How to use this atlas
This page is designed to be traversed as a pipeline. The key is not “collecting tools” — it is composing a minimal stack that can be executed locally, audited, and iterated without fragile dependencies.
1) Establish the data spine
Start with open datasets and time series. Snapshot what matters locally. Treat cloud-hosted repositories as sources to mirror, not infrastructure to depend on.
2) Lock demand and constraints
Buildings and appliances define the demand reality. Model them explicitly before attempting “optimal” system builds.
3) Run grid physics and failure modes
Planning outputs are not credible until power flow, voltage, dynamics, and islanding constraints are tested.
4) Execute planning models (scenario engines)
Use optimisation tools to generate scenario spaces and stress tests — not as governance templates.
5) Operate (control + dispatch)
Convert designs into operational logic: energy management systems, microgrid control, EV charging scheduling.
6) Ground in materials (LCA)
Operational “green” claims collapse without embodied impacts, supply chains, and toxicity accounted for.
Legend: tags & sovereignty tiers
Each resource card includes tags for Scale, Function, and an S-tier (sovereignty tier) describing dependency risk.
S-tiers
S2 items are preferred as foundations. S1 items can be core if run fully locally and solver/data constraints are controlled. S0 items are treated as ingest/reference or constrained-use utilities.
Scale & Function (examples)
Scale indicates where the tool lives in the stack. Function indicates what role it plays (data, physics simulation, optimisation planning, operations/control, or material impact accounting).
Tier-0 backbone (minimal executable core)
The Tier-0 backbone is the smallest set that can generate designs, test physics, and operate local systems while remaining auditable. Everything else in this page either feeds into Tier-0 or extends it.
PyPSA Python for Power System Analysis
A core planning and optimisation framework for modern power systems with renewables, storage, and sector coupling (electricity + heat + hydrogen + transport).
Stack role
Calliope multi-scale optimisation framework
Energy system optimisation focused on flexible model design, high spatial/temporal resolution, and repeatable scenario runs with clear separation of framework code and model data.
Stack role
OSeMOSYS long-horizon energy planning
A long-run integrated assessment and energy planning model generator designed for low barriers to adoption; widely used for policy and national-scale planning. Multiple toolchains exist; some deployments depend on proprietary solvers, others do not.
Sovereignty note
TEMOA Tools for Energy Model Optimization and Analysis
Energy system optimisation framework built for transparent scenario analysis and reproducible datasets; designed to support rigorous planning and uncertainty studies.
Stack role
oemof Open Energy Modelling Framework
Modular Python framework for building custom energy system models. The ecosystem includes specialised packages such as oemof.solph for LP/MILP optimisation.
Stack role
OnSSET Open Source Spatial Electrification Tool
GIS-based least-cost electrification planner selecting between grid extension, minigrids, and stand-alone options across settlements. Designed for energy access and real-world rollout planning.
Stack role
Data & measurement
This layer supplies datasets, grid time series, and structured energy databases. The core rule is simple: snapshot and mirror critical data locally; never architect “live dependence” on a cloud portal.
OpenEI Open Energy Information
Semantic wiki + data hub for energy datasets, mappings, and related information. Best used as a gateway to discover and pull datasets.
Sovereignty note
OEDI Open Energy Data Initiative
Large, centralized repository of energy datasets aggregated from DOE programs and national labs, published through a data lake model. Licensing is often CC BY 4.0 unless otherwise noted.
Sovereignty note
Open Power System Data (OPSD)
Curated power system datasets (especially Europe): time series, power plant lists, network-related data packages — processed, documented, and published for modelling use.
License note
Open Energy Platform (OEP)
A structured platform for energy-system modelling and data: datasets with metadata, quality indicators, and APIs for programmatic access.
Use pattern
Renewables.ninja
Wind and PV hourly output simulation + downloadable datasets. Used frequently as an input generator for system models.
License / sovereignty note
System planning & optimisation
These tools generate scenario spaces: capacity expansion, dispatch, and long-horizon trajectories. The key constraint: they must be paired with grid physics and demand realism (buildings + appliances), or they will output elegant impossibilities.
PyPSA
Optimisation and simulation for modern power systems with sector coupling. Strong ecosystem and high scalability.
Best used for
Calliope
Scenario exploration framework designed to run many variants from the same base model data.
Best used for
OSeMOSYS
Long-horizon energy planning for policy contexts. Designed for accessibility with low upfront costs.
Best used for
TEMOA
Planning tool built for transparent, reproducible databases and scenario comparisons.
oemof
Modular modelling framework for composing bespoke energy system models; supports LP/MILP via oemof.solph.
ElecSim agent-based electricity market model
Agent-based simulation of wholesale electricity markets (GenCos + demand). Useful where “single omniscient planner” assumptions fail.
Grid physics, dynamics & co-simulation
This layer tests feasibility. It answers: “will voltages stay inside limits?”, “what fails first?”, “does islanding work?”, and “what happens under transients?”. Planning models should feed into this layer before anything is treated as real.
pandapower
Power system modelling and analysis in Python: power flow, OPF, short-circuit, state estimation, time series control loops.
GridLAB-D
Distribution and smart-grid simulation: end-use loads, DERs, automation, and market effects in one environment.
OpenDSS
Open-source distribution system simulator (EPRI). Widely used for DER integration, hosting capacity, and distribution planning studies.
Cross-platform enhancement (recommended)
HELICS co-simulation framework
Federated co-simulation framework for coupling transmission, distribution, communications, markets, and other simulators into one run.
PowerModels.jl
Julia/JuMP package for steady-state power network optimisation; a strong alternative when Julia toolchains are preferred.
