Estimation of Chromaticity Coordinates for LEDs Array by Modulation of Red or Yellow LEDs with Artificial Neural Network

Abstract

The modulation of color temperature of illumination always use the red or yellow LED to adjust the white LED. The research was to use the Artificial Neural Network to estimate the chromaticity coordinates of 3x2 LED array mixing by red and yellow LEDs. There are four algorithm methods: Levenberg-Marquardt (LM), Broyden Fletch Goldfard Shan no (BFGS), Fletcher-Reeves Conjugate Gradient (FRCG) and Scaled Conjugate Gradient (SCG) to train the learning function. The modulation color temperature was control at 2700K. The result showed that the artificial neural network model has a good and stable predictive ability by FRCG method.