Great video!
Great video!
Thanks!
Nice Video — sets out the issues clearly.
Thank you!
Really cool work Akshay! I am curious to know what the end goal of the ANN simulation is. In the end, are you trying to build a more accurate prediction model of energy consumption then models that already exists and what are the applications of this new model?
Rishee,
Thanks for the question! Our aims are two-fold regarding the ANN model. Firstly, we do aim to develop a better and more granular model of residential demand. We are able to do this due to the robust data set provided by collaboration with Pecan Street Inc., as well as manipulation of ANN structure and input variables. Secondly, we aim to deploy these forecast models in actual demand response applications, such as curtailment incentive programs, etc.
Furthermore, the onset of such technologies such as energy storage and electric vehicles provide rich dynamics to the modeling problem, and we are very excited to handle these modeling challenges!
Akshay,
You mention that ambient temperature, relative humidity, and hour of the day were highly correlated to residential energy consumption. Did you study any other parameters to determine their correlation?
Cory,
Thanks for the interest! We did investigate other parameters, such as day of the week, which provided a poor fit, and we hope to include month of the year in the future. With only a year of data so far, we were unable to include the seasonal effects that a month input would have,
When we incorporate solar generation into the model, we’ll include phenomena such as solar irradiation in as well.
Further posting is closed as the event has ended.
Judges and Presenters may log in to read queries and replies.