Tesla Virtual Power Plant in CA

[cwied]

It will not be meter data but PW data. You could technically have PV sending back power outside of PW. But none the less, it's about what the PWs are doing for VPP.

Frankly I think this is just someone at Tesla doing a bad job of describing the process. The agreement uses the term "ELRP A.4" to describe the program we're joining. This is a program that is available to more than just Tesla customers. I can't find any mention in the standard ELRP documentation that only the storage device participates in the VPP (nor would it make sense, since even those without any solar or batteries can participate in ELRP).

Here is the section on calculating the incentive from the ELRP documentation (https://elrp.olivineinc.com/wp-content/uploads/sites/5/2022/05/ELRP-Group-A-Terms-and-Conditions.pdf):

Steps for calculating a participant’s ELRP baseline
1) Calculate the Energy Baseline (EB) – A service account must have at least 10 days of interval meter data available in PG&E’s billing system to have a valid baseline. Only the hourly average usage for the hours included in the event will be included in the EB determination.
2) The EB and Adjusted Energy Baseline (AEB) will all be calculated at the service account level. The EB and AEB will be calculated on an hourly basis using the average of the preceding similar days8 , excluding those days when the customer: (1) was subject to an ELRP event, or (2) an event for a dually enrolled DR program, if applicable, or (3) was subject to a grid outage. Critical Peak Pricing event days will not be excluded from the similar days in order to capture incrementality for ILR.
3) Calculate the Day-Of Adjustment Value (DOAV) – A DOAV shall not be less than 1.00 or greater than 1.40. The DOAV is a ratio of (a) the average load of the first three hours of the four hours prior to the event to (b) the average load of the same hours from the days selected in accordance with Step 2 above.
4) Calculate the Adjusted Energy Baseline (AEB) – A service account’s AEB for an ELRP event is calculated by multiplying the EB by the DOAV.

The reference to the interval meter implies to me that that is the data that will be used for calculation. In addition, this is the description on how exports work:

If the customer has a Rule 21 interconnected device with export capability and permit, the customer may choose to count exported energy, up to their export rating, in ILR. In that case, the applicable ELRP baseline is modified to account for exported energy during nonevent days and count exported energy in ILR.

I am assuming that the site export limit setting that Tesla updated for the VPP corresponds to the export rating that includes both Powerwall and solar exports.

 

[cwied]

That's my main concern as well, other entities having more control of my batteries.

One important thing to note about the California program is that unlike the East Coast VPPs, you still control the backup reserve for these events, so you still can override the discharge any time you want.

I'm also a net exporter so gaming use/VPP vs. my TOU is probably not worth the hassle at this point. If anything, I've raised my reserve and started to pull from the grid off hours to cycle my batteries less during the summer now.

The VPP is not an NEM credit, it is a separate payment from Tesla, so it's a way to get money back even if you're a net exporter. In addition, if you discharge less during the event hours on non-event days, you'll get a higher payment due to the higher baseline.

 

[aesculus]

@cwied : Thanks for the info. At this point I give up. I had a hard enough time figuring out how NBCs played into the ESS/Solar game. This is far more complex. And Teslas's estimate of me getting $180 for a year of participation (never mind that they did not factor in any losses from my regular peak solar returns) means this program is not for me. It would take more than $180's worth my time to just try to account for what's going on differentially between my current give back and this program. And the chance of anyone/thing making a mistake is huge based on the reading above. Its crazy that they were even allowed to come up with such a program for consumers vs energy load reduction companies.

I too will now be sitting on the sidelines lurking to see what comes of all of this. But since participants are not going to be paid for a great while, that shell game may be hard to follow.

 

[cwied]

I too will now be sitting on the sidelines lurking to see what comes of all of this. But since participants are not going to be paid for a great while, that shell game may be hard to follow.

Fair enough, although I think that the program is structured so there's no downside risk for participants. Any extra discharge during events is compensated at $2.00/kWh and worst case can only cause additional NBCs. Since the events are always during peak, they will generate the most NEM credits possible.

I too am curious to see what the end result is and do wonder why Tesla chose to keep the details so vague. I am logging enough data to verify the payment calculation when it comes out.

 

[charlesj]

I just hope that I don't have to hound Tesla for that check if it doesn't get here by the end of March as I had to for that 4% rebate for missing the 30% ITC.

 

[astrorob]

@charlesj hey at least they agreed to give you that 4%. they told me to pound sand and i never got it.

given that i'm exporting most of my stored solar at peak time already, it seems like i won't ever get paid much for a VPP event; probably $10 or less. i wonder if it makes more sense to switch over to one of those demand reduction programs like OhmConnect since i'm mostly off the grid past 4PM. anyone have experience with those programs? or are they structured similarly such that if they see that i'm always off the grid in the afternoon then there will be no payment during an event?

