A utility from a polygon
Draw territories, search a place or ZIP code, or describe the scenario in plain English and review the spec Claude drafts. Premises land on real buildings and assets on real roads — urban, suburban, or rural density.
Simulate everything
Grid Data Simulator A whole utility from a polygon.
Draw a territory, search a place, or describe it in plain English, and Grid Data Simulator (GDS) generates a complete synthetic utility — electric, gas, and water networks, customers, meters, and telemetry — then delivers it in your MDMS's own data model, the way the real estate would.
Map-driven
draw, search, or describe
Millions
of meters, proven at scale
GDM-native
entities, relationships, history
Multi-commodity
electric · gas · water
Why simulate
The utility your systems need to be tested against
Real utility data is scarce, sensitive, and slow to obtain. Grid Data Simulator builds a believable utility instead — premises on real addresses, assets on real roads, meters with their channels, partners, contracts, and service agreements — and delivers it to each system on its own Kafka cluster. Every scenario is reproducible from its spec and seed.
GDM's data model, natively
Premise, location, device, channel, partner, and contract entities with their full relationship families. Every attribute is dated, and every change carries its history.
Integration-ready master data
Deliver per entity or as one combined message per service point; target GDM or SAP, with profile sync for SAP; stamp every envelope with the source system a CI/CD pipeline expects.
Several brokers, one routing table
Meter master data and reads to GDM, asset master data and SCADA to Enhanced Analytics, commands to the MDMS — each topic routed to its own Kafka cluster.
Telemetry feeds & historical loads
AMI cohorts on staggered cadences with their own channels and lag; SCADA on its own clock; meters silent under outages. Historical backfills of any length run as monthly job chains.
Realism on demand
Storms restored by finite crews with ETRs and nested outages; DER and EV adoption; VEE fault patterns with an expected-findings manifest; and dirty data injected at the broker edge.
Actions & lifecycle events
Change one record and publish exactly the transaction a CIS would send — meter exchange, move-out and move-in, attribute changes — each with the reads that follow, in a replayable ledger.
Command & control
The full CIS → MDMS → HES round trip — disconnect, reconnect, on-demand read, power status, ping — with the simulator on both ends and your MDMS in the middle.
How it works
See it, end to end
Step 1
Define the territory
Draw a polygon, search a place, or describe the scenario in plain English and review the generated spec.
Step 2
Generate the utility
Networks, premises, customers, contracts, meters, and channels are built — reproducible from the spec and seed.
Step 3
Deliver it
Master data, reads, SCADA, lifecycle transactions, and commands go to each system on its own Kafka cluster, in the target's own model.
Step 4
Measure downstream
Watch the MDMS ingest, validate, and act — with the Testing Platform verifying the results.
Use cases
Where Grid Data Simulator earns its keep
MDMS acceptance & VEE certification
Every rule fires on exactly the meters it should — the manifest says which.
GDM and SAP integration testing
Master data, profile sync, and reads in the exact entity model.
Load and scale proving
Millions of meters, published at broker speed, before go-live.
CIS lifecycle & outage operations
Exchanges, move-ins, and storms with crews and ETRs.
Command-and-control integration
A real MDMS's disconnect, reconnect, and read flows — no head-end, no CIS.
CI/CD test environments
Deployed by ControlPlane, stamped with each pipeline's source system.
Under the hood
Technology
Operable at scale: partition manifests, banded spill-to-disk processing, monthly job chains, and per-band pause and cancel — deployed with Docker Compose, or on Kubernetes with Helm through ControlPlane.
Angular + MapLibre GL
Python / FastAPI
DuckDB (spatial)
Postgres + Parquet
Arq + Redis workers
Kafka, multi-broker
Claude (NL → spec)
Keycloak
OpenAPI
Docker Compose / Helm
Better together
Part of the Grid Data Family
Every family application shares one identity platform, one design system, one registration pattern, and one deployment path.
Receives master data and reads in its own entity model.
Receives asset master data and SCADA on its own broker.
Runs its suites against environments stamped with each pipeline's source system.
Hosts its shared Postgres, Keycloak, Kafka, and Redis, and registers it.
FAQ
Common questions about Grid Data Simulator
What is Grid Data Simulator?
Grid Data Simulator (GDS) builds a complete synthetic utility from a territory you draw, search, or describe in plain English — networks, premises, customers, contracts, meters, and telemetry — and delivers it to your systems the way a real utility would, so you can test without customer data.
What does it generate?
Electric, gas, and water networks from substation to meter; premises and service points on real addresses; partners, contracts, and service agreements; meters and their channels; and interval, register, and SCADA telemetry.
Which systems can it deliver to?
Master data and reads in the Grid Data Management Platform's own entity model, or to SAP with profile sync; asset master data and SCADA to Enhanced Analytics; and commands to your MDMS — each routed to its own Kafka cluster.
Can it test VEE rules and outages?
Yes. It writes VEE fault patterns into chosen meters with an expected-findings manifest, so you know exactly which rules should fire, and it simulates storms restored by finite crews with ETRs and nested outages.
Is every scenario repeatable?
Yes. Every scenario is reproducible from its spec and seed, so a test can be run again exactly — before and after a change.
How is it deployed?
With Docker Compose, or on Kubernetes with Helm as a managed application on the Platform Manager's shared Postgres, Keycloak, Kafka, and Redis — built, imported, and upgraded through ControlPlane.
See Grid Data Simulator live
A working demonstration takes thirty minutes — on our environment or yours, with realistic simulated data.
Connect With Us