Energy Disaggregation from Low Sampling Frequency Measurements Using Multi-Layer Zero Crossing Rate

Abstract

Non-Intrusive Load Monitoring aims to disaggregate the energy consumption measurements of a smart-meter from household to device level. Most commercial smart-meters measure at 1 Hz or slightly higher, in order to be able to wirelessly transmit the acquired data to a server for further data analysis. The measured aggregated signal is a low sampling frequency signal, which is prohibitive for spectral analysis based disaggregation. In this article we propose an extension of the zero crossing rate as a time domain measurement of frequency content and compare it to the previously proposed Karhunen Loeve Expansion, in terms of energy disaggregation performance and in terms of computational time. The proposed feature was evaluated on the well-known and widely used REDD database and the experimental results shown that our approach significantly improves the baseline performance and has much lower computational cost than the previously proposed Karhunen Loeve Expansion.