Flexible Zoom Telescopic Optical System Design Based on Genetic Algorithm

Round 1

Reviewer 1 Report

The authors used the genetic algorithm to design the flexible zoom telescopic optical system.

This paper is interesting and combines the current hot topics. But for the publication, there are some questions that need to be addressed.

1. The GA section is missing references. Because this manuscript only uses genetic algorithms, citations need to be added when the algorithm is introduced.

2. As the manuscript mentioned, six structural parameters are real numbers. Compared with GA’s binary coding, why don’t you use real number encoded genotype of evolution strategy (ES) or covariance matrix adaptation evolution strategy (CMA-ES)?

 

3. The layout of Figure 9 and Figure 10 is too messy. Please reorganize the figures and provide new figures with higher resolution.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

-The work of Liu Z. et al. presents the application of genetic algorithms to determine the initial structure and solution of a liquid lens configuration to provide a starting point for complex configurations and design. The manuscript is well written, and I consider it relevant for developing liquid lens-based systems. However, the description of the central topic, genetic algorithms, is a bit light and needs to be expanded.

Therefore, I consider that the manuscript can be accepted after minor revisions.

I am sending some specific questions related to

-  I suggest improving the introduction by mentioning some   works that use genetic algorithms to solve problems in  lens design, optics, and photonics

 

 -        Although genetic algorithms are a well-known optimization method, authors should support section 3.3 Genetic Algorithm with appropriate citations.

 -        In the Genetic algorithms section, the authors do not use the common names or usually names of parameters. For example, the genotype, or the set of parameters that define a proposed solution to the problem,  is called a chromosome, and each value in the chromosome is called a gene. I recommend that the authors review various works on genetic algorithms to use the most appropriate language so readers can understand it more easily.

 -        In the part of the individuals' evolution and crossover (section 3.3.1), the authors perform a genotype conversion from decimal to binary. There are two types of individuals (e, ). However, it is not clear to me if those two types of individuals can be reproduced between them or not. Moreover, can individuals e_1 be produced with e_2, e_3, and e_4 individuals? Or just between those of the same type? It would be helpful if the author made this clear of explained it more explicitly in the manuscript

 -      In section 3.3.4, the authors mention that the parameters, in the fitness function were experimentally adjusted. I think it would be helpful if the authors extended their explanation of this topic and explained how they did it

 -        What was the algorithm programming language or environment?

-        Given that it corresponds to the initial configuration of a possible device, what is the convergence time of the algorithm?

-        Did you consider that the search space may be degenerate?

-        Can you ensure a unique solution, or are there multiple values of the genes that lead to the same solution?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

 The authors have modified the manuscript and added more content. The manuscript is greatly improved from the previous one. I recommend acceptance.