New Data-Driven Models of Mass Flow Rate and Isentropic Efficiency of Dynamic Compressors

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The reviewer notes the thoroughness of the text and the analysis of a large number of previously developed models. There is no reason to doubt that the authors have solved the problem of more accurate simulating the characteristics of existing compressors. However, the process of simulating characteristics is not related to the workflow of dynamic compressors. The authors could apply the same techniques to the characteristics of a turbine, internal combustion engine, and any other technical device, the working process of which is described by the dependence of one parameter on another. Anyway, this kind of work is needed to solve some problems using dynamic compressors, and the work should be published.

The reviewer is not an expert regarding this type of work. As a compressor specialist and simply as an attentive reader, the reviewer could make the following comments.

The introduction does not fully represent the scope of application of dynamic compressors and has no relation to the essence of the work.

The authors of the work, like the authors of most of the presented alternative models, simulate the parameters of dynamic compressors, presented to the user as gas-dynamic characteristics. Such naivety brings a sad smile to a compressor specialist: - dimensional gas-dynamic characteristics can be simulated formally but it is long and dark way.

The right way is to simulate non-dimensional gas-dynamic characteristics. They are one-way connected to dimensional characteristics. When simulating the non-dimensional gas-dynamic characteristics, you may to relate on known facts:

- work coefficient characteristic  is linear,

- efficiency characteristic  is parabolic (for subsonic compressors, at least),

- the influence of the Mach number is well known; simulation of its influence does not present any great difficulties. The remaining similarity criteria can be ignored in this case.

So the problem posed can be solved more correct from a scientific point of view, and more simply by relying on the fundamentals of the theory of dynamic compressors.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The study discusses the development and initial validation of a data-driven approach for compressors mass flow rate and efficiency prediction. The manuscript is in general well-structured and well-motivated. However, additional effort is necessary to enhance the paper's quality before it can be considered for publication.

0) ABSTRACT

The key feature, differentiating the present technique from the others available, should be clearly stated. While the outcomes are compared, it is not clear how the methodological approach is different.

1) INTRODUCTION

While the literature review is substantial and comprehensive, the part concerning the present study falls short in describing the methodology and how it is superior to the existing alternatives.

3) EVALUATION OF THE REVIEWED MODELS FOR APPLICATIONS TO AERO ENGINE AND ECS COMPRESSORS

Figures 1 to 5: if they are taken from the reference literature, the authors should clearly state it in the captions.

4) DEVELOPMENT OF NEW EMPIRICAL MODELS FOR DYNAMIC COMPRESSOR PERFORMANCE

The authors should detail how the genetic algorithm is employed and the corresponding computational setup (is it NSGA? How many generations and individuals?). The obtained results show improved prediction accuracy regarding specific cases for which data are available, denoting the necessity to calibrate the models at any case study (e.g. by eliminating some coefficients). However, since the method is intended as a design tool, it would be interesting to consider how the models perform over novel design configurations. Have the authors considered this analysis? If so, what are the criteria to employ the models with non-calibrated coefficients to produce characteristic maps of non-existing compressors?

 

 

 

Comments on the Quality of English Language

I am not a native speaker, but I think that a general check for minor grammatical mistakes and typos should be done.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Until Chapter 4, the work is truly a review-type paper.

The new methods proposed by the authors do not seem to be sufficiently developed; the models should be presented in more detail, including a comprehensive presentation of the compressors and the experimental data used in these models to more clearly demonstrate the validity of these models.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I think the authors have done a remarkable job in revising the manuscript. In my opinion, the paper can be published in present form.

Comments on the Quality of English Language

I am not a native speaker, but I think that a general check for English language before publication may benefit readability.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have addressed my requirements, and the paper has been appropriately improved.