Behind one meter: what should one home or one plant install?

Build a plan, test what could change it, and defend it in a one-page note

Capstone C0 of Energy Systems, Step by Step: choose what the Prairie Town home or the ridge wind farm should install, run every hour of 2025 on the Site Bench’s engine, and defend the plan with its NPV, its CO₂, a sensitivity chart and what the site cannot do alone, in a one-page note.
  • Analyst to Modeler · L2–L3
  • Capstone
  • 60 min guided · 45 min lab
  • After lesson 9 and A1 and F4 and SB0

The brief

What should one site install behind its meter: the Prairie Town home, or the ridge wind farm? You advise its owner.

  1. Start from: the owner's goal, in one line: the most money over the plan's life, or less CO₂ at a cost you state. Either can pass, with its trade-off in numbers.
  2. Hand in: a plan link and a one-page note; its sensitivity part lists every bar of the chart in numbers (Modeler version: one M4 comparison too).
  3. From: the Site Bench's engine, below, with every hour of 2025: solar, a battery and a heat pump, or an electrolyzer and a hydrogen tank.
  4. Done when: every check under the note is in place, and your note meets each "Look for" line of the self-check.

About 60 minutes: 25 to build and test the plan, 25 to write the note, 10 to check it.

Build and run your plan

Control Default plan Range
Site Prairie Town home one of: Prairie Town home, Ridge wind farm
Rooftop solar 6 kW 0 to 15 kW, in steps of 0.5 kW
Home battery 0 kWh 0 to 30 kWh, in steps of 0.5 kWh
Heat pump none (keep the gas furnace) one of: none (keep the gas furnace), standard heat pump, high-efficiency heat pump
Tariff flat price one of: flat price, time-of-use
Export credit 4 ¢/kWh 0 to 8 ¢/kWh, in steps of 0.5 ¢/kWh
Wind farm 150 MW 0 to 300 MW, in steps of 10 MW
Line to the grid 100 MW 10 to 300 MW, in steps of 10 MW
Battery (4 hours) 0 MWh 0 to 1,000 MWh, in steps of 10 MWh
Electrolyzer 0 MW 0 to 200 MW, in steps of 5 MW
Hydrogen tank 0 t 0 to 200 t, in steps of 1 t
Buyer takes at most 1,000 kg/h 0 to 5,000 kg/h, in steps of 100 kg/h
Hydrogen price 4 $/kg 1 to 8 $/kg, in steps of 0.25 $/kg
Power contract fixed $50/MWh one of: fixed $50/MWh, time-of-use
Discount rate (real) 4 % 0 to 10 %, in steps of 0.5 %
Equipment cost 100 % of sourced 50 to 150 % of sourced, in steps of 5 % of sourced

With JavaScript on, you change the plan, run it in your browser and copy a link to it. Without it, write the note on the default plan: its link, results and sensitivity are below.

The default plan, without JavaScript

Its plan link, for your note: this page's address followed by #c0-v1-s0-pv6-hb0-hp0-t0-x4-w150-l100-fb0-e0-k0-o1000-h4-p0-r4-c100

Input (the equipment held fixed) Low case NPV High case NPV
Equipment cost 75% of sourced −$10,237 125% of sourced −$25,389
Discount rate (real) 2.0% −$15,162 7.0% −$20,510
Export credit 0.0¢/kWh −$20,909 8.0¢/kWh −$14,717

The NPV at the default plan is −$17,813. The equipment is held fixed. Re-run over 52 weeks and 31 December: the export credit. Computed on 2026-11-05.

Result for the default plan Value
Net present value, 25 years (against the same home with nothing new, at 4.0% real) −$17,813
Saved on bills and gas, a year ($2,114 a year before, $1,314 with the plan) $800
Equipment, up front (upkeep $120 a year) $28,430
CO₂ a year, bought power and gas (down 1.3 t from 5.9 t; net of power sent back, credited at the same factor: 3.1 t) 4.6 t
Own solar covers (of the home's electricity use, at once or through the battery) 48%
Hours that need the grid (of 8,760; the busiest hour draws 1.5 kWh) 5,240 h

Run on every hour of 2025, as 52 weeks and 31 December, with this engine on 2026-11-05.

Hourly patterns: SPP's 2025 demand, wind and solar (EIA-930), reshaped to EIA's 2025 capacity factors. The home's use and heat are illustrative. Costs: EIA, DOE and NLR, in 2025 dollars; sources below.

Write your one-page note

Before you write: which number defends the plan?

On the wind farm, the bench shows two numbers that can disagree: the cost per kg of hydrogen, and what the extra equipment adds to the NPV.

