YAG: Ce3+ phosphor/silicone composites are widely used in solid-state lighting as a light converter to achieve white lighting. However, because of high thermal resistance and low thermal stability, the luminous performance of YAG: Ce3+ phosphor/silicone composite deteriorates rapidly when excited by high-power-density blue-laser. To explore the potential of blue laser-excited YAG: Ce3+ phosphor/silicone composites, the luminous performances under different blue laser power conditions were characterized by both the reflective and transmissive excitations using a self-built three-integrating-sphere system. Furthermore, the Monte-Carlo Ray-tracing simulation was used to illustrate the light-transmission and energy conversion mechanism in the phosphor/silicone composites. The results showed that: (1) The YAG: Ce3+ phosphor/silicone composite could be excited by the 0.292W laser light with the peak wavelength of 445nm, excessive laser power will cause phosphor thermal quenching and silicone carbonization. (2) The luminous flux of the composite under both the reflective and transmissive excitations gradually increased with the increase of phosphor concentration; correspondingly, the color coordinate moved to the yellow region, and the Correlated Color Temperature (CCT) gradually decreased. (3) The simulation results indicated that under the same phosphor concentration, the luminous flux obtained by reflection excitation was largely higher than that by the transmission excitation, as the light re-conversion and strong back-scattering were occurred in the reflective and transmissive laser excitation respectively.

The emission spectra of high color rendering phosphors, mixed with the yttrium aluminium garnet, silicon based oxynitride and nitride based phosphors, were predicted by the Lambert-Beer theory and back propagation neural network (BP NN). Firstly, the modified Lambert-Beer model was used to calculate the proportional coefficient of the emission spectra of the mixed phosphors in ratios. Next, the BP NN was implemented to train and predict the proportional coefficients. Finally, the prediction of the emission spectra of the mixed phosphors was estimated and verified by the experimental measurements. The results show that the prediction error fraction of the proportional coefficients can be controlled within 5%; the predicted emission spectra by BP NN keep high agreement with the experimental measurements with lower RMSE and Δxy as 0.019 and 0.0016, respectively.