Introducing Edge Intelligence to Smart Meters via Federated Split Learning

Abstract The ubiquitous smart meters are expected to be a central feature of future smart grids by enabling the collection of massive fine-grained consumption data to support demand-side flexibility. However, the current smart meters are still not smart enough. They can only perform basic data collection and communication functionalities but fail to carry out any on-device intelligent data analytics due to hardware constraints in terms of memory, computation, and communication capacity. Moreover, privacy concerns have hindered the utilization of data from distributed smart meters. Here, we present an end-edge-cloud federated split learning framework to enable collaborative model training on resource-constrained smart meters with the assistance of edge and cloud servers in a resource-efficient and privacy-enhancing manner. The proposed method is validated on a hardware platform to conduct building and household load forecasting on smart meters with only 192KB of static random-access memory (SRAM). We show that the proposed method can reduce the memory footprint by 95.5%, the training time by 94.8%, and the communication burden by 50% under the distributed learning framework, and achieve comparable or even superior forecasting accuracy compared to resource-unlimited methods.