Improving accuracy and convergence of federated learning edge computing methods for generalized DER forecasting applications in power grid
This work addresses forecasting challenges for power grid operators managing distributed energy resources, but appears incremental as it builds on existing federated learning extensions.
The paper tackles the problem of forecasting distributed energy resources in power grids by developing federated learning methods with improved accuracy and faster convergence, specifically targeting non-IID data and heterogeneous clients.
This proposal aims to develop more accurate federated learning (FL) methods with faster convergence properties and lower communication requirements, specifically for forecasting distributed energy resources (DER) such as renewables, energy storage, and loads in modern, low-carbon power grids. This will be achieved by (i) leveraging recently developed extensions of FL such as hierarchical and iterative clustering to improve performance with non-IID data, (ii) experimenting with different types of FL global models well-suited to time-series data, and (iii) incorporating domain-specific knowledge from power systems to build more general FL frameworks and architectures that can be applied to diverse types of DERs beyond just load forecasting, and with heterogeneous clients.