Transfer Learning for Non-Intrusive Load Monitoring

Abstract

Non-intrusive load monitoring (NILM) is a technique to recover source appliances from only the recorded mains in a household. NILM is unidentifiable and thus a challenge problem because the inferred power value of an appliance given only the mains could not be unique. To mitigate the unidentifiable problem, various methods incorporating domain knowledge into NILM have been proposed and shown effective experimentally. Recently, among these methods, deep neural networks are shown performing best. Arguably, the recently proposed sequenceto-point (seq2point) learning is promising for NILM. However, the results were only carried out on the same data domain. It is not clear if the method could be generalised or transferred to different domains, e.g., the test data were drawn from a different country comparing to the training data. We address this issue in the paper, and two transfer learning schemes are proposed, i.e., appliance transfer learning (ATL) and crossdomain transfer learning (CTL). For ATL, our results show that the latent features learnt by a 'complex' appliance, e.g., washing machine, can be transferred to a 'simple' appliance, e.g., kettle. For CTL, our conclusion is that the seq2point learning is transferable. Precisely, when the training and test data are in a similar domain, seq2point learning can be directly applied to the test data without fine tuning; when the training and test data are in different domains, seq2point learning needs fine tuning before applying to the test data. Interestingly, we show that only the fully connected layers need fine tuning for transfer learning. Source code can be found at https://github.com/MingjunZhong/transferNILM.

A Non-Intrusive Load Monitoring Algorithm Based on Non-Uniform Sampling of Power Data and Deep Neural Networks

“…This procedure allows for determining the most performing configurations, that, however, are specific for the target dataset. This represents a general problem for neural networks-based algorithms [33,34], and here it has not been taken into account, since the main focus of this paper is the evaluation of different sampling strategies regardless the network topology.…”

Non-Intrusive Load Monitoring via Deep Learning Based User Model and Appliance Group Model

“…One of the most effective work to improve the generation capability is transfer learning. In Reference [26], a deep learning model is trained to perform NILM at user A, and applying it to user B. If user B is similar to user A, we can use the model trained in user A as a initial model, then collect a small amount of labeled data from user B, and fine-tune it to perform NILM at the new user.…”

Automated Non-Intrusive Load Monitoring System Using Stacked Neural Networks and Numerical Integration

“…al. [23] proposed sequence to point algorithms for convolutional DNNs with appliance transfer learning (ATL), cross-domain transfer learning (CTL) and crossdomain transfer learning with fine tuning (CTL-FT) to perform energy disaggregation. The study used one convolutional DNN per appliance.…”

“…Considering appliance identification and load disaggrega-tion, different techniques have been proposed in literature, such as: Discrete Fourier Transform [12]; Decision trees [13]; Principal Component Analysis (PCA) [14], [15], Genetic Programming [16], Artificial Neural Networks (ANN) [17]- [19], Deep Artificial Neural Networks (DNN) [20]- [26], Hidden Markov Models (HMM) [27]- [31], Integer and Quadratic Programming [32]- [34], Transfer learning [23], among others. Many of these techniques make use of several consumption parameters obtained from data collected at high frequency, which require expensive meters, that are not viable for residential use.…”