Problem Statement

When a rooftop solar array produces peak surplus at midday, the energy is back-fed to the distribution utility at a fraction of its retail value, while a neighbor across the street pays full commercial tariff for that exact electricity. Centralized distribution grids penalize micro-producers and experience severe phase imbalance and over-voltage trips when multiple solar inverters inject power simultaneously into the same low-voltage substation feeder. This sprout develops an open-source peer-to-peer (P2P) local microgrid simulation engine and smart meter telemetry gateway for neighborhood energy cooperatives.

Open Hardware Bill of Materials (BOM Hints)

  • Smart Meter Optical Interface: ESP32 optical probe (IEC 62056-21) or galvanically isolated RS-485 Modbus-RTU converter capturing granular 15-minute OBIS load curves directly from residential and commercial meters.
  • Local Simulation Server: Low-power edge mini-PC (refurbished HP T630 x86 thin client or Raspberry Pi 4 drawing <15W).
  • Software Simulation Stack: Discrete-event simulation daemon written in Python utilizing pandapower for AC power-flow and voltage-drop calculations; embedded SQLite / DuckDB local time-series database.
  • Secure Mesh Networking: WireGuard peer-to-peer encrypted mesh connecting neighborhood nodes to exchange telemetry securely without any centralized cloud dependency.

Schematics & Experiment Notes

  1. 15-Minute P2P Settlement Algorithm: The simulator matches local micro-producers with active consumers in 15-minute intervals. Surplus solar power is routed virtually to neighbors at an equitable mid-market rate ($Tariff_{\text{P2P}} = \frac{Tariff_{\text{import}} + Tariff_{\text{export}}}{2}$), maximizing economic savings for both producer and consumer.
  2. Substation Feeder Constraint Modeling: Using low-voltage electrical distribution network impedance models, the simulator verifies that concurrent solar injection does not breach statutory $\pm 10\%$ voltage envelopes on $230\text{V}$ lines, preventing inverter trips.
  3. Automated Demand-Side Flexibility: Dispatches local MQTT control signals to trigger flexible residential loads (heat pumps, hot water thermal stores, home battery storage, EV chargers) precisely during peak midday solar surplus.
  4. Zero-Cloud Privacy Architecture: Highly sensitive high-resolution household consumption profiles never leave the local neighborhood perimeter; the engine exposes only cryptographically aggregated net-settlement values.

Call for Contribution (How to Join)

  • Power Systems Engineers: Provide real-world wire impedance parameters and low-voltage topology models typical of Balkan residential distribution grids.
  • Python / Data Science Developers: Contribute to our pandapower optimization routines and machine-learning load forecasting pipelines.
  • Citizen Energy Collectives: Share anonymized 15-minute smart-meter profile logs to benchmark and validate our open simulation algorithms.