Virtual energy demand data: Estimating energy load and protecting consumers´ privacy

To address global warming, the construction of energy-efficient houses that have low environmental impact is essential. Variations in energy load and energy efficiency can be estimated by numerical simulation using measured data on energy demand, for example, for a house newly equipped with solar panels and storage batteries. Data on energy demand is useful for estimating the variation in energy load, and this data can be acquired with a smart meter. However, high-resolution data on energy demand can expose consumers´ behavior patterns through the identification of specific appliances and the times that they are used within the household. Although the introduction of smart grid technology has been applauded, its deployment has been slowed by serious security vulnerabilities that could compromise consumers´ privacy. Furthermore, although cooperation among specialists from various fields is necessary for achieving environmental goals, data on energy demand is difficult to share between research institutes. We propose a method for generating data, which we call “virtual demand data”, which will allow data to be shared while protecting consumers´ privacy. Appliances used in consumers´ houses are identified differently in virtual energy demand data and in measured demand data. Moreover, virtual energy demand data is useful for estimating the variation of energy load and energy efficiency instead of measured demand data, because the distribution of measured demand data for a single day taken from a measurement period is statistically consistent with the distribution of virtual energy demand data.