DEDDIAG: Domestic Electricity Demand of 50 Appliances in 15 German Homes

50 individual appliances were recorded at a 1 Hz frequency in 15 German homes for up to 3.5 years. The dataset, created by Marc Wenninger, focuses on appliances significant for load-shifting like dishwashers and washing machines. Manual ground truth event annotations for 14 appliances provide precise start and stop timestamps.

Use Cases

• Training energy disaggregation models based on the high-frequency (1 Hz) appliance-level recordings.
• Analyzing load-shifting potential based on the detailed usage patterns of appliances like dishwashers and washing machines.
• Validating event detection algorithms using the manual ground truth annotations for 14 appliances.
• Studying long-term electricity consumption behavior based on the up to 3.5-year recording period across multiple homes.

Strengths

• Contains up to 3.5 years of longitudinal data from 15 homes, providing a substantial temporal scope.
• Includes manual ground truth event annotations for 14 appliances, offering a precise validation resource.
• Records 50 individual appliances at a high 1 Hz frequency, enabling detailed temporal analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.
• Row count is unknown, which may limit suitability assessment for some modeling tasks.

Provenance

Source

Marc Wenninger

Collection Method

Recordings from 15 homes in Germany.

Time Range

Up to 3.5 years (specific dates unknown).

Geography

Germany