The transition to the new generation power grid, or “smart grid”, requires novel ways of using and analyzing data collected from the grid infrastructure. Fundamental functionalities like demand response (DR), that the smart grid needs, rely heavily on the ability of the energy providers and distributors to forecast the load behavior of appliances under different DR strategies. This paper presents a new model of aggregated water heater load, based on dynamic Bayesian networks (DBNs). The model has been validated against simulated data from an open source distribution simulation software (GridLAB-D). The results presented in this paper demonstrate that the DBN model accurately tracks the load profile curves of aggregated water heaters under different testing scenarios.
Revised: September 2, 2013 |
Published: July 19, 2012
Citation
Vlachopoulou M., G. Chin, J.C. Fuller, S. Lu, and K. Kalsi. 2012.Model for Aggregated Water Heater Load Using Dynamic Bayesian Networks. In Proceedings of the 2012 International Conference on Data Mining (DMIN 2012). WORLDCOMP '12, July 16-19, 2012, Las Vegas, Nevada, edited by R Stahlbock and GM Weiss, 17-23. Athens, Georgia:CSREA Press.PNNL-SA-86275.