Optimization of Multistage Coilgun Based on Neural Network and Intelligent Algorithm
Round 1
Reviewer 1 Report
Title
Optimization of Multistage Coilgun Based on Neural Network and Intelligent Algorithm
Abstract:
The abstract is clear and provides an overview of the methodology and results.
Introduction
Need to provide a clear hypothesis and objective, it is said that “ In this paper, the particle swarm optimization (PSO) [15] algorithm based on the prediction model of the gated recurrent unit (GRU) neural network [16] is used to optimize the feed time of MSSICG”, but it’s not clear how this optimization will influence the MSSICG
There is a brief description of the methodology and exected results at the end of the introduction: “The training and test data sets are selected by the standard orthogonal design table. After the training is completed, the error of the prediction model on the test data set is within a tolerable range. Finally, through algorithm optimization, the number of coils required is reduced to a minimum under the premise that the muzzle velocity of SICG reaches the target value. ” It needs to be replaced or better explained in order to prevent confusion for the reader
Simulation Model
It is presented a Singlestage Coilgun, and the modeled is one Multistage Coilgun. In the Multistage Coilgun, the distance between the coils is crucial because one coil interferes in the current wire of others. This interference is not presented in the experimental verification.
Another problem is on line 138 “After repeating the experiment three times, they are still almost the same.”. The results are compared without any experimental methodology, such as ANOVA. Therefore it can not be concluded that they are “almost the same”.
Recurrent Neural Network
Present an overview of the RNN used, with citations only in the first paragraph; even though all section presents an explanation, it's not clear if the attempt at improvement was already implemented in previous works or how the parameters of the GRU wore chosen.
Predictive Model
It is used a numerical simulation to generate the data set for training and testing the proposed model.
Figure 11 has an error in the legend
Parameter Optimization
PSO is used to optimize the feed time of the drive coil of MSSICG.
Presents a comparison of two methods for calculating the muzzle velocity of a coilgun and provides quantitative results to support the use of the prediction model and PSO in terms of computational efficiency. 2.1%, with only 0.015s, although its not present the design of experiments used for this analysis.
Conclusions
The conclusion suggests that the method can be effective but misses the analysis and comparison with previous works, besides it says that it can be used for 2 - 10 coils it only presents results regarding 3 coils.
Please clearly indicate the innovation of this paper in the conclusion part.
Clearly indicate the improvements over other similar research work in the conclusion part.
Overall
-
The manuscript is clearly articulated, although the level of originality is not yet fully established. Further investigation could be conducted to strengthen the novelty of the findings.
-
The study's significance is introduced in the introduction, and the feasibility of implementation is demonstrated through computational analysis.
-
To provide a more comprehensive understanding, additional details regarding the experimental design and simulations involving multiple coils should be included in the manuscript.
-
The manuscript would benefit from a thorough review to minimize the use of colloquial language and enhance its academic tone.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
see the atached file.
Comments for author File: Comments.doc
Author Response
Please see the attachment.
Author Response File: Author Response.docx