Application of hybrid learning algorithm for optimization of LED lens design

In this study was performed optimization procedure of LED lens design by adaptive neuro fuzzy inference system (ANFIS). There are two objective functions in the optimization procedure: viewing angle and luminance uniformity. Optical design software was used for experimental procedure in order to ext...

Full description

Saved in:
Bibliographic Details
Published in:Multimedia tools and applications 2022-11, Vol.81 (28), p.40469-40488
Main Authors: Petković, Biljana, Resic, Sead, Petkovic, Dalibor
Format: Article
Language:English
Subjects:
Adaptive systems
Algorithms
Artificial neural networks
Blended learning
Computer Communication Networks
Computer Science
Data Structures and Information Theory
Design optimization
Design parameters
Fuzzy logic
Lens design
Lenses
Light emitting diodes
Machine learning
Multimedia
Multimedia Information Systems
Neural networks
Optical design
Optimization
Software
Special Purpose and Application-Based Systems
Viewing
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this study was performed optimization procedure of LED lens design by adaptive neuro fuzzy inference system (ANFIS). There are two objective functions in the optimization procedure: viewing angle and luminance uniformity. Optical design software was used for experimental procedure in order to extract training data for the ANFIS optimization procedure. In the first step of the optimization procedure the viewing angle was used as the optimization objective function. Afterwards the initial optimization of the lens shape was found. In the next step the luminance uniformity was used as the second optimization objective function. ANFIS model was used to find the optimal parameters for the LED lens design. The optimal parameters were found based on minimization of the ANFIS prediction accuracy of the viewing angle and luminance uniformity. The selected optimal parameters could be used further for improvement of the LED lens design.
ISSN:1380-7501
1573-7721
DOI:10.1007/s11042-022-13116-3