Forecasting electricity load: Interview with Prof. Tao Hong

Load forecasting is an important application of forecasting in energy. Prof. Tao Hong is the Director of the Big Data Energy Analytics Laboratory (BigDEAL) at University of North Carolina at Charlotte. In the interview, Prof. Hong (www.drhongtao.com/) discusses the special characteristics of load forecasting and what is needed for successful solutions.
This video supports the textbook Practical Time Series Forecasting.
www.forecasting...
www.galitshmuel...

[@BattleBotOfficial]

Thanks to our lovely professor.

[@wasiusulaimon1147]

Hello Prof, given an electric utility dataset with demand and temperature, can short-term load forecasting be done with only the demand data without considering the temperature data?

[@marcosyamasaki]

I would say definitely yes. I'm working on load forecasting using the US utility dataset which prof. Dr. Tao Hong mentioned (is free to get btw), and I could predict a demand output without temperature data. However, the results are not great, I got low accuracy, but it's probably due to lack of seasonal data and others which may increase the prediction scores.
As the professor said, using temperature data makes more sense, cause lots of power demand it's related to heaters and coolers. But, sometimes you just don't have them, which makes a bit harder to do accurate prediction so.
Anyhow, you can get the dataset from here:
www.iso-ne.com/isoexpress/web/reports/load-and-demand

[@chriscockrell9495]

Too bad that chainsaw was running in the background.