Prediction of Drilling Efficiency for Rotary Drilling Rig Based on an Improved Back Propagation Neural Network Algorithm

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, BP network is selected as the basic model of prediction and improved, and good results are obtained. However, there are still the following key problems:

1. Although many intelligent prediction models based on machine learning exist, the author chose the more common BP model without comparing its effectiveness with other models in solving this problem.

2. Pearson correlation analysis was used to obtain the correlation values of screening drilling features, but the initial data set features were not clearly described, which may cause confusion to readers.

3. The format of R2 in Table 3 should be modified.

4. The ANN in line 638 is not directly related to the above BP network, and the author should make a distinction.

5. GA and PSO optimization methods are used to optimize BP network parameters, but they also have hyperparameters. How to determine them? And ask why choose these two methods the author does not explain clearly.

6. In general, the paper seems to use existing methods to predict drilling efficiency, and does not have the key points to attract readers.

Comments on the Quality of English Language

The language expression of the article is clear.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The work, in general, holds promise but still require a slight improvement to meet publication standards. The topic is interesting; nevertheless, the presentation of the work requires slight changes to make the concept easy to understand and useful for others. Therefore, I would like to offer some suggestions for improvement. Please consider them as positive feedback and try to implement them.

·        The current structure of the abstract makes it very difficult for the reader to understand the problem and main objectives of the research, which are considered to be very important parts that must be explained well at the high-level details. Therefore, please reconstruct the abstract, ensuring that it covers the following in the listed order: problem description, primary objective, methodology, and finally, please share the main findings of the study.

·        Please make sure to check the English of whole paper, there are a few grammar mistakes , for example this sentence “it is difficult to fully the impact of  many variables on the prediction”  

·        The topic overview part in the introduction section is a little long, please try to make it shorter and focus on the problem rather than general issues.

·        Section2.1 “ Overview of drilling hydraulic system of rotary drilling rig” it should be a part of the introduction

·        The last part of the introduction must be specified for your claims , the current one does not focus on that

·        For all the figures please try to add more details to them , for example fig1, it shows a diagram but with no details.

·        You used 4 evaluation parameters ( MSE, MAE etc) However, the question here , which one you will use at the end as main ?

·        Please provide a table which shows the features and the target of the model

·        Why they are a negative value of drilling rate ( the ones shown in fig 15)

·        In section 6.1 , you mentioned fig 25 and 26 , however these figs are not there, I think you may meant fig 17 and 18 .  

·        You conducted only one case study to validate your model, I would recommend you to add 3 or more cases

·         

Comments on the Quality of English Language

NA

Author Response

Please see the attachment.

Author Response File: Author Response.pdf