VPP-Core: Mini Virtual Power Plant Orchestrator

VPP-Core is an MVP that demonstrates:

• high-concurrency IoT telemetry ingestion in Go
• event-driven processing with Kafka + Benthos + n8n
• AI-assisted optimization via a Python gRPC service
• a small operator UI to observe simulation outputs live

Architecture (MVP)

producer (Go) -> Redpanda/Kafka (telemetry.raw) -> Benthos -> Postgres (telemetry_events + alerts) + n8n webhook

orchestrator (Go REST) -> optimizer (Python gRPC + OpenAI)

Services

• redpanda: Kafka-compatible broker
• postgres: app data (vpp) + n8n DB (n8n)
• n8n: low-code workflow endpoint for low-battery events
• producer: simulated smart-home device telemetry generator
• benthos: stream transform/routing pipeline
• optimizer: Python service that generates optimization plans
• orchestrator: REST API (/healthz, /devices/{id}/latest, /devices/{id}/strategy, /alerts)
• ui: React operator console (http://localhost:3000)

What this simulation actually does

1. producer simulates many smart homes sending telemetry continuously.
2. benthos enriches events and stores them in Postgres.
3. low-battery events are also persisted as alerts and pushed to n8n webhook.
4. orchestrator exposes read APIs over that live stream state.
5. optimizer uses your OpenAI key to produce a strategy on demand.
6. ui shows latest telemetry, recent alerts, and generated strategy for a device.

Prerequisites

• Docker
• Go 1.23+ (for local build)
• Python 3.12+ (for local syntax checks / tooling)
• OpenAI API key

Quick Start

cp .env.example .env
# set OPENAI_API_KEY in .env

make up
make demo

Then open:

• UI: http://localhost:3000
• n8n: http://localhost:5678

UI Preview

Stop everything:

make down

Useful Make Targets

• make up - build and start all services
• make down - stop and remove containers/volumes
• make logs - tail logs
• make ps - list services
• make proto - regenerate Go/Python gRPC stubs
• make build - local Go build sanity check
• make demo - run end-to-end API smoke flow

One-time n8n Workflow Import

Import n8n/workflows/low-battery.json in the n8n UI:

1. open http://localhost:5678
2. create/import workflow from file
3. activate it

Note: alerts are also persisted directly by Benthos into Postgres, so /alerts works even if n8n workflow import is skipped.