LAD-BNet高效预测能源消耗，边缘设备上1小时预测仅需18毫秒。
LAD-BNet: Lag-Aware Dual-Branch Networks for Real-Time Energy Forecasting on Edge Devices
- 双分支结构分别处理时序滞后与长短程依赖，提升预测精度。
- 1小时预测MAPE达14.49%，推理速度比CPU快8-12倍。
- 适合部署在资源受限的智能电网与楼宇系统中。
边缘设备上的实时能源预测对智能电网优化和智能建筑至关重要。本文提出LAD-BNet（滞后感知双分支网络），一种专为Google Coral TPU优化的神经架构。该混合方法结合一个显式利用时间滞后的分支与采用空洞卷积的时序卷积网络（TCN），可同时捕捉短期与长期依赖关系。在10分钟分辨率的真实能耗数据上测试，LAD-BNet在1小时预测上实现14.49% MAPE，Edge TPU上推理时间仅18ms，相比CPU加速8-12倍。多尺度架构支持长达12小时的预测，性能下降可控。相较于LSTM基线提升2.39%，纯TCN架构提升3.04%，且内存占用仅180MB，符合嵌入式设备约束。结果为实时能源优化、需求管理与运营规划提供了工业应用路径。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。