency-reliability tradeoff. The analogy here is cost savings of avoiding winterization make the power companies more competitive (more efficient) but less reliable.

You are, of course, right in the all specific details you've brought up vis-a-vis the winterizing and the nature of the power grid. I just think you missed the parent's point slightly.

Re: On power markets, snow storms, and $16k power bills

#135

Post by amluto » Mon, Feb 22, 2021, 9:19 PM UTC

> If you pair a retail energy provider like Griddy with an advanced suite of home efficiency technology, like programmable or price-responsive thermostats, you can have a building that automatically tailors its energy usage to the market environment

One major piece of this puzzle that is, IMO, missing, is the lack of any reasonable, standardized way for a hypothetical smart home to communicate with the grid. With a smart meter, a properly ZigBee certified (why? probably some silly reason involving “security”) end user device can get current usage data from a smart meter. (And this data is blatantly wrong in any nontrivial circumstance.)

But and end-user device has no way to negotiate pricing or load reduction with the grid short of going through some intermediary on the Internet. If I want to program my house to reduce consumption when prices are high, I can’t.

An an example, a car should be able to tell the grid that it wants to buy x kWh at a maximum power of y kW between now and 5 AM, and the grid should be able to tell the car when to buy it. I don’t mean Tesla negotiating this on behalf of itself or its customers - I mean the car or the house talking to the meter or the utility. No unnecessary or unavoidable intermediaries please.

It doesn’t help that PG&E does not accurately know the topology of its own grid. Oops.

Re: On power markets, snow storms, and $16k power bills

#136

Post by billylindeman » Mon, Feb 22, 2021, 9:19 PM UTC

ahepp wrote: post #66

>Those that chose Griddy, chose higher risk, which comes with high potential rewards but also high potential losses When I lived in the poor part of town, "energy deregulation" companies were always coming around, aggressively trying to get me to sign up on the spot. They never talked about risk, they just told me I could sign here to save $100s on my power bill. I asked them about the details and they just lied thro…

"This article seems to suggest the answer is a fully automated smart home, with some kind of AI to intelligently manage your power usage. Sounds awesome, but I don't think that's ever going to be a reality outside the valley."

I actually worked on a system like this https://powerley.com it's based in the Detroit area. The energy grid is going to face growing demand / supply issues over the next decade and I do believe systems like this will become more important to manage load.

Re: On power markets, snow storms, and $16k power bills

#137

Post by dctoedt » Mon, Feb 22, 2021, 9:21 PM UTC

jeffrallen wrote: post #90

woeirua wrote: post #54

The big takeaway from all of this should be the critical cyclic dependency of natural gas power plants. That is, if your primary source of power is natural gas, then you must have reliable power to the natural gas wells to produce natural gas. If the power becomes unreliable, then natural gas pressures in the pipelines drop to unsuitable levels to run large generators. As the generators trip due to adverse safety con…

Nuclear is not: it needs grid power to be able to run.

> Nuclear is not: it needs grid power to be able to run.

But the "grid power" needed to run the reactor's control systems, etc., can almost surely be generated by the reactor itself while it's online, right?

A naval nuclear reactor needs external power only to get started; when the ship is in port, this will typically be shore power from the grid, and either in port or at sea the external power could come from the onboard emergency diesel generators. When the reactor is started up, it will eventually "go critical," i.e., achieve self-sustaining fission. After that happens, one or more in-plant, steam-turbine electrical generators will be started and brought online to provide electrical power for the ship, including the reactor control systems, etc. [0] A similar arrangement seems likely to be used for civilian power plants. [1]

[0] e.g., https://fas.org/man/dod-101/sys/ship/eng/reactor.html

[1] e.g., https://www.elprocus.com/what-is-a-nuclear-power-plant-worki...

Re: On power markets, snow storms, and $16k power bills

#138

Post by phicoh » Mon, Feb 22, 2021, 9:23 PM UTC

nicoburns wrote: post #115

ahepp wrote: post #66

>Those that chose Griddy, chose higher risk, which comes with high potential rewards but also high potential losses When I lived in the poor part of town, "energy deregulation" companies were always coming around, aggressively trying to get me to sign up on the spot. They never talked about risk, they just told me I could sign here to save $100s on my power bill. I asked them about the details and they just lied thro…

> This article seems to suggest the answer is a fully automated smart home, with some kind of AI to intelligently manage your power usage. Sounds awesome, but I don't think that's ever going to be a reality outside the valley. It's not the answer yet. But I'd be very surprised if it wasn't the norm in developed nations in 10-20 years. It's an obvious response to inconsistent generation from renewables, and the techno…

I don't see how that would work. Suppose there is no wind for a couple of days, do you expect homes to shutdown everything and wait for wind to return?

Solar in a desert is easier, if you only have to cover the night then install a big enough battery. But a battry in every home is probably less efficient than a huge battery next to the solar panels.

So most likely, renewables come with storage. And the smart grid idea with mostly fade. Possibly with the exception of car chargers that charge faster or slower depending on conditions.

Re: On power markets, snow storms, and $16k power bills

#139

Post by brandmeyer » Mon, Feb 22, 2021, 9:24 PM UTC

ahepp wrote: post #66

>Those that chose Griddy, chose higher risk, which comes with high potential rewards but also high potential losses When I lived in the poor part of town, "energy deregulation" companies were always coming around, aggressively trying to get me to sign up on the spot. They never talked about risk, they just told me I could sign here to save $100s on my power bill. I asked them about the details and they just lied thro…

> I suspect you could get 99% of the effect by only passing a small price window to consumers.

I've been thinking about this problem for some years now, since I first became involved in microgrid controls. Rather than change the dynamics by limiting amplitude, I would change the dynamics by adjusting information latency.

Schedule customer pricing based on the day-ahead market clearing price instead of the spot price. End-customers are still exposed to price fluctuations for both good and ill. But now they can make decisions based on them well in advance. Momentary interruptions in communication no longer render the customer blind to the market price. Similarly, low-latency communications are not required for smart devices to take advantage of price changes. The most volatile portion of the market risk is also still borne by the utility.

Re: On power markets, snow storms, and $16k power bills

#140

Post by pierrebai » Mon, Feb 22, 2021, 9:26 PM UTC

Combining real-time data, market with speculators and a product that demand immediate, real-time delivery is a recipe for disaster.

When demand exceed supply, given the real-time nature of the product speculators can come in a take you to the cleaner if they had any chance to beat you to foretelling this would happen.

As I understand it, ERCOT puts a $9000 per MWh to put a limit on speculation, but then what happens is you have to take people off the grid.

The desire to "maximize efficiency" by surfing the demand curve with real-time data is not worth worth the expense of smart meters and risks. Dumb demand curves have another advantage: the utility is forced to over-estimate demand, creating a built-in safety margin.