As the hydrogen price rose, the optimizer built a larger electrolyzer to use more surplus wind.

The larger unit ran at lower utilization, so each kilogram carried more capital and the levelized cost of hydrogen rose, while hydrogen revenue lowered total system cost.

Among the cases studied, concentrated solar power with thermal storage running a high-temperature electrolyzer at variable power gave the lowest levelized cost of hydrogen; constant-power operation was the most expensive.

Approved excerpt of: Li et al. (2023), EEM 2023.

Yardstick here: the cost per kilogram across designs. Holds for: one West Texas site, cost projections from NREL's Annual Technology Baseline, a conference-length study.

Say which yardstick your note uses. Case file CF2 takes the first case apart.

  1. Title and plan link: the plan's link from the bench, and the date
  2. Recommendation: Two sentences: the owner's goal, what this site should install or not, and the number that decides it.
  3. The plan: each choice, with its unit
  4. Results: each result, with its unit and boundary
  5. Assumptions: one sentence each
  6. Sensitivity: every bar of the sensitivity chart in numbers, and what was held fixed
  7. What the site cannot do alone: One or two sentences: what it still needs from the grid or a buyer, and when.
  8. What would change the answer: One sentence from your sensitivity chart: does any tested input change the recommendation, and between which values?
  9. Limits of this analysis: what the model leaves out
Go deeper · ModelerThe Modeler version: one comparison from the M4 notebook
  • Storage and coupled carriers, modelling power, hydrogen and heat together (M4, in the Modeler tier), writes a plant linear program like the bench’s in Python.
  • It is another plant: two weeks, power bought and sold, a fixed hydrogen demand and a heat store; it has no NPV.
  • Hand in, with the note: the notebook’s weekly saving from adding the electrolyzer and tank (its section 3), winter and summer.
  • Then two sentences: how its model differs from the bench’s, and which difference could change your recommendation.
  • About 45 minutes. Work through M4 first; B3 shows how to read a formulation in words.

Check your note

Weak note · misses the Look for lines

Recommendation: Install 6 kW of solar. It saves $800 a year. [1] CO₂: Solar is clean, so the home goes green. [2] Alone: With a battery it could leave the grid. [3]

  1. Falls short: A saving without its cost: the panels cost $28,430 up front, and the NPV over 25 years is −$17,813.
  2. Falls short: No number and no boundary: bought power and gas fall from 5.9 t to 4.6 t a year (3.1 t net of the power it exports), at one average grid factor.
  3. Falls short: A claim the bench never tested: the home still buys from the grid in 5,240 hours a year.

Strong note · meets every Look for line

Recommendation: The owner wants panels that pay for themselves, so do not buy 6 kW of solar at the sourced costs: NPV −$17,813 over 25 years at 4% real [1], though it saves $800 a year. CO₂: Bought power and gas fall from 5.9 t to 4.6 t a year (Kansas's average grid factor); 3.1 t net of the power it exports. [2] Alone: The home still needs the grid in 5,240 hours a year, every night. [3] Sensitivity: No tested input turns the NPV positive: −$10,237 at 75% of the equipment cost, −$14,717 with an 8¢ export credit. [4]

  1. Works: The owner's goal first, then the decision and the number that decides it, with its horizon and rate.
  2. Works: CO₂ with its size and its boundary, and the credit for exported power kept apart.
  3. Works: What the site cannot do alone, in the bench's own numbers.
  4. Works: Whether the tested ranges change the answer, from the chart's bars: here they do not, so no switching point is claimed.
Weak note Strong note
Goal Unstated Stated: panels that pay for themselves
Decision Install, because it saves money For that goal, do not install: the NPV is negative
Numbers A yearly saving alone NPV, saving, CO₂ and grid hours, each with its unit
Sensitivity None Whether the tested ranges change the answer
Competence Developing Proficient (the bar for this note) Strong
Consistent accounting
Look for: NPV and CO₂ for the same plan, against the same site with nothing new, in one dollar year, with the CO₂ boundary said.		 Costs and emissions on different boundaries. Cost and emissions on one boundary; emissions question named; who pays and who benefits named. Distributional and environmental side effects estimated with stated assumptions.
Uncertainty
Look for: Every bar of the sensitivity chart in the note, the equipment held fixed, and whether any tested input changes the answer.		 Single numbers. Uncertainties classified; scenario or global sensitivity analysis; results as ranges. Decision-relevant uncertainty identified and its value discussed.
Realism
Look for: What the site still needs from outside (the grid, a buyer) and what the bench leaves out (outages, storage across weeks).		 Assumes announced projects exist. Evidence status of key inputs; a 'what the analysis leaves out' list. Realism checked with a practitioner or primary documents.
Communication
Look for: A page an owner can act on: the recommendation first, then the numbers that back it.		 Results without limits. A one-page note with the plan link: the recommendation first, then its NPV, its CO₂, the sensitivity chart and what the site cannot do alone. A briefing a decision-maker could act on without misreading.
Go deeper · ResearcherFrom this capstone to your own study
  • The same plan, note and self-check grow into the research project (CR): a question, a model at your chosen scale, uncertainty done properly and a short paper.
  • Start with the question studio (R10), which turns “what should this site install?” into a question you can test.

