Computation speeds of different optical thin-film synthesis methods

Li, Li; Dobrowolski, J. A.

doi:10.1364/ao.31.003790

Cited by 43 publications

(23 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The population size is 50. These parameters are decided after the experiments have been conducted on some optical coating problems (Aguilera et al, 1988;Li and Dobrowolski, 1992;Sullivan and Dobrowolski, 1996) with various values.…”

Section: Resultsmentioning

confidence: 99%

“…The two basic approaches to the design of numerical optical coatings are re nement methods (Aguilera et al, 1988;Dobrowolski and Kemp, 1990) and synthesis methods (Dobrowolski, 1986;Li and Dobrowolski, 1992;Tikhonravov, 1993;Bovard, 1988). Re nement methods normally require a starting design, which is gradually modi ed to a desired solution.…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, we studied a lter with 0.0, 0.5, and 1.0 transmission regions within the m wavelength range. This problem was used to compare the FCEA to several synthesis methods (Dobrowolski, 1986;Li and Dobrowolski, 1992) on the performance and computation speed. The nal problem is the synthesis of a tristimulus lter matching the CIE used in the tristimulus colorimeters (Sullivan and Dobrowolski, 1996).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Optical Coating Designs Using the Family Competition Evolutionary Algorithm

Yang

Horng

Lin

et al. 2001

Evolutionary Computation

View full text Add to dashboard Cite

A robust evolutionary approach, called the Family Competition Evolutionary Algorithm (FCEA), is described for the synthesis of optical thin-film designs. Based on family competition and adaptive rules, the proposed approach consists of global and local strategies by integrating decreasing mutations and self-adaptive mutations. The method is applied to three different optical coating designs with complex spectral quantities. Numerical results indicate that the proposed approach performs very robustly and is very competitive with other approaches.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Optical Coating Designs Using the Family Competition Evolutionary Algorithm

Yang

Horng

Lin

et al. 2001

Evolutionary Computation

View full text Add to dashboard Cite

show abstract

“…A lot of efforts have been done to find the most effective optimization methods in order to obtain solutions close to the global minimum of the merit function. One can follow this dramatic process by a series of papers devoted to comparison of various modern design techniques [7][8][9] .…”

Section: Introductionmentioning

confidence: 99%

Usage of simulated annealing algorithm in design of optical thin film

Wang

Rong

2011

J. Shanghai Jiaotong Univ. (Sci.)

View full text Add to dashboard Cite

Simulated annealing algorithm is a mathematic model, which imitates the physical process of annealing. And optical thin film is widely used in many industry. Its design is difficult and can be regarded as an optimization problem. In this paper, we use the simulated annealing algorithm to design an edge filter, which is composed of 20 dielectric thin film layers with TiO2 and SiO2. The simulated annealing algorithm is a very robust algorithm for optical thin film design.

show abstract

“…One is called refinement or local method which needs a starting design and aims at speeding as fast as possible [1] . The other can be called synthesis or non-local method which can create a starting design by itself and mainly aims at enhancing optimal search capability [2][3][4][5][6][7] . In most of local methods based on merit function optimization, the accurate calculation and fast realization of merit function's gradient and Hesse matrix plays a key role in determining the actual speed of film design [8] .…”

mentioning

confidence: 99%

Accurate calculation and Matlab based fast realization of merit function’s Hesse matrix for the design of multilayer optical coating

Long

Yang

2009

Optoelectron. Lett.

View full text Add to dashboard Cite

To improve the current status of home multilayer optical coating design with low speed and poor efficiency when a large layer number occurs, the accurate calculation and fast realization of merit function's gradient and Hesse matrix is pointed out. Based on the matrix method to calculate the spectral properties of multilayer optical coating, an analytic model is established theoretically. And the corresponding accurate and fast computation is successfully achieved by programming with Matlab. Theoretical and simulated results indicate that this model is mathematically strict and accurate, and its maximal precision can reach floating-point operations in the computer, with short time and fast speed. Thus it is very suitable to improve the optimal search speed and efficiency of local optimization methods based on the derivatives of merit function.It has outstanding performance in multilayer optical coating design with a large layer number.The optimality and speed of multilayer optical coating design has always been one of the focuses in the field of thin film optics. Dozens of optical coating optimization methods have been put forward with proved practical significance. And all these methods can be mainly distinguished into two types. One is called refinement or local method which needs a starting design and aims at speeding as fast as possible [1] . The other can be called synthesis or non-local method which can create a starting design by itself and mainly aims at enhancing optimal search capability [2][3][4][5][6][7] . In most of local methods based on merit function optimization, the accurate calculation and fast realization of merit function's gradient and Hesse matrix plays a key role in determining the actual speed of film design [8] . However, for the complexity of the calculation of merit function's Hesse matrix of multilayer coatings, many kinds of film design software only take advantage of gradient information to realize first order optimization. Researches about second order optimization method are seldom reported. And there is obvious speed difference between second and first order methods especially when a large layer number occurs. And then low speed and poor efficiency of home film design software is caused.In order to improve the computation efficiency, based on the matrix method to calculate the spectral properties of multilayer optical coating, an analytic model is established theoretically for absolutely accurate calculation of the merit function's gradient and Hesse matrix with respect to the design parameters. And the corresponding fast computation is successfully achieved by programming with Matlab, with the maximal precision of floating-point operations in the computer. The model is very suitable to improve the optimal search speed and efficiency of local optimization methods. And it has outstanding performance in multilayer optical coating design with a large layer number. The concept of modern film design is generally featured with constructing a merit function, with the structure and param...

show abstract

Computation speeds of different optical thin-film synthesis methods

Cited by 43 publications

References 17 publications

Optical Coating Designs Using the Family Competition Evolutionary Algorithm

Optical Coating Designs Using the Family Competition Evolutionary Algorithm

Usage of simulated annealing algorithm in design of optical thin film

Accurate calculation and Matlab based fast realization of merit function’s Hesse matrix for the design of multilayer optical coating

Contact Info

Product

Resources

About