HCQ-LSTM: Physics-Informed Hybrid Classical-Quantum LSTM

預測 Forecast vs. actual (synthetic test set, 10-min steps)

actual model (computed live in JS) persistence

風 Try it: next-step forecast

Current wind speed Wind direction Recent trend (m/s per 10 min)

Forecast (next 10 min)–

Persistence–

The sliders build a 8-step history ending at the chosen speed; the power columns follow a generic cubic power curve.

衡 Test-set metrics (synthetic)

	Model	Persistence

Persistence predicts "next = current". At 10 minutes ahead it is already very accurate, so it is the honest yardstick.

道 How it works

Each LSTM gate gets an extra term from its own 2-qubit variational quantum circuit, trained with a physics-informed loss. The full repo can also tune hyperparameters with the CLQGWO grey wolf optimizer. The model predicts the 10-minute change in wind speed and adds it to the current reading.

Weights were trained in PyTorch, exported to JSON and re-implemented in plain JavaScript, including a small statevector simulator for the quantum circuit. A test checks that the JS and Python predictions match within 1e-4.