 

[Redhill_qik]

@cwied : Thanks for the info. At this point I give up. I had a hard enough time figuring out how NBCs played into the ESS/Solar game. This is far more complex. And Teslas's estimate of me getting $180 for a year of participation (never mind that they did not factor in any losses from my regular peak solar returns)

It isn't that complex, you will get at least what you exported to the grid during the event period * $2.00. There are no losses from your regular solar returns as everything is still under NEM as well.

You may get more depending on what you used in the comparison days (the prior 10 weekdays if the event is a weekday or the prior 4 weekend days if the event is on a weekend).

1. During the event what was your grid usage kWh? (Event Grid Usage or EGU [my term as it isn't otherwise defined])
   1. Mine was -17.61 kWh from PG&E Green Button data which matches the Tesla App data
2. During the comparison days during the same period what was your average grid usage? (Energy Baseline or EB)
   1. Mine was -0.79 kWh from the PG&E Green Button data
3. Calculate the Incremental Load Reduction or ILR = max(0,EB) - EGU
   1. My ILR = max(0,-0.79) - -17.61 = 17.61 kWh
4. Compensation is ILR * ECR (ELRP Compensation Rate)
   1. Mine =17.61 kWh * $2.00/kWh = $35.22

If your EB is positive then the Day Of Compensation Value (DOAV) multiplier which has a range of 1.0-1.4 based on the 3 hours before the event period might apply, but this is very unlikely as the grid usage should be negative (exporting) which makes it 1.0, so just ignore this possible increase to the EB value.

 

[Redhill_qik]

@charlesj hey at least they agreed to give you that 4%. they told me to pound sand and i never got it.

given that i'm exporting most of my stored solar at peak time already, it seems like i won't ever get paid much for a VPP event; probably $10 or less. i wonder if it makes more sense to switch over to one of those demand reduction programs like OhmConnect since i'm mostly off the grid past 4PM. anyone have experience with those programs? or are they structured similarly such that if they see that i'm always off the grid in the afternoon then there will be no payment during an event?

If your grid usage for the same period during the comparison days is negative this does not count against you. If your average during the previous days was -30 kWh and during the event you exported -31 kWh then you get 31 kWh * $2.00 or $62.00.

People that participated in the Beta program last year were paid by Tesla automatically from the reports here.

 

[wwu123]

@charlesj hey at least they agreed to give you that 4%. they told me to pound sand and i never got it.

given that i'm exporting most of my stored solar at peak time already, it seems like i won't ever get paid much for a VPP event; probably $10 or less. i wonder if it makes more sense to switch over to one of those demand reduction programs like OhmConnect since i'm mostly off the grid past 4PM. anyone have experience with those programs? or are they structured similarly such that if they see that i'm always off the grid in the afternoon then there will be no payment during an event?

You'd still have to game the system with Demand Response Programs like OhmConnect, the energy baseline calcs are well-known for OhmConnect, and likely the ERLP calcs are the exact same. If you're already exporting daily, you won't gain much, but seems like you wouldn't lose much either if you do nothing for the events, and you can choose to game it before events selectively.

It seems mainly a difference of event profiles, between VPP <20 events/year of multi-hour duration at $2/watt, or OhmConnect more like 100 events/year typically 1 hour duration at $1-1.40/watt.

 

[charlesj]

If your grid usage for the same period during the comparison days is negative this does not count against you. If your average during the previous days was -30 kWh and during the event you exported -31 kWh then you get 31 kWh * $2.00 or $62.00.

People that participated in the Beta program last year were paid by Tesla automatically from the reports here.

For this time of year I don't use anything from the grid from 6PM-9PM as battery is fully charged. I will keep an eye out for future events and keep track.

 

[astrorob]

If your grid usage for the same period during the comparison days is negative this does not count against you. If your average during the previous days was -30 kWh and during the event you exported -31 kWh then you get 31 kWh * $2.00 or $62.00.

People that participated in the Beta program last year were paid by Tesla automatically from the reports here.

well that is good news then. but i’ve read different things here, so i guess i’m just going to have to go to the program documents and figure it all out.

typically i would export about 30kwh from 3-8pm. but during this event, because it started at 6pm and it was so cloudy that all the solar went into the powerwall, i only exported 16kwh from 6-9pm. upthread i think someone else said i’d only get paid for the 8-9pm export under vpp since typically i am pulling from the grid at that time.

