Skip to article content

Revised 23 October 2026 · v1.0.0 · what changed

Code: github.com/herrrickshaw/hydrogen-electrolyser-control

Watch the walkthrough

Watch on YouTube: youtu.be/1MpY6jGGDkE

Why it is useful

• Who benefits: students and engineers who want a runnable sketch of how an electrolyser controller is structured (setpoints, ramps, safety limits, logging, an API) before touching hardware.
• What it replaces: writing the plumbing yourself. One command compares three electrolyser types and another exposes control over HTTP.
• Evidence from the run: 7 of 8 bundled examples completed; the PEM, AEM and alkaline comparison printed efficiencies of 75.0, 72.0 and 68.0 percent.
• It also shows the limits honestly: I found a crash, a voltage setpoint that is silently clamped, and a production table that is physically wrong. Those are listed below so you do not trust the wrong numbers.

What it is for

The repository holds a control system for three electrolyser types, an advanced layer with ramp profiles, data logging and safety interlocks, a Flask REST API with a Python client, and eight runnable examples. Everything runs against a software model, not a device.

Its own validation notes say that every test used synthetic power profiles and that no hardware testing was done. Simulation only, not for operating real electrolysers. The repository is archived, so it will not receive fixes; treat it as a teaching reference.

Who it is for

• Students learning how electrolyser control is organised
• Energy analysts who want a toy model to explain ramping and safety limits
• Non-programmers who want an AI assistant to walk through each module

What you need

Get the code and an AI assistant

You do not have to be a programmer. Get the repository, give it to an AI assistant, and ask in plain language. Any of these routes works:

git clone https://github.com/herrrickshaw/hydrogen-electrolyser-control.git
cd hydrogen-electrolyser-control
claude      # or: gemini

Install

1. Get the code (the repository is archived, so you can clone it but not push)

   git clone https://github.com/herrrickshaw/hydrogen-electrolyser-control.git
   cd hydrogen-electrolyser-control

2. Create an environment and install

   python3 -m venv .venv && source .venv/bin/activate
   pip install -r requirements.txt

Quick start: commands and what you should see

Run one example. The menu is interactive, so pipe the choice in

echo 5 | python electrolyser_examples.py

Three runs (PEM, AEM, alkaline) and a comparison summary.

Run the ramp profile example

echo 4 | python electrolyser_examples.py

A time series of voltage, power and efficiency as a ramp is applied.

Start the REST API and drive it with curl

python electrolyser_web_api.py PEM 5000
curl -X POST localhost:5000/api/control/start
curl -X POST localhost:5000/api/control/voltage -H 'Content-Type: application/json' -d '{"voltage": 45.0}'
curl localhost:5000/api/status

JSON replies. Note the README mentions a /api/production path, but the code serves /api/monitoring/production.

Run example 7 (safety limits) to see the crash

echo 7 | python electrolyser_examples.py

The limits print, then an AttributeError traceback.

Sample outputs

Real output from a run on the date shown in the sources line; trimmed to fit.

Comparing the three electrolyser types (example 5)

echo 5 | python electrolyser_examples.py
Type            Efficiency      H2 (Nm³)        Water (L)
------------------------------------------------------------
PEM             75.0%           0.0079           0.0075
AEM             72.0%           0.0075           0.0083
ALKALINE        68.0%           0.0071           0.0100

Best Efficiency: PEM (75.0%)
Best H2 Production: PEM (0.0079 Nm³)

Efficiencies are the model's configured maxima. The hydrogen quantities are for a short simulated run.

A power ramp (example 4, trimmed rows)

echo 4 | python electrolyser_examples.py
   t=0.0s: V=30.3V, P=1027W, Eff=63.2%
   t=4.0s: V=37.0V, P=1383W, Eff=64.3%
   t=8.0s: V=44.5V, P=1987W, Eff=66.2%
   t=12.0s: V=47.2V, P=3040W, Eff=69.5%
   t=16.0s: V=47.2V, P=3998W, Eff=72.5%
   t=18.0s: V=47.2V, P=3998W, Eff=72.5%

Voltage saturates at 47.2 V while power keeps rising, and efficiency climbs with power.

Production at different power levels (example 8): wrong

echo 8 | python electrolyser_examples.py
Power(W)     Voltage(V)   Current(A)   Eff(%)     H2(Nm³/h)    O2(Nm³/h)
------------------------------------------------------------------------------
500          0.00         0.00         61.6       316.4381     158.2191
1000         0.00         0.00         63.1       316.4381     158.2191
2000         0.00         0.00         66.2       316.4381     158.2191
4000         0.00         0.00         72.5       316.4381     158.2191
4500         0.00         0.00         74.1       316.4381     158.2191

The hydrogen rate is the same 316 Nm³/h at 500 W and 4,500 W, and the voltage and current columns are zero. The stack's own spec says about 0.95 Nm³/h. Do not use this table.

Safety limits and a crash (example 7)

echo 7 | python electrolyser_examples.py
PEM Safety Limits:
  Voltage: 30.0V - 60.0V
  Current: 10.0A - 150.0A
  Temperature: 5.0°C - 80.0°C
  Pressure: 1.0 - 35.0 bar
   Setting voltage to 45V (normal)...   Safe: True
   Attempting to set voltage to 65V (exceeds 60.0V limit)...
   Safe: False
   Alert: Voltage too high: 65.0V > 60.0V
AttributeError: 'ElectrolyserControlSystem' object has no attribute 'safety_monitor'

The limit check itself works, then the script fails on a missing attribute.

REST API session

curl against python electrolyser_web_api.py PEM 5077
POST /api/control/start     -> {"status":"started", ...}
POST /api/control/voltage   -> {"success":true,"voltage":45.0}
GET  /api/status            -> "mode":"RAMP_UP", "voltage":0.0, "power":0.0,
                               "safety":{"alerts":[],"safe":true}, "type":"PEM"
POST /api/control/stop      -> {"status":"stopped", ...}

The voltage call reports success, yet the status afterwards shows 0.0 V in the RAMP_UP mode.

Ways to use it

Sample prompts for your AI assistant

Copy, paste and edit the parts in [brackets].

• Claude: Read electrolyser_control_system.py and explain in plain language how the voltage regulator and the safety monitor work together.
• Gemini: Run example 5 and tell me why the alkaline type produces less hydrogen than PEM in this model.
• ChatGPT: I pasted the output of example 8. Why would the hydrogen rate be identical at 500 W and 4,500 W, and what should the formula depend on?
• GitHub (Copilot Chat): Fix example_safety_monitoring so it no longer calls system.safety_monitor.clear(), and re-run example 7.
• Claude: List every place where the README and the code disagree, for example the production endpoint path.

Limits and honest caveats

Sources: Cloned github.com/herrrickshaw/hydrogen-electrolyser-control (archived) into a clean folder and ran electrolyser_examples.py choices 1 to 8 and electrolyser_web_api.py in a throwaway virtualenv with Python 3.14 on 23 October 2026. Output is copied from those runs and trimmed. The repository was not modified.

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.