The bus-split method for residential load aggregation

This paper presents the implementation and performance testing of the bus-split method for aggregating power demands of energy storage appliances (i.e. heating units and water heaters) in residential loads. The tested method is based on modifying the bus-split formulation to replace load models with the power demands of constant-impedance and constant-power (P_Z and P_S) load components. The values of P_Z and P_S are determined by the ZIP phaselet method, which can provide P_Z and P_S for each reading of the household power meter. The combination of the ZIP-phaselet method and bus-split aggregation can eliminate the need for measuring the power demands of individual energy storage appliances, thus simplifying the implementation of the the bus-split aggregating of residential loads. The bus-split method has been implemented for performance evaluation using data collected from 20 households during the Fall, Winter, Spring, and Summer seasons. Performance results show that the developed aggregation method can provide accurate, simple, and non-intrusive aggregation of the power demands for energy storage appliances. Moreover, test results show that the bus-split method has minor sensitivity to the type and/or ratings of aggregated appliances, along with negligible sensitivity to seasonal variations of household power demands.