Recurve is excited to announce Energy Differential Privacy, the first open-source library purpose built for computing differentially private energy data statistics.
Over the past two years, Recurve has been working as the technical lead on a Department of Energy initiative to develop a set of tools to solve the problem of how to unlock energy data while preserving fundamental principles of customer privacy. With COVID shutdowns, this task has taken on new urgency.
Recurve’s Energy Differential Privacy (EDP) library is open-source and available for any third party to utilize for their own research or implementation purposes. Over the next few months, we will be integrating EDP into our existing products so that all of our utility partners will have confidence that whether they are deploying comparison groups to account for the effects of COVID or helping implementers deliver more cost-effective programs, customer energy data protections will meet the highest standards of data privacy.
We would like to thank the US Department of Energy for funding these important efforts, and our partners, the National Renewable Energy Laboratory, the Lawrence Berkeley National Laboratory, and the City of San Francisco, for helping to make possible this important advance.
The combination of rapid decarbonization goals, pressures from COVID, and proliferation of diverse behind the meter distributed resources has created increasing critical needs to access the rich and valuable data created by our country's investment in smart meters.
At the same time, protecting data privacy is important: energy consumption can predict business health, and exposing energy data could enable insider trading; personal lives can be upended when energy use contradicts supposed whereabouts.
For utilities, these challenges come at a time when they are under pressure to make customer energy use data more accessible while at the same time facing increasing scrutiny and legal liability over customer data privacy.
Until now, utilities have relied on antiquated privacy approaches such as simple aggregation and anonymization, in an attempt to strike a balance between these critical public use cases. However, these approaches were designed for a time before interval data was available, and lack the sophistication necessary to provide needed insights while robustly protecting customer privacy.
Our work on differential privacy isn’t done. We will be extending the core software library to cover more use cases and we are excited to welcome new collaborators in contributing to these efforts.
Recurve is excited to contribute to this renaissance of privacy-protecting tools and believe that Energy Differential Privacy (EDP) is foundational to the decarbonized grid.
Read the methods, download the code, and contribute to Energy Differential Privacy at edp.recurve.com.