On-Device Training of PV Power Forecasting Models in a Smart Meter for Grid Edge Intelligence

TL;DR

This paper explores on-device training of PV power forecasting models directly on smart meters, demonstrating feasible and resource-efficient methods for grid-edge intelligence using limited hardware.

Contribution

It introduces a practical framework for on-device training on resource-constrained smart meters, including mixed- and reduced-precision schemes for PV forecasting models.

Findings

Feasibility of on-device training in resource-limited smart meters.

Effective mixed- and reduced-precision training schemes.

Potential for economically deploying grid-edge intelligence.

Abstract

In this paper, an edge-side model training study is conducted on a resource-limited smart meter. The motivation of grid-edge intelligence and the concept of on-device training are introduced. Then, the technical preparation steps for on-device training are described. A case study on the task of photovoltaic power forecasting is presented, where two representative machine learning models are investigated: a gradient boosting tree model and a recurrent neural network model. To adapt to the resource-limited situation in the smart meter, "mixed"- and "reduced"-precision training schemes are also devised. Experiment results demonstrate the feasibility of economically achieving grid-edge intelligence via the existing advanced metering infrastructures.