The same real energy-balance model as the 2D version — diurnal solar
curve with mean-reverting cloud cover, an Ornstein-Uhlenbeck wind speed feeding
a cubic turbine power curve, battery round-trip losses, time-of-use grid
pricing and carbon intensity — drives a 3D microgrid instead of a chart. The
sun arcs across the sky at the simulated hour and lights the panels; the
turbine's blade speed tracks the live wind speed; glowing particles stream
along the wires from whichever source is active toward the battery, the house,
or out to the grid pylon, sized by the actual kW flowing at that instant.
panel glow ∝ solar_kW / solar_capacity
blade ω ∝ wind_speed (cubic power region highlighted)
particle flow: source → battery (surplus) or battery/grid → house (deficit)
cost, CO2, self-sufficiency: identical running sums to the 2D model
- Sun position — swings from horizon to horizon over the simulated day, and panel glow tracks real solar output, not just daylight.
- Turbine — spins faster as wind speed rises toward its rated speed, then holds steady output beyond it — the same cubic power curve as a real small turbine.
- Flow particles — direction and density show where energy is actually going each tick: battery charging, house consumption, or grid export/import.
- Evening demand shifted — the same "wise decision" lever as the 2D model: fewer red (grid-import) particles at dusk once shifted load is met from batteries or off-peak grid overnight.