Can demand and small resources act like a power plant? Demand response, EV charging, virtual power plants and local trading

Shift a town’s load, pool small devices into a virtual power plant, and let neighbours trade

A7 of Energy Systems, Step by Step: shift a small town’s electricity use to cheaper hours and tell energy shifted from energy saved; pool batteries, thermostats, EV chargers and farm pumps into a virtual power plant and see what each owner earns and where a feeder’s limit stops it; then let neighbours trade.
  • Analyst · L2
  • 30 min guided · about 30 min practice
  • After lesson 9 and A4

Prairie Town's flexible demand on 9 June 2025 (illustrative)

  • None: the town still uses the same 10 MWh, at 14:00 instead of 20:00. Spare sun replaces the peaker, so the running cost falls $1,000 and the plants' CO₂ 5.5 t (metric tonnes).
  • At 5%, 99 MWh move from gas hours into spare-sun hours: the running cost falls $5,995 and the plants' CO₂ falls 42.8 t. Energy saved: 0 MWh.
  • The attributed goal moves 161 MWh, 62 MWh more, and books 22.0 t less attributed CO₂, against 17.5 t. The plants emit the same 42.8 t less: the extra moves swap hours with the same plant on the margin.
  • 21.3 MW for three hours: the peaker runs 43 MWh less, but 45 MWh of use comes back after 23:00. The town saves $3,360 that day.
  • All at once: Feeder 7 carries 3.25 MW at 23:00, over its 3.0 MW rating by 0.25 MW.
  • Staggered: Feeder 7 peaks at 2.58 MW after the event, under its 3.0 MW. The town saves more too, $3,576 instead of $3,360, as the peaker stays off at 23:00.
  • At 5.0¢, the seller gains $8 a month and the buyer $8, while other customers pay $16 more for plants and wires. The plants run exactly as before.
The virtual power plant from 20:00 to 23:00, with everyone back at once at 23:00 and a coordination cost of $30 per device or farm.
Device In the pool Owner earns a year
Home batteries 5.1 MW $127 per home battery
Smart thermostats 6.7 MW $6 per thermostat
EV chargers 8.4 MW $90 per EV charger
Dairies' pumps 1.1 MW $11,566 per dairy

With JavaScript on, you move the town's use yourself, enrol devices in the pool, set its coordination cost, restart a feeder and set the neighbours' price.

Illustrative: Prairie Town is fictional; its flexible share, devices, values, coordination cost, Feeder 7 and neighbours are invented.
Real hourly patterns: demand, wind and sun follow 9 June 2025 on the Southwest Power Pool grid (EIA-930, accessed 2 November 2026), the day of lesson 9 and A4.
Boundary: CO₂ is in metric tonnes (t), counted at the plants only; the plants run cheapest first, with no ramp limits.

About 30 guided minutes. Practice, about 30 more: the practice set under Check.

What you just did, as diagrams

Shifted, not saved
Moving use to another hour keeps the energy; with extra energy needed it takes more.
Why it still pays
The moved MWh are made by cheaper, cleaner plants: spare sun instead of the peaker.
Attributed against emitted
Average factors can book a cut that the plants never make; marginal ones follow the plants.
A virtual power plant
An aggregator pools many small devices and offers their cut as one resource.
Who earns
Value per kW is similar; a fixed cost per device eats small devices' share.
The wires
Use that all comes back at once can overload a feeder, a local distribution line; a staggered restart fits.
Local trade
It moves money, not electricity: the kWh flow the same way either way.
Go deeper · AnalystFour numbers a shift report should keep apart

A 5% flexible share in Prairie Town on 9 June 2025 (CO₂ in metric tonnes, t):

Without losses Cheapest Lowest attributed CO₂
Moved out 99 MWh 161 MWh
Energy saved 0 MWh 0 MWh
Running cost −$5,995 −$5,995
Plants’ CO₂ −42.8 t −42.8 t
Attributed CO₂ −17.5 t −22.0 t
With 10% extra energy Cheapest Lowest attributed CO₂
Moved out 90 MWh 150 MWh
Energy saved −9 MWh −15 MWh
Running cost −$5,634 −$5,396
Plants’ CO₂ −39.5 t −37.3 t
Attributed CO₂ −16.7 t −20.8 t
  • Moved against saved. A shift moves MWh; only using less saves them.
  • Attributed against emitted. Attributed CO₂ counts each MWh at its hour’s average factor; the plants’ CO₂ is what they burn, re-dispatched.
  • Which question. Inventories use the first; asking what a change does to the plants needs the second (case file CF6).
Go deeper · AnalystBaselines: measuring use that did not happen
  • A cut is a difference. It is what a home would have used minus what it used; the first number is never metered.
  • A baseline. Programs estimate it from similar recent days, often adjusted to the morning of the event.
  • Errors both ways. A baseline set too high pays for cuts that never happened; one set too low misses real ones.
  • Another way. Compare enrolled homes with a similar group that was not called (M14).
Go deeper · ModelerThe shift as a linear program

Take use \(d_t\) in hour \(t\) (MWh), flexible share \(s\), extra energy \(\ell\) and users’ cost \(u\) per MWh moved. Decide \(x_t \ge 0\) (moved out) and \(y_t \ge 0\) (added), with the gas plants \(c_t\) and \(p_t\) (MWh):

\[\min \sum_t \left(40\, c_t + 100\, p_t\right) + u \sum_t x_t\]

  • Rules. \(x_t, y_t \le s\, d_t\); \(\sum_t y_t = (1 + \ell) \sum_t x_t\); each hour balances, with \(0 \le c_t \le 150\) and \(0 \le p_t \le 100\), and spare sun may be spilled.
  • The page’s rule. It moves use from the dearest hour to the cheapest, one straight piece of the merit order at a time. For every setting on the page it reaches this program’s minimum (checked with HiGHS).
  • Ties. Hours with the same price can trade places, so the hours moved are one choice among equally cheap ones; the cost and the plants’ CO₂ are the same in all of them.
  • Next. Carbon in models comes in M6; a microgrid that runs through an outage in M5.

