Application of artificial neural-network to control the light of multi-color LED system

This paper presents the application of artificial neural-network (ANN) algorithm to control the light of multi-color light-emitting-diode (LED) system. Compared with conventional control methods, the proposed method has the merits of 1) not requiring an accurate system model, 2) achieving quality lighting with higher color rendering index (CRI), 3) requiring only one red-green-blue (RGB) color sensor for feedback, and 4) handling the change of flux, shift of wavelength with temperature and aging. The proposed method is introduced based on a buck-type single-inductor-multiple-output (SIMO) LED driver with channels connected in series. A prototype has been built and experimental results are given.