OpenSynth - LF Energy

OpenSynth is an open data community democratizing access to AI-generated synthetic and real energy data, accelerating the decarbonization of global energy systems.

Initially focused on synthetic energy demand data, OpenSynth drew on its originator Centre for Net Zero’s AI expertise to generate and share synthetic data and models. This allows researchers, industry innovators and policy-makers to understand how energy demand profiles are changing, in a system that requires greater real time optimization of demand and supply on the grid. Access to demand data is highly restrictive, as a result of privacy protections – but generating synthetic datasets is a fast way to achieve widespread, global access to smart meter datasets.

OpenSynth has since broadened its scope. As energy systems seek to transition, the need for granular, interoperable, and transparent data grows – not just for demand, but for grid infrastructure, supply, flexibility, and more. OpenSynth now includes both synthetic and real datasets, across a wider range of domains, designed to support researchers, developers, and system operators in building robust, AI-ready models of the energy system.

A major milestone in this expansion is the inclusion of the D-GITT & RTE7000. D-GITT (Detailed Grid Inner Topology Timeseries) is a centralized, open dataset supporting collaborative research and modeling of power systems. It launches with its first major contribution: RTE7000 is an open-source dataset representing the 7000 nodes of the French transmission grid. This dataset is a joint effort between France’s Transmission System Operator, RTE, and CRESYM, a non-profit supporting open-source energy system simulation.

RTE7000 captures realistic load and topology data suitable for large-scale system studies and AI model development. The dataset offers a series of snapshots of the French transmission network in node-breaker topology, with 5-minute resolution, spanning a three-year period from January 2021 to December 2023.

What’s next for OpenSynth?

The RTE7000 dataset is the first of a new wave of system-level datasets on OpenSynth. Our goal is to establish OpenSynth as the definitive platform for:

AI-generated synthetic datasets for energy research and modelling
Real or synthetic system data enabling training of advanced AI models

The OpenSynth Technical Steering Committee includes Alliander, TU Delft, Hydro-Québec and others. Several further datasets – synthetic and real – are in the pipeline from international partners with unique access.

OpenSynth Videos

17:21

The OpenSynth project, powered by Centre for Net Zero and LF Energy, aims to democratize access to synthetic energy data, fostering a collaborative community to advance energy research while ensuring privacy. In this video, Gareth Jones, Chief Operating Officer, and Gus Chadney, Data Lead at Centre for Net Zero, discuss the project and its key focuses to address the climate crisis. Jones says, “OpenSynth is a project looking to democratize access to synthetic energy consumption data. So what we're hoping to achieve is to build a community around users and creators of synthetic data.”

YouTube Video UExLeUZmMUo5WGtwc2lBb2kxNE5tZEVSb0ZwT2k3SXJsaC41NkI0NEY2RDEwNTU3Q0M2

OpenSynth democratizes access to synthetic energy data for research and innovation

July 26, 2024 1:46 pm

22:22

This video was recorded at the Open Sustainability Policy Summit in Washington, DC. It was presented by Gareth Jones and Gus Chadney from the Centre for Net Zero.

Abstract: Access to raw smart meter data is essential to a rapid and successful energy transition. Yet little of this data is available for research and modelling purposes, largely because of consumer privacy protections. Synthetic data can solve this problem.

We want to liberate global access to smart meter data for research and innovation purposes, by creating an open community of synthetic smart meter data contributors and users.

Centre for Net Zero is already engaged in work that advances these ambitions. Our generative AI model, Faraday, outputs synthetic half-hourly consumption data for specific household archetypes. These are adapted for user-specified inputs, such as low carbon technology, property type and season, allowing users to simulate the entire distribution of load profiles of a bespoke population. The model is designed to help grid operators, innovators and researchers understand the impacts of electrification, and optimise the design of the future energy system.

In this session, we will be exploring what “good” looks like for synthetic smart meter data, as well as some interesting real-life and proposed use cases for the data.

The Open Sustainability Policy Summit was hosted May 2-3, 2024 by the Johns Hopkins University Whiting School of Engineering.

YouTube Video UExLeUZmMUo5WGtwc2lBb2kxNE5tZEVSb0ZwT2k3SXJsaC4yODlGNEE0NkRGMEEzMEQy

OSPS 2024 - OpenSynth - An open source community for synthetic energy data

May 29, 2024 4:04 pm