Multi-color LED spectrum and illumination matching method

Abstract

The invention discloses a multi-color LED spectrum and illumination matching method. The method is characterized by, to begin with, obtaining spectrum sample data of a single LED in each color through sampling and measuring; then, carrying out transition on spectrum and illumination to be matched to obtain corresponding target spectrum radiance distribution; and finally, carrying out iterative matching through a differential evolution algorithm to obtain a driving value of each LED. The invention also provides corresponding spectrum and illumination matching evaluation indexes for evaluating the matching results. The method solves the defects of an existing spectrum matching method, can realize illumination matching while carrying out spectrum matching, can obviously reduce calibration cost due to LED mixed light source maintenance and updating, and is simple to operate and easy to implement. Besides, the problem of spectrum change due to heat of an LED light source is solved in the matching process, and the method can meet requirement for spectrum and illumination simultaneous matching in the practical application.

Description

technical field [0001] The invention relates to the field of LED lighting, in particular to a method for matching spectrum and illuminance of multicolor LEDs. Background technique [0002] With the advantages of environmental protection, energy saving, longevity, and small size, LED light sources are gradually penetrating into various application fields, such as display, general lighting, plant cultivation, biomedicine, photovoltaics, and calibration of radiometric, photometric, and chromatic instruments. Most applications require the light source to provide a certain brightness or illuminance and also have certain spectral distribution characteristics. Therefore, it is of great practical application value to accurately provide the light source spectrum and illuminance that meet the specific application requirements. [0003] The narrow-band characteristics of the LED light source spectrum and its rich color distribution provide an important basis for realizing the tunable l...

AI Technical Summary

Problems solved by technology

At present, the methods to achieve spectrum matching through LED light sources mainly include: 1) The least square method is used to minimize the error between the target spectrum to be matched and the matching spectrum. This method is simple and easy, but for some matching situations, the least square method can obtain The matching ratio of some of the LEDs will have a negative value. Obviously, the negative value cannot be realized with the actual LED, and simply discarding the LED light source with a negative value will cause an increase in the matching error; 2) Using the iterative method of gradient descent, the The method is more sensitive to the setting of iteration parameters. Once the setting is unreasonable, it often causes the disadvantage of non-convergence of the iteration; 3) the genetic algorithm is used. Although this method can avoid the shortcomings of the above two methods, it is not easy to implement

[0004] The spectral characteristics of the LED light source will change with its heat generation, and the existing methods often assume that the spectral characteristics of the LED light source are constant, resulting in matching errors

In addition, none of the existing methods mentioned how to deal with the problem of illuminance matching while matching the LED spectrum

Images