Non-imaging Optical System Design Method And Its Applications In The Led Road Lighting

Light Emitting Diode (LED) is considered as the fourth-generation light source after Flame, Incandescent and Fluorescent in the history of mankind. Compared with traditional light sources, LED has high energy conversion efficiency, long life, small size, environment friendly and other advantages. Non-imaging Optics with its short history (borned in 1960s), has developed quickly with the development of solar energy technology. And it has been widely used in solar energy collection, illumination system design and even in imaging optics. In the illumination design the main issues concerned is how to guide the light from the source to the target efficiently and with expected pattern. There are three methods to solve the problem: multiparameter optimization, partial differential equations and the simultaneous multiple surface method (SMS). Through these methods we can design LED illumination systems with high efficiency, thereby make the LED come into people's life possible. But the only efficient optical design is not enough, because LED heat sensitive, how to conduct the heat out from LED must also be resolved. This paper mainly focuses on these two aspects of LED illumination system design.In the research, we adopted the idea of region division for the optical design of LED street lights. By using the currently available lenses to adjust the radiation pattern of LED, and let each LED illuminates one specific region, then add them altogether to achieve the uniform illumination on the street. Base on the multiparameter optimization method we have invented a method of sampling points dynamicly changing with the light sourse increasing step by step. By using the method we invented, it reduces the number of variable parameters in the process of optimization, improves the pertinency of the illumination merit function, so that substantially enhances the efficiency of optimization.Then we use the thermal design software Qfin4 to design and optimize a heat sink for a single LED with a minimum mass under restriction of maximum thermoresistance. In the end we made a integrated cooling structure of the LED lights, and tested its related properties.