Characterizing energy flexibility of buildings with electric vehicles and shiftable appliances on single building level and aggregated level. (September 2022)

• Record Type: Journal Article
• Authors: Azizi, Elnaz; Ahmadiahangar, Roya; Rosin, Argo; Bolouki, Sadegh
• Is Part Of: Sustainable cities and society. Volume 84(2022)
• Publication Date: 2022-09
• Subjects: Demand response -- Electric vehicle -- Load monitoring -- Flexibility -- xg-boost method; Sustainable urban development -- Periodicals; Sustainable buildings -- Periodicals; Urban ecology (Sociology) -- Periodicals; 307.76
• DOI: 10.1016/j.scs.2022.103999
• Languages: English
• ISSNs: 2210-6707

Abstract

Abstract: Residential energy flexibility is considered one of the efficient concepts to alleviate the ever-increasing concerns of better balancing supply and demand. A positive assumption that all buildings have the same energy flexibility potential, is not applicable in a realistic situation especially when direct load control is not applied for each. This paper proposes a novel approach to characterize the energy flexibility of shiftable appliances and EVs (as two main sources of energy flexibility) protecting consumers' data privacy and considering the usage behavior. First, an xg-boost regression is utilized to non-intrusively extract the consumption of appliances. Then, the uncertainty of the power values and the operation time of each appliance is computed based on the extracted consumption patterns. Finally, a price-based DR model is used to determine their energy flexibility and prioritize them according to the change in their operating time before and after optimization. Case studies are conducted and results prove that the proposed method is computationally cost-effective and outperforms other methods in terms of accuracy, customer privacy and comfort. Moreover, the results show that the proposed model can significantly decrease the flattening signal up to 3% for each residential building and up to 25% in the residential aggregated level. Highlights: The xg-boost algorithm not only eliminates the need for a large amount of training dataset but also increases the disaggregation accuracy compared to previous studies. Residential buildings manifest high differences in the provision of the demand flattening potential. The power values distribution and the consumer's usage behavior are two main factors that affect the probability of energy patterns occurring during a day. The extracted results can be utilized by aggregators in pre-defined price-based DR models to prioritize the consumers based on their energy flexibility potential and incentive them more smartly.