LAD-BNet: Lag-Aware Dual-Branch Networks for Real-Time Energy Forecasting on Edge Devices

TL;DR

LAD-BNet is a novel lag-aware dual-branch neural network designed for real-time energy forecasting on edge devices, achieving high accuracy and speed with low memory usage, suitable for smart grid and building management.

Contribution

The paper introduces LAD-BNet, a hybrid neural architecture combining lag exploitation and TCNs, optimized for edge inference with significant improvements over existing models.

Findings

Achieves 14.49% MAPE at 1-hour horizon with 18ms inference time on Edge TPU

Outperforms LSTM and pure TCN models by 2.39% and 3.04% respectively

Maintains a 180MB memory footprint suitable for embedded devices

Abstract

Real-time energy forecasting on edge devices represents a major challenge for smart grid optimization and intelligent buildings. We present LAD-BNet (Lag-Aware Dual-Branch Network), an innovative neural architecture optimized for edge inference with Google Coral TPU. Our hybrid approach combines a branch dedicated to explicit exploitation of temporal lags with a Temporal Convolutional Network (TCN) featuring dilated convolutions, enabling simultaneous capture of short and long-term dependencies. Tested on real energy consumption data with 10-minute temporal resolution, LAD-BNet achieves 14.49% MAPE at 1-hour horizon with only 18ms inference time on Edge TPU, representing an 8-12 x acceleration compared to CPU. The multi-scale architecture enables predictions up to 12 hours with controlled performance degradation. Our model demonstrates a 2.39% improvement over LSTM baselines and 3.04% over pure TCN architectures, while maintaining a 180MB memory footprint suitable for embedded device constraints. These results pave the way for industrial applications in real-time energy optimization, demand management, and operational planning.