 

[Redhill_qik]

well that is good news then. but i’ve read different things here, so i guess i’m just going to have to go to the program documents and figure it all out.

typically i would export about 30kwh from 3-8pm. but during this event, because it started at 6pm and it was so cloudy that all the solar went into the powerwall, i only exported 16kwh from 6-9pm. upthread i think someone else said i’d only get paid for the 8-9pm export under vpp since typically i am pulling from the grid at that time.

This is the relevant section of the detailed ELRP document that reclassifies the exported energy to ILR (Incremental Load Reduction) instead of counting against you as being part of your EB (Energy Baseline).

3.2.1.4 Settlement Calculations for Sub-Group A.4. (aka Virtual Power Plants)
The aggregator selected CPUC approved baseline for PG&E’s Capacity Bidding Program is utilized and modified to account for exported energy, to the extent allowed by a site’s Rule 21 export permit, during non-event days and count exported energy in ILR

 

[cwied]

This is the relevant section of the detailed ELRP document that reclassifies the exported energy to ILR (Incremental Load Reduction) instead of counting against you as being part of your EB (Energy Baseline).

I interpret this a little differently. It says the program is modified to account for the exported energy during non-event days, meaning to calculate the baseline. There is no mention that that exported energy is then ignored if the net is negative. It seems to me that PG&E just sees the meter amounts, so exported energy by their definition only happens when generation (from Powerwall and solar) exceeds load.
My intuitive argument is as follows: the purpose of the program is "load reduction" from the grid's perspective, and that's what they are paying incentives for. If you normally export 31 kWh but export 32 kWh during the event, that is only a reduction of 1 kWh load on the grid, not 32 kWh.

 

[swedge]

I interpret this a little differently. It says the program is modified to account for the exported energy during non-event days, meaning to calculate the baseline. There is no mention that that exported energy is then ignored if the net is negative. It seems to me that PG&E just sees the meter amounts, so exported energy by their definition only happens when generation (from Powerwall and solar) exceeds load.
My intuitive argument is as follows: the purpose of the program is "load reduction" from the grid's perspective, and that's what they are paying incentives for. If you normally export 31 kWh but export 32 kWh during the event, that is only a reduction of 1 kWh load on the grid, not 32 kWh.

One other detail, though. Do they do it hour by hour, or include import/export for just the period of the event. And does the event mean the period when we export (6-9pm) or the entire event (5am - 9pm).

I ask because my Export Everything starts full power ~5 kW export at 4 pm, but then discharge slows before 6 to only cover house draw till 9. So my baseline from 6 to 9 is mostly zero, but my baseline from 4-6 is solar plus around 10kWh from PW minus a couple kWh for house load.In VPP mode, the PW export started at 6. So, I hope the word/aconym salad in those official explanations means the the 10 day average is of grid use from 6-9 pm.

Is anyone clear on this facet.

Also, I am not clear on Tesla's calculation vs the one PG&E describes/obfuscates. Does Tesla follow the same rules? Tesla has the PW data, rather than just the PG&E meter data, so different calculations are possible.

And here we used to think NEM was complicated... ;-)

SW

 

[cwied]

One other detail, though. Do they do it hour by hour, or include import/export for just the period of the event. And does the event mean the period when we export (6-9pm) or the entire event (5am - 9pm).
...
Also, I am not clear on Tesla's calculation vs the one PG&E describes/obfuscates. Does Tesla follow the same rules? Tesla has the PW data, rather than just the PG&E meter data, so different calculations are possible.

And here we used to think NEM was complicated... ;-)

SW

The agreement section on Tesla's VPP page references ELRP A.4, which is not Tesla-specific. That is the document I linked above. I don't believe that Tesla will be doing anything different than any other VPP aggregator would, so I don't think the PW data would be used. There is a telemetry requirement for generators over 10 MW, but it looks like that can be avoided by bundling 10 MW at a time.

My expectation is that Tesla will follow PG&E's rules as documented in the ELRP. I believe there is an aggregator choice as to whether to use a 10 day baseline or highest 5 in 10 day baseline, but it seems like Tesla has chosen 10 days based on their description.

Based on my experience with OhmConnect's demand-reduction program, to which this seems to be similar, only the period during which you export is counted. This would mean 6-9 pm. My expectation is your export from 4-6pm will not affect your baseline. I have a lower export limit, so I probably won't be compensated for the first 45 minutes of the event since my Powerwalls don't usually hit reserve until about 6:45pm.

 

[Redhill_qik]

I interpret this a little differently. It says the program is modified to account for the exported energy during non-event days, meaning to calculate the baseline. There is no mention that that exported energy is then ignored if the net is negative. It seems to me that PG&E just sees the meter amounts, so exported energy by their definition only happens when generation (from Powerwall and solar) exceeds load.
My intuitive argument is as follows: the purpose of the program is "load reduction" from the grid's perspective, and that's what they are paying incentives for. If you normally export 31 kWh but export 32 kWh during the event, that is only a reduction of 1 kWh load on the grid, not 32 kWh.

