Application of Local Search Particle Swarm Optimization Based on the Beetle Antennae Search Algorithm in Parameter Optimization
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors propose a cooperation of two Nature-inspired algorithms and its application.
- eq4 - as I know, there is only one global best solution in PSO - why there is counter i=1:N?
- eq5 - why three different symbols for Dimensionality is used?
- eq11 - it is not clear what distribution was applied to sample random number
- fig1 - fig3 use different fonts and graphics for the same diagrams - why?
- What does the initial best position g = 1?(in Alg1)
- The English is at a low level ('dimention', etc.)
- The manuscript contains a lot of mistakes ('The dimensionality was Ddim = 301005001000')
- Fig.4 does not specify which dimension is mentioned
- Fig.4 h - why did the red curve stop before 1000 generations? It stagnates?
- The statistical comparison for the simulated results is necessary (compare the methods by '<' symbols is not acceptable
- The results do not contain some interesting conclusions/recommendations
Comments on the Quality of English Language
The English needs a significant update
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper presents an approach consisting of stochastic optimization (Particle Swarm Optimization (PSO)) and heuristic algorithm (Beetle Antennae Search (BAS)). As a result, the aim of the paper is to solve the challenges in the PSO approach, such as the failure to achieve the overall optimal solution, low search speed, and low accuracy.
The general points that can be mentioned after reviewing the paper are as follows:
1) In the abstract section, the issues and drawbacks of the optimization approach and how to solve these challenges are not clearly stated.
2) In the introduction section, the review of past studies is too long, while the challenges and solutions to the problem are not stated.
3) In optimization approaches, one of the existing challenges is managing existing uncertainties and how to model them. In this paper, in the innovative algorithm introduced, the location of the food (points of search) that is detected by the antennas has uncertainty, and there is no mention of modeling this uncertainty in this paper.
4) The PSO approach has shortcomings such as not achieving an optimal solution, low search speed, and low accuracy. Combining this approach with heuristic and meta-heuristic algorithms leads to improvement and improvement of the results. In the simulation section, the introduced approach should be compared with other innovative algorithms to show the superiority of this approach.
5) The flowcharts designed in Figures 2, 3, and 4 do not have a clear quality.
6) Table 1 should be placed in the middle of the page.
Comments on the Quality of English LanguagePlease polish the paper and make the narration fluent
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsStill, the statistical comparison is not provided - the scientific soundness is low.
Comments on the Quality of English LanguageI am not qualified to evaluate English on the professional level.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsNo further comments
Comments on the Quality of English LanguageNo further comments
Author Response
Thanks for your suggestions.
Round 3
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors still did not provide statistical comparison - it means testing of the statistical hypothesis that is standard in the field of newly prepared optimisation algorithms.
Comments on the Quality of English LanguageI am not an expert of English, but I am rather satisfied.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf