Privacy-Preserving Distributed Learning for Renewable Energy Forecasting

Abstract

Data exchange between multiple renewable energy power plant owners can lead to an improvement in forecast skill thanks to the spatio-temporal dependencies in time series data. However, owing to business competitive factors, these different owners might be unwilling to share their data. In order to tackle this privacy issue, this paper formulates a novel privacypreserving framework that combines data transformation techniques with the alternating direction method of multipliers. This approach allows not only to estimate the model in a distributed fashion but also to protect data privacy, coefficients and covariance matrix. Besides, asynchronous communication between peers is addressed in the model fitting, and two different collaborative schemes are considered: centralized and peer-topeer. The results for a solar energy dataset show that the proposed method is robust to privacy breaches and communication failures, and delivers a forecast skill comparable to a model without privacy protection.

Distributed Learning Applications in Power Systems: A Review of Methods, Gaps, and Challenges

“…However, the proposed method was not online and only learned model parameters were exchanged between agents. Moreover, Goncalves et al [55] used the same approach for solar energy forecasting.…”

Privacy-preserving Spatiotemporal Scenario Generation of Renewable Energies: A Federated Deep Generative Learning Approach

“…Such kind of centralized data handling approaches will suffer from the following problems: 1) these approaches depend on a central workstation with powerful computing power and storage space; 2) frequent data exchanges between renewable energy plants and a central workstation will inevitably cause heavy communication overheads; 3) data stored in a centralized form is vulnerable to data leakage and cyberattacks; 4) most importantly, some renewable energy data owners, such as independent system operators (ISOs), are not willing to share data with others in practical applications due to data privacy. In this context, data privacy can refer to either commercially sensitive data from grid-connected renewable energy plants with commercial competition or personal data from households with renewable energy technology [13]. At present, many countries have issued special laws such as the General Data Protection Regulation (GDPR) [14] enforced by the European Union and the China's Cyber Security Law and the General Principles of the Civil Law to regulate the management and use of data.…”

“…Some distributed structures considering data security and privacy protection have been studied in [13], [15], [16]. Reference [13] formulates a privacy-preserving framework that combines data transformation techniques with the alternating direction method of multipliers for renewable energy forecasting.…”

Prediction Markets for Probabilistic Forecasting of Renewable Energy Sources

“…This way, consumers have incentives to forecast as accurately as possible or adjust their electricity consumption to match their own forecast and thereby, improve the performance of the forecasting model. In [18], a distributed learning framework for exchanging res-related forecast data between agents has been developed, in which the data privacy is preserved and therefore the agents are willing to share their data. In [19], a data marketplace for renewable forecasting has been proposed that allocates a reward to the data providers, proportional to the forecasting accuracy reflected by the reduction in the electricity market imbalance costs.…”