Check yourself

Myth or fact? “Moving electricity use to another hour reduces the total electricity used.”

  • Myth
  • Fact
Answer

Myth. The 5% cheapest shift moves 99 MWh and saves 0 MWh; with 10% extra energy the town uses 9 MWh more. Using less is a different response: the pool's warmer homes skip 4 MWh an event.

Myth or fact? “Moving use to hours with a lower average CO₂ factor always cuts what the plants emit.”

  • Myth
  • Fact
Answer

Myth. The attributed goal moves 62 MWh more than the cheapest shift and books 4.5 t more attributed savings, yet the plants emit the same: those moves swap hours with the same plant on the margin.

Myth or fact? “Every small device is worth enrolling.”

  • Myth
  • Fact
Answer

Myth. At $40 a device for coordination, a thermostat's owner loses $4 a year, while a dairy still earns $11,556.

Myth or fact? “A virtual power plant can overload a feeder after the event is over.”

  • Myth
  • Fact
Answer

Fact. Released at once, Feeder 7 carries 3.25 MW at 23:00 against its 3.0 MW rating; a staggered restart keeps it at 2.58 MW.

Air-conditioned homes are called on the month's hottest evening. Their baseline, an estimate of what they would have used, averages ten cooler, otherwise similar weekdays at the same hours. How does the measured cut compare with the real one?

  • Smaller
  • Larger
  • The same
Answer

Smaller: in the hotter event hours the homes would have used more than on the cooler days, so the baseline is too low. Programs may adjust it for weather or for the same day's earlier use.

Practice · about 30 minutes on your own

  1. Shifted or saved? A program moves 10 MWh from 20:00 to 03:00, when the combined-cycle plant is on the margin. What does the town save in energy, money and CO₂?

    Answer

    Energy: none. Money: $600 (the peaker's $100 for the combined-cycle plant's $40, per MWh). CO₂: 1.8 t (0.55 − 0.37 t per MWh).

  2. Extra energy. With 10% extra energy needed, the 5% cheapest shift moves 90 MWh. How much more energy does the town use, and why does it still save money?

    Answer

    9 MWh more (99 MWh in for 90 MWh out). Each MWh leaves a $40 or $100 hour and arrives in a $0 one, so the running cost still falls $5,634.

  3. The thermostat's year. One thermostat cuts 1.2 kW and 70% of homes take part. At $50 per kW a year for being ready, what is that part of its value?

    Answer

    1.2 × 0.7 = 0.84 kW × $50 = $42. With the energy it saves, $45 a year.

  4. Where coordination bites. At what coordination cost per device does a thermostat's owner earn nothing?

    Answer

    At $36 a year: 80% of its $45 value. A dairy earns $11,566 at $30.

  5. The feeder. Feeder 7 is rated 3.0 MW; released at once, its load reaches 3.25 MW at 23:00. Name two fixes.

    Answer

    Stagger the restart over several hours, or enrol fewer chargers on this feeder; a feeder-aware aggregator plans with the limit.

  6. The neighbours' trade. At 5¢ a kWh, 168 kWh a month change hands. Who gains and who pays?

    Answer

    The seller $8, the buyer $8; other customers pay the $16 of plant-and-wire charge the trade skips. The kWh and the plants do not change.

Numbers, sources and what’s next

Numbers to remember

Number What it means Source
33,272 MW · 6.5% Demand response enrolled in the seven U.S. wholesale markets in 2024, and the share of their peak demand it could meet. FERC staff, 2025 Assessment of Demand Response and Advanced Metering, pp. 2, 14–15 (PDF) · December 2025; data for 2024
968 MW · 1.8% Demand response in the Southwest Power Pool market, Prairie Town's real-data home, in 2024, and its share of SPP's peak demand. FERC staff, 2025 Assessment of Demand Response and Advanced Metering, pp. 15–16 (PDF) · December 2025; data for 2024
100 kW FERC Order No. 2222 requires the regional grid operators it regulates to let aggregations of small resources into their markets, with a minimum size of no more than 100 kW. FERC, Order No. 2222, p. 138 (PDF) · issued 17 September 2020

Keep going

Back to the course map

In the real world

What a shift is worth depends on how it is counted.

In the reported scenarios, reductions in attributed emissions were on the order of one to two percent of regional totals or less.

Reductions are computed under average-emission-intensity accounting with perfect foresight; they are not estimates of the consequential (marginal) system response.

The full case: case file CF6 (how much shifting load to cleaner hours can cut emissions, and when).

From the course author's research: in the modeled supply chains, each open factory or warehouse could run its own microgrid, with on-site wind, solar and storage, an energy balance every hour, and energy bought from and sold to the grid.

Because the transactive energy market brought a new revenue stream, locations with stronger renewables became competitive to establish facilities despite higher fixed and logistics cost.

Holds when: stylized networks with synthetic distances and costs, candidate sites in California, Arizona and Nevada, and one year, planned week by week for products and hour by hour for energy. Leaves out: grid outages, which the studies do not model.

Sources and further reading

Spotted an error, or stuck? Write to hlli@ksu.edu. Corrections are made in place and dated here.

Text licensed CC BY 4.0, code MIT: teachers and students may reuse and adapt it, with credit to Honglin Li.