I had a completely different response written defending my position, but then I received a response to my ELRP contact form submission which changes a lot of what I thought were the governing rules.

The reply said that the metering isn't the PG&E SmartMeter, but the Tesla Powerwall meter, so the house load isn't a factor. What is consider for compensation is the change in the battery behavior during the event hours. So, say the battery during the 10 prior weekdays exported nothing during 6:00-9:00pm (maybe because it exported everything from 4:00-6:00pm) and during the event the Powerwall holds off until 6:00pm and discharges 20 kWh then you would get all 20 kWh compensated. But, if the battery was averaging 8.0 kWh discharge during those 3 hours and during the event you discharged 20 kWh you would be compensated for only 12 kWh.

I've asked for some clarification on how 3.2.1.4 is being applied and exactly what is being modified. Until I get an answer on that I'm not going to say anything more on this topic.

 

[cwied]

The reply said that the metering isn't the PG&E SmartMeter, but the Tesla Powerwall meter, so the house load isn't a factor. What is consider for compensation is the change in the battery behavior during the event hours. So, say the battery during the 10 prior weekdays exported nothing during 6:00-9:00pm (maybe because it exported everything from 4:00-6:00pm) and during the event the Powerwall holds off until 6:00pm and discharges 20 kWh then you would get all 20 kWh compensated. But, if the battery was averaging 8.0 kWh discharge during those 3 hours and during the event you discharged 20 kWh you would be compensated for only 12 kWh.

That's very interesting. I am surprised Tesla managed to negotiate using their data. What's interesting is that the discharge limit does take into account house load (and solar generation). If what they say is true, it would mean you could get compensated for charging an EV from the Powerwall during the event.

 

[sunwarriors]

One important thing to note about the California program is that unlike the East Coast VPPs, you still control the backup reserve for these events, so you still can override the discharge any time you want.

The VPP is not an NEM credit, it is a separate payment from Tesla, so it's a way to get money back even if you're a net exporter. In addition, if you discharge less during the event hours on non-event days, you'll get a higher payment due to the higher baseline.

That's good that you can still set a higher reserve. For me, during these hot summers, I simply don't have enough ESS to last a full power outage overnight if I drained faster than I already do so if I export faster till my reserve from 6-9pm, I'd hit my reserve well before the window is finished (I now barely make it to me reserve at 9pm as it is) and I'd have to pull from the grid anyways at full TOU rates (59/69 cents) so I'd probably not save as much as other folks with either more storage or are ok with draining batteries down to a much lower reserve.

Maybe like having companies know your bank account number, I'd rather just not let more companies have me in their system I suppose. The $$ still isn't much (to me) relative to everything else since I don't have that much ESS anyways and I'd probably rather skimp somewhere else.

Maybe still minor, but there is still the added "use/wear/tear" of more charge/discharge cycles (like renting out your cars on Turo) which adds incremental wear/tear.

 

[h2ofun]

That's good that you can still set a higher reserve. For me, during these hot summers, I simply don't have enough ESS to last a full power outage overnight if I drained faster than I already do so if I export faster till my reserve from 6-9pm, I'd hit my reserve well before the window is finished (I now barely make it to me reserve at 9pm as it is) and I'd have to pull from the grid anyways at full TOU rates (59/69 cents) so I'd probably not save as much as other folks with either more storage or are ok with draining batteries down to a much lower reserve.

Maybe like having companies know your bank account number, I'd rather just not let more companies have me in their system I suppose. The $$ still isn't much (to me) relative to everything else since I don't have that much ESS anyways and I'd probably rather skimp somewhere else.

Maybe still minor, but there is still the added "use/wear/tear" of more charge/discharge cycles (like renting out your cars on Turo) which adds incremental wear/tear.

I have zero desire to have folks get into my stuff anymore than they already are. The small amount of money is not worth the risk. If calif really cared about stuff like this, NEM3 would not be doing what it might.

 

[Dave[N]]

I noticed during the VPP event, that I was sending exactly 4 kW to the grid for the full 3 hours.
That is the site export limit set in my gateway, matching my 4 kWp Solar.
If PG&E needs more energy, why not pull more, up to my reserve limit?
Since I have 2PW, presumably the PW could sustain 10 kW, and send that minus the house load during the event.
I get that the grid may not want us exporting normally at full boar, and could have limits based on local equipment, but it makes sense to have a 2nd export limit when its to their (PG&E's) advantage.
It would certainly let them source more MW, without needing more users opting in.