Figure 6
A bar graph shows the Edge A I modelʼs error distribution in energy forecasting, peaking at 0.00 kilowatt-hours.The bar graph is labeled “Error distribution - Edge A I Model (Random Forest).” The vertical axis is labeled “Frequency,” ranging from 0 to 50 in increments of 10. The horizontal axis is labeled “Prediction Error (kilowatt-hour),” ranging from negative 0.05 to 0.10 in increments of 0.05. The solid bar starts at (negative 0.08, 3.48), peaks at (0.00, 57.46), and ends near (0.13, 1.10). The overlaid curve starts at (negative 0.08, 1.74), peaks at (0.00, 39.5), and ends near (0.13, 0.78). Note: All numerical data values are approximated.

Distribution of prediction errors from the edge AI model in forecasting energy consumption. Source(s): Author’s work

or Create an Account

Close Modal
Close Modal