Energy Consumption Dataset for Smart Homes with Individual Appliance Metering (Zenodo 2025)

Overview

The Energy Consumption Dataset for Smart Homes published on Zenodo in January 2025 provides granular appliance-level energy data collected from a controlled smart home testbed for energy disaggregation and home automation research.

Testbed Configuration

• Five common household appliances instrumented with individual smart meters: refrigerator, washing machine, microwave oven, air conditioner, and television.
• Each smart meter records real-time power consumption, voltage, current, power factor, and cumulative energy usage.
• Data collected over an extended period capturing diverse usage patterns (weekdays, weekends, seasonal variations).
• Controlled environment ensuring clean ground-truth labels for each appliance's energy signature.

Measured Parameters

• Active power (W): Real-time power consumption of each appliance at sub-minute resolution.
• Reactive power (VAR): Non-working power for characterizing inductive/capacitive loads.
• Voltage (V) and Current (A): Electrical parameters for each device.
• Power factor: Ratio of active to apparent power, useful for appliance fingerprinting.
• Energy (kWh): Cumulative consumption over time for billing simulation and total usage analysis.
• Timestamps: High-resolution time markers enabling time-series analysis and event detection (appliance on/off transitions).

Appliance Characteristics

• Refrigerator: Cyclic on/off compressor pattern with baseline always-on consumption.
• Washing machine: Multi-phase operation (fill, wash, rinse, spin) with distinct energy signatures per phase.
• Microwave: High-power short-duration bursts, easy to detect but variable usage frequency.
• Air conditioner: Variable load depending on thermostat settings and ambient temperature, major contributor to peak demand.
• Television: Relatively constant low-power consumption with standby modes.

Use Cases

• Non-Intrusive Load Monitoring (NILM): Training disaggregation algorithms to infer individual appliance usage from whole-house aggregate consumption.
• Energy management systems: Developing smart home controllers that optimize appliance scheduling to reduce costs and peak demand.
• Behavioral analysis: Understanding household energy consumption patterns and identifying opportunities for conservation.
• Anomaly detection: Identifying unusual appliance behavior that may indicate malfunctions or inefficiencies.
• Demand response: Simulating and evaluating strategies for load shifting and peak shaving in residential settings.