Aspects of illumination system optimization

Koshel, R. John

doi:10.1117/12.562917

Cited by 8 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[2] and [3], while employing the techniques discussed in Ref. [1]. The algorithm is the simplex method of optimization, which is akin to a ball rolling down a hill to locate the local minimum, but these references proposed an improved simplex method that converges more rapidly.…”

Section: Optimization Proceduresmentioning

confidence: 99%

“…11 As Ref. [1] indicates, there are a multitude of additional limitations in illumination system optimization. The following ones in conjunction with parameterization are studied within this paper:…”

Section: Introductionmentioning

confidence: 98%

“…As Ref. [1] indicates this procedure can be quite tedious, if not impossible, thus a method to parameterize the illumination components is needed a priori.…”

Section: Introductionmentioning

confidence: 99%

“…1 Chief among the limitations is parameterization of the software model comprising the illumination system. Typically, illumination systems are constructed from complex geometrical forms, including Boolean solids, non-uniform rational B-splines (NURBS), faceted optics, and small-scale structure.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Optimization of parameterized lightpipes

Koshel

2007

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

Parameterization via the bend locus curve allows optimization of single-spherical-bend lightpipes. It takes into account the bend radii, the bend ratio, allowable volume, thickness, and other terms. Parameterization of the lightpipe allows the inclusion of a constrained optimizer to maximize performance of the lightpipe. The simplex method is used for optimization. The standard and optimal simplex methods are used to maximize the standard Lambertian transmission of the lightpipe. A second case presents analogous results when the ray-sample weighted, peak-to-average irradiance uniformity is included with the static Lambertian transmission. These results are compared to a study of the constrained merit space. Results show that both optimizers can locate the optimal solution, but the optimal simplex method accomplishes such with a reduced number of ray-trace evaluations.

show abstract

Section: Optimization Proceduresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

“…As Ref. [1] indicates this procedure can be quite tedious, if not impossible, thus a method to parameterize the illumination components is needed a priori.…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Optimization of parameterized lightpipes

Koshel

2007

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

show abstract

“…The main steps in the nonimaging optimization procedure are (1) the parameterization of the optical system, including the definition of the constraints in the parameters; (2) the definition of the merit function (MF) to be minimized or maximized; 4 and (3) the selection of the optimization algorithm, for which the Nelder-Mead algorithm particularly produces a robust and convergent method in nonimaging optimization problem. 5 In this paper, we focus our attention on the merit function, as it has the role to drive the optimization procedure; improvements in the capabilities of the merit function will improve the results of the optimization procedure. The most common way to build merit functions involves the weighted sum of squares of the differences between a set of objectives and their associated target values:…”

Section: Introductionmentioning

confidence: 99%

Application of dynamic merit function to nonimaging systems optimization

et al. 2015

View full text Add to dashboard Cite

Abstract. Automatic optimization algorithms have been recently introduced as nonimaging optics design techniques. Unlike optimization of imaging systems, nonsequential ray tracing simulations and complex noncentered systems design must be considered, adding complexity to the problem. The merit function is a key element in the automatic optimization algorithm; nevertheless, the selection of each objective's weight, fw i g, inside the merit function needs a prior trial and error process for each optimization. The problem then is to determine appropriate weights' values for each objective. We propose a new dynamic merit function with variable weight factors fw i ðnÞg. The proposed algorithm automatically adapts weight factors during the evolution of the optimization process. This dynamic merit function avoids the previous trial and error procedure by selecting the right merit function and provides better results than conventional merit functions.

show abstract