Energy consumption assessment of LED displays based on mapping method

High-intensity large LED displays are prevalent in outdoor advertising lighting and account for a significant portion in building energy consumption. However, the accurate measurement of their power usage is challenging, since it varies with the display outputs. Typically, for tunable multi-spectral LED displays, different light colors and luminance correspond to different energy consumption, but the exact relationships are not clear. This study investigates the energy consumption of LED displays under a series of spectral settings. We collected hyperspectral data from LED display modules under different display conditions, analyzed the correlation between display spectra and energy consumption estimations, and established mapping relation between color and energy consumption. The results show a great linearity and additivity between energy consumption and the radiance of each color channel. We extend these correlations to display colors under different luminance. This work also proposes a convenient energy consumption metric to address the estimation issue of multi-color display outputs, with an average error of 2.52%. This work provides a convenient and non-destructive method for measuring the energy consumption of LED displays, significantly reducing facility management costs in urban commercial districts with numerous displays, and thereby facilitating a balance between urban development and environmental monitoring.