Distribution companion
ANDES dynamics & stability
Python library for power system modelling and dynamics: transient stability, small-signal stability, and symbolic-numeric prototyping.
Components & resources (physics building blocks)
Component libraries prevent “magic inputs”. They allow direct modelling of PV output, wind power, marine energy, and battery behaviour so system models can be grounded in transparent physics rather than opaque coefficients.
pvlib python
Photovoltaic system performance modelling in Python (irradiance → DC → inverter → AC), with benchmark implementations and validated methods.
windpowerlib
Wind turbine and wind farm power output modelling library; converts weather data and turbine characteristics into generation time series.
MHKiT Marine & Hydrokinetic Toolkit
Toolkit for marine renewable energy: data ingestion, QC, resource assessment, device performance, and loads (Python & MATLAB).
Operations & control (EMS / microgrids / EV)
This layer turns designs into operational logic. It is where sovereignty becomes tangible: local control, local optimisation, local scheduling, and minimal dependence on cloud orchestration.
OpenEMS Open Source Energy Management System
Modular platform for monitoring and controlling storage, PV, EV charging, heat pumps, electrolysers, and tariff-aware optimisation. Built explicitly around local integration requirements.
pymgrid microgrid simulator (control)
Microgrid simulation environment (rule-based, MPC, and RL-friendly control). Useful for control strategies and stress tests at micro scale.
Maintenance note
ACN-Sim EV charging simulator
Data-driven simulator for EV charging systems, capturing battery charging behaviour and infrastructure constraints; used for scheduling algorithm research.
PowerSimulations.jl operations simulation
Julia tool for quasi-static power system operations simulation (production cost models, scheduling and dispatch studies), designed for modular decision/emulation models.
Buildings, districts & cities (demand reality)
Buildings are the demand engine and the thermal battery of civilisation. This layer anchors “energy system” work in heat flows, envelopes, HVAC, and urban form.
EnergyPlus whole-building simulation
DOE’s open-source building energy modelling engine (BEM). Models HVAC, lighting, equipment, water, and advanced controls.
Stack role
OpenStudio EnergyPlus workflows
Open-source SDK and tools to build and automate EnergyPlus workflows, including graphical interfaces and model management.
ESP-r multi-domain building simulation
Building performance simulation modelling heat, air, moisture, light, and electrical power flows with high control over resolution and coupling.
LCA & material grounding (embodied reality)
This layer prevents energy modelling from collapsing into pure operational accounting. It answers: “what is embedded in the hardware?”, “what is the upstream burden?”, “what is displaced or shifted?”
openLCA
Open-source life cycle assessment software for footprinting and sustainability modelling, from simple products to large process systems.
openLCA Nexus LCA databases
Data portal for LCA inventory databases and sustainability datasets used by openLCA and other workflows.
License note
Brightway Python LCA framework
LCA and environmental impact framework designed for large datasets and programmatic experiments; strong for integration into modelling pipelines.
Humanitarian, access & off-grid appliances
This layer forces energy modelling to confront fragile contexts, displacement settings, weak-grid markets, and real appliance performance. It contains both field knowledge hubs and the market / performance intelligence needed to spec systems that actually work.
Humanitarian Energy Knowledge Hub
Living knowledge base for energy access in humanitarian and displacement settings: planning tools, case studies, and sector updates.
Practical Action — Energy resources
Practitioner-grade resources on energy access, clean cooking, off-grid systems, and real-world barriers/opportunities in low-income contexts.
Global LEAP Awards (Clean Energy Ministerial)
International competition identifying high-quality, energy-efficient off-grid appliances; winners/finalists undergo accredited lab testing.
Efficiency for Access — Market series & surveys
Reports and datasets describing off-grid and weak-grid appliance markets, demand, and impacts — a critical complement to purely technical modelling.
CLASP — Global LEAP programs & tools
Research, tools, and procurement mechanisms supporting efficient appliances in off-grid and weak-grid settings.
Integration rules (what plugs into what)
The atlas becomes operational when each layer’s outputs are treated as inputs for the next layer. The rules below define the default wiring.
Rule A — Data → Planning
Time series and structured datasets from Data & measurement feed planning engines (PyPSA, Calliope, OSeMOSYS, TEMOA, oemof, OnSSET). Any dataset used more than once should be snapshotted and versioned locally.
Rule B — Demand realism first
Build demand constraints using EnergyPlus, OpenStudio, ESP-r, and CEA. Off-grid contexts should cross-check appliance-level power and usage patterns with Global LEAP and Efficiency for Access.
Rule C — Planning → Physics
Every planning output must be tested in grid physics tools: pandapower, GridLAB-D, OpenDSS, and when needed for dynamics: ANDES. For coupled simulations (T&D + comms + markets), use HELICS.
Rule D — Design → Operations
Operational control stacks use OpenEMS (site/microgrid management), pymgrid / python-microgrid (microgrid control simulation), and ACN-Sim (EV charging scheduling simulation). Operations at system level can be stress-tested with PowerSimulations.jl.
Rule E — Materials reality always on
Any technology pathway should be grounded with life cycle assessment using openLCA and/or Brightway, pulling inventories as appropriate from openLCA Nexus. If inventories are proprietary, treat them as constraints with explicit licensing and provenance rather than hidden assumptions.
Rule F — Access & fragility are not optional
Energy access work should be cross-validated against: Humanitarian Energy Knowledge Hub and Practical Action resources, with appliance performance and market intelligence from Efficiency for Access. Without this, models drift into high-capacity assumptions and fail under real constraints.