COLD: Concurrent Loads Disaggregator for Non-Intrusive Load Monitoring

TL;DR

This paper introduces COLD, a transformer-based model for non-intrusive load monitoring that accurately disaggregates high-frequency household electricity data with multiple simultaneous devices, supported by a new dataset and performance analysis.

Contribution

We propose COLD, a novel transformer-based NILM model capable of handling up to 42 devices and 11 concurrent loads, along with a new high-frequency dataset for load disaggregation.

Findings

COLD achieves 95% load identification accuracy.

COLD handles up to 11 concurrent loads effectively.

Performance declines as the number of concurrent loads increases.

Abstract

The global effort toward renewable energy and the electrification of energy-intensive sectors have significantly increased the demand for electricity, making energy efficiency a critical focus. Non-intrusive load monitoring (NILM) enables detailed analyses of household electricity usage by disaggregating the total power consumption into individual appliance-level data. In this paper, we propose COLD (Concurrent Loads Disaggregator), a transformer-based model specifically designed to address the challenges of disaggregating high-frequency data with multiple simultaneously working devices. COLD supports up to 42 devices and accurately handles scenarios with up to 11 concurrent loads, achieving 95% load identification accuracy and 82% disaggregation performance on the test data. In addition, we introduce a new fully labeled high-frequency NILM dataset for load disaggregation derived from the UK-DALE 16 kHz dataset. Finally, we analyze the decline in NILM model performance as the number of concurrent loads increases.