Smart meter data is essential to rapid and successful energy transitions. Energy modelling is currently still largely based on static and highly aggregated data from the past, whilst access to demand data is highly restricted as a result of privacy protections.

Rather than joining widespread calls to unlock raw smart meter data through existing mechanisms, by challenging current data regulations and legislation, we believe generating synthetic data is the fastest way to achieve widespread, global access to smart meter datasets.

Synthetic data is generated using AI, replicating the statistical properties and characteristics of real datasets. It models the patterns, structures, and distributions inherent to the real data, to be used when real-world data is not readily available, or to supplement datasets that are incomplete.

Synthetic data increases the ease through which highly sensitive, personal data can be shared between organisations, providing realistic profiles for consumer archetypes that cannot be attributed to individuals.

• Faraday: a synthetic smart meter generator trained on over 1.8 billion readings from the largest energy supplier in the UK. It produces household-level synthetic load profiles consisting of half-hourly kWh consumption - modified on defined user-specified inputs. Find out more here.
• OpenSynth: we’re democratising global access to smart meter data via this open data community, sourced under The Linux Foundation’s LF Energy. You can access data, algorithms, best practices, benchmarking and more from CNZ and other community members. Find out more here.
• Frameworks: we’ve collaborated with technical experts at MIT, University of Oxford and Georgia Tech to propose a common evaluation framework to benchmark algorithms which generate synthetic smart meter data, drawing inspiration from work already done in areas like health and finance. It applies three tests - fidelity, utility and privacy - to consider whether the data meets privacy requirements whilst still being sufficiently accurate for its intended purpose. You can read it here.

Synthetic data offers a breakthrough solution and is already trusted across several industries globally including technology, healthcare and finance - with an estimated global value of $300 million, rising to $2.5 billion by 2030.

However, use of synthetic data is still in its infancy in the energy sector. In a recent policy paper, we’ve identified actions that must be taken to realise the potential benefits - and five applications for synthetic smart meter data:

1. Consumer Products and Services. Integration in consumer facing products can help people optimise their energy use, compare tariff rates, and save money on bills.
2. Policy and Regulatory Design. Synthetic data improves distributional analysis when designing energy policies, ensuring interventions target those most affected.
3. Electricity System and Network Modelling. From strategic forecasting, to addressing data gaps in real-time operations, synthetic data enables a more resilient and cost-effective power system.
4. Financial Products and Services. Alternative data sources can enhance the development of sustainable lending products and services and support regulatory compliance for emissions reporting, or assessing risk and portfolio resilience.
5. Housing Development Planning. Electricity grid connection requests can be made more accurately, reflecting how energy will be consumed in the homes of the future.