ADF & TransApp: A Transformer-Based Framework for Appliance Detection Using Smart Meter Consumption Series

TL;DR

Detecting appliance ownership from very-long, low-frequency smart-meter series is challenging due to data length and sparsity. The authors propose ADF, a framework that fragments consumption series into subsequences and merges per-subsequence predictions, together with TransApp, a Transformer-based classifier pretrained in a self-supervised manner to learn robust representations. Across two real-world datasets, ADF+TransApp outperforms state-of-the-art time-series classifiers and scales to long series, with larger gains when leveraging self-supervised pretraining (TransAppPT-l). The approach offers a practical, scalable solution for electricity suppliers to infer appliance ownership and tailor personalized energy offers, aiding the energy transition.

Abstract

Over the past decade, millions of smart meters have been installed by electricity suppliers worldwide, allowing them to collect a large amount of electricity consumption data, albeit sampled at a low frequency (one point every 30min). One of the important challenges these suppliers face is how to utilize these data to detect the presence/absence of different appliances in the customers' households. This valuable information can help them provide personalized offers and recommendations to help customers towards the energy transition. Appliance detection can be cast as a time series classification problem. However, the large amount of data combined with the long and variable length of the consumption series pose challenges when training a classifier. In this paper, we propose ADF, a framework that uses subsequences of a client consumption series to detect the presence/absence of appliances. We also introduce TransApp, a Transformer-based time series classifier that is first pretrained in a self-supervised way to enhance its performance on appliance detection tasks. We test our approach on two real datasets, including a publicly available one. The experimental results with two large real datasets show that the proposed approach outperforms current solutions, including state-of-the-art time series classifiers applied to appliance detection. This paper appeared in VLDB 2024.

Figures (7)

• Figure 1: Example of a consumption series of 12hours, containing a dishwasher and a plugin heater at two sampling frequencies (1 second vs. 30min).
• Figure 2: Overview of our proposed Appliance Detection Framework (ADF).
• Figure 3: Overview of the TransApp architecture.
• Figure 4: Overview of the TransApp two steps training.
• Figure 5: (a) Impact of amount of unlabeled data used for pretraining; (b) Sensitivity study for $w$.