Free, in your browser
One HTML file. Nothing to install, no sign-up, and your numbers never leave the page.
MOEA subsidy built in
Applies the storage subsidy of NT$5M/MWh (up to 50%, for 1–10 MWh systems).
Web · Python modellive
Own your power.
A free calculator that tells a Taiwan factory in 60 seconds whether on-site battery storage pays back, using the real Taipower tariff and the MOEA subsidy.
NT$6.86per kWh, summer peak vs off-peak
60 sto a first answer
6 yrpayback, reference case
14.8%IRR, reference case
Why it exists
Peak power just got expensive.
Taiwan just made peak electricity expensive.
Most factory owners never run the numbers because a consultant quote takes weeks.
In short
Free, real, honest.
Real Taiwan numbers
Taipower's October 2025 three-stage high-voltage tariff, MOEA storage subsidy and current BESS prices.
Check any quote
Shows the maximum fair price per kWh, so you know if an integrator's offer pays. We don't sell batteries.
How it works
Use the app
- 1
Open Peakwise
Go to juw12.github.io/peakwise (中文 or English, auto-detected; switch any time).
- 2
Enter your bill numbers
Click Open the calculator and enter your contract capacity, peak load, monthly kWh and working days.
- 3
Read the answer
Read the recommended size, payback, IRR and 15-year NPV, then Print / save PDF.
What's inside
What it works out for you.
Real Taipower tariff
Uses the October 2025 three-stage high-voltage tariff: NT$9.39/kWh summer peak against NT$2.53/kWh off-peak.
Size, payback, IRR, NPV
Recommends a battery size and shows payback, IRR and 15-year NPV for it.
Shows the downside
Re-runs the case with cost +20%, savings −20% and no subsidy, and says plainly when a battery is not worth it.
Check any quote
Shows the maximum fair price per kWh, so you know if an integrator's offer pays.
Share and print
Gives a shareable link for your boss and a printable report (Print / save PDF).
中文 or English
Language is auto-detected and you can switch at any time.
Python-checked maths
The browser engine is checked against the Python reference model by the test suite.
Screens
See more of it.
Run it yourself
Run the model yourself.
From the README’s “Development” section. Full options, requirements and troubleshooting are there too.
Open the README$ python tests/test_model.py # model regression tests plus calculator parity check
$ python model/taiwan_pilot.py # full scenario table (JSON + chart in model/taiwan_pilot_output/)
# web: no build step; open the file directly, or serve docs/
$ python -m http.server 8000 -d docs # then visit http://localhost:8000/calculator/#selftestYour data
Your numbers stay with you
- One HTML file. Nothing to install, no sign-up, and your numbers never leave the page.
- No login and no meter-data upload.
- We don't sell batteries.
Good to know
Honest limits.
Straight from the README, so you know what you're getting before you try it.
- This is a screening tool, not an engineering study.
- Taipower dReg ancillary-service revenue is excluded on purpose, because clearing prices are near zero after the market filled up.
- Model assumptions: 88% round-trip efficiency, 90% usable depth, 2%/yr fade, 1.5%/yr O&M, NT$12.5M/MWh installed, 15-year life and a 6% discount rate.
Under the hood
How it's built.
Built with
Language
HTML/CSSPython
Data
NumPyMatplotlib
Testing
Automated tests
12 files in the repo · last updated 6 Nov 2026.
Browse the codeRecently shipped
Calculator: verdict right after inputs on phone/tablet, table scroll cue, even preset row, roomier Details labels (#2)
Design: fix mobile overflow, LINE contrast, tap targets, small text, calculator favicon (#1)
Move to juw12/peakwise: update all URLs
Battery Check: fix seasonal demand-charge rates, verdict, quote check, sensitivity, LINE share, presets
Watt Reserve: Taiwan factory battery-storage payback calculator
Questions
Good questions.
How do I get it running?
The commands in the “Development” box above come straight from the README. Paste them into a terminal; the README covers options and troubleshooting.
Where does my data go?
One HTML file. Nothing to install, no sign-up, and your numbers never leave the page. No login and no meter-data upload.
What does it run on?
Any modern browser: it is one HTML file with nothing to install, and its maths is checked against a Python reference model.
What does a typical result look like?
In the README's reference case (a 2 MWh / 1 MW system on the October 2025 tariff with the MOEA subsidy) the battery pays back in 6 years with a 14.8% IRR and an NPV of +NT$8.2M, while the same system on the 2023 tariff without a subsidy never pays back.
What is planned next?
The roadmap lists uploading a Taipower bill (PDF) to fill in the inputs, 15-minute load-profile import, a solar + storage case and an integrator directory.
What doesn't it do yet?
This is a screening tool, not an engineering study. Taipower dReg ancillary-service revenue is excluded on purpose, because clearing prices are near zero after the market filled up.
Is it free?
Yes. The source is public on GitHub under the MIT licence.
Who made it?
Juwen Lie, an R&D mechanical engineer in Batam, Indonesia, as an independent personal project.
Keep exploring
More from the same workbench.
Check your factory's numbers.
Own your power.