Numbers, sources and what’s next

Inputs the plan uses

Input Value Source
Price index $1 of 2022 = $1.093 of 2025; $1 of 2024 = $1.028 EIA STEO, Table 9a
Discount rate 4% real (you can change it) NLR 2025 ATB, financial cases
Home: household use 9,000 kWh a year, the town's 2025 demand shape (illustrative) EIA-930, SPP 2025
Home: heating need 14,000 kWh of heat a year, from a smooth made-up temperature (illustrative) illustrative
Home: rooftop solar $4,336/kW in 2022 dollars = $4,738/kW; 25 years; upkeep $20/kW a year (illustrative). Output: SPP's 2025 hourly solar pattern, scaled month by month to EIA's 2025 capacity factors and to a 17.6% year EIA AEO2026, residential assumptions, Tables 2-3; EIA, distributed generation costs (2024), Table 2-2; EIA Electric Power Monthly, Table 6.07.B
Home: battery $1,000/kWh (illustrative); 2.5 hours; 85% round trip; 15 years; upkeep 4% a year NLR 2025 ATB, residential battery
Home: heat pumps standard: about 2.4 kWh of heat for each kWh of electricity over a season (HSPF 8.2, the heating seasonal performance factor), $6,730 in 2022 dollars = $7,355; high efficiency: about 3.6 (HSPF 12.4), $8,620 = $9,420. HSPF ÷ 3.412 = the season's heat over electricity EIA AEO2026, residential assumptions, Tables 2-3
Home: gas furnace and gas 80% efficient; $14.81 per thousand cubic feet = $14.30 per MMBtu (a million British thermal units, about 293 kWh of heat); 52.91 kg CO₂ per MMBtu burned EIA, Kansas residential gas price; EIA, British thermal units; EIA, CO₂ emissions coefficients
Home: tariffs 14¢/kWh flat; time-of-use 9.66¢ and 28.98¢ (16:00–21:00), illustrative EIA Electric Power Monthly, Table 5.6.B
Home: export credit what the utility pays for each kWh the panels send to the grid (net billing): 4¢ by default, 0 to 8¢ on the bench (illustrative) illustrative
Home: grid CO₂ 680 lb/MWh = 0.308 kg CO₂/kWh (Kansas, 2024) EIA, Kansas Electricity Profile 2024
Farm: wind $1,712/kW and $34.57/kW a year (2025 dollars); 30 years: $133,575 a year per MW at 4%. Output: SPP's 2025 hourly wind pattern, reshaped month by month to EIA's 2025 capacity factors and to a 33.8% year, the best hours at full size EIA AEO2026, electricity assumptions, Table 3; EIA Electric Power Monthly, Table 6.07.B
Farm: battery $1,521/kW for 4 hours = $380.25/kWh; $41.84/kW a year; 85% round trip; 15 years EIA AEO2026, electricity assumptions, Table 3; NREL battery cost projections, 2025
Farm: electrolyzer $2,000/kW in 2022 dollars = $2,186/kW; upkeep 5% a year; 57.5 kWh/kg. Stack wear $6.01/MWh used: an illustrative allowance per MWh, based on DOE's stack replacement (11% of the cost after 40,000 operating hours) as if every hour ran at full power DOE Hydrogen Program Record 24005
Farm: hydrogen tank $800 per kg held (illustrative); 30 years illustrative
Farm: prices power $50/MWh fixed, or $40 and $100/MWh (16:00–21:00); hydrogen $4.00/kg by default (all illustrative) illustrative
Farm: potential CO₂ avoided 0.18 MMBtu of gas per kg of hydrogen made from natural gas × 52.91 kg CO₂ per MMBtu = 9.52 kg CO₂ per kg, if the farm's hydrogen replaces it EIA AEO2026, hydrogen assumptions, Table 2; EIA, CO₂ emissions coefficients

Keep going

Back to the course map

Related lab: the M4 notebook: power, hydrogen and heat in one linear program, about 45 minutes.

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.