A simple design of uniform LED illumination using catadioptric collimator and freeform lenslet array

We introduce a compact lenslet array principle that takes advantage of freeform optics to deploy a light distributor, beneficial for highly efficient, inexpensive, low energy consumption light-emitting diode (LED) lighting system. We outline here a simple strategy for designing the freeform lens that makes use of an array of the identical plano-convex lenslet. The light is redistributed from such lenslet, hinging on the principle of optical path length conservation, and then delivered to the receiver plane. The superimposing of such illumination area from every lenslet occurs on the receiver plane, in which the non-uniform illumination area located in the boundary should have the same dimension as the size of the freeform lenslet array. Such an area, insofar, is negligible due to their small size, which is the crux of our design, representing a large departure from the former implementations. Based on simulations that assess light performance, the proposed design exhibited the compatibility for multiple radiation geometries and off-axis lighting without concern for the initial radiation pattern of the source. As simulated, the LED light source integrated with such proposed freeform lenslet array revealed high luminous efficiency and uniformity within the illumination area of interest were above 70% and 85%, respectively. Such novel design was then experimentally demonstrated to possess a uniformity of 75% at hand, which was close to the simulation results. Also, proposed indoor lighting was implemented in comparison with the commercial LED downlight and LED panel, whereby the energy consumption, number of luminaires and illumination performance were assessed to show the advantage of our simplified model.

Nonimaging optical design with supporting quadric method and deep neural network

Background: The supporting quadric method(SQM) is a versatile method for designing freeform optics for desired irradiance redistribution, but the consumed time of solution optimization increases rapidly with the refinement of the mapping grid. Objective: As the complexity of light distribution is getting higher and higher, time-consuming will also increase exponentially. This paper proposes an idea of applying the deep neural network method to optical design. Methods: In this article, we established a special corresponding relationship and made a data set. Hand over to deep network learning and training. Finally, a hybrid design method of deep learning and optical design was realized and verified. Results: Compared with the traditional method, this method is more efficient. Here we use a deep neural network(DNN) to accelerate the freeform optical design. After the DNN was trained by a sample set consisting of a uniform pattern and eight different Chinese characters represented by an array with 11 × 11, it can generate a character's reflector within few milliseconds. Conclusion: As proof of this new method, a character pattern reflector was manufactured and tested, and the experimental irradiance distribution is closed to the expectation, which means the neural network has the excellent capability to memorize all of the learned characters. SQM combined with DNN has the potential to establish a particular “optical font library” and even offers a promising path for rapid freeform optical design to realize the function of “optical typography”.