Artificial Neural Network (ANN) Modelling for Biogas Production in Pre-Commercialized Integrated Anaerobic-Aerobic Bioreactors (IAAB)

Round 1

Reviewer 1 Report

The manuscript summarizes the pre-commercialised Integrated Anaerobic-aerobic bioreactor (IAAB) operational condition and the Biogas Production situation. The authors used Artificial Neural Network assessment, empirical models and experimental measurements to elucidate the mechanisms and performance. But some revisions should be made before accepted for publication. Detailed comments are listed as follows:

1. Please pay attention to the format through the manuscript.

2. The description needs to be more concise in introduction section.

3. The scientific hypothesis should be clearly elucidated to emphasize the importance of this research.

4. Please pay attention to the format for Figure 6, Figure 8, Figure 11, Figure 13, Figure 15, Figure 17. It is recommended to modify these Figures due to the yellow line (the experimental data line) cover the blue line (the prediction value line).

5. The indicator of ammonium nitrogen and total phosphorus is important for anaerobic and aerobic bioreactor system. It is recommended to compare the prediction value and experimental data for ammonium nitrogen and total phosphorus.

6. There are some grammar mistakes in the manuscript and the authors should check the whole manuscript very carefully to avoid any mistakes. (e.g. Line 47 “such as such as”). 

7. Some references cited are old and irrelevant to this research. Please replace with the recent related publication reference. (e.g.  doi: 10.1016/j.watres.2020.116576; doi: 10.1016/j.scitotenv.2019.05.060; doi: 10.1016/j.jhydrol.2020.125440). 

Author Response

Please refer to the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Overall, the paper is interesting and contains novel points. I have only two minor remarks to the authors:

1. Based on which analysis the authors chose ANN’s input parameters?
2. The paper is too long and should be reduced. Some of the Figures can be omitted.

Author Response

Please refer to the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Artificial Neural Network (ANN) Modelling for Biogas  Production in pre-commercialised Integrated Anaerobic aerobic bioreactor (IAAB) - 1499043

 

General comment

This manuscript studied the use of artificial neural network to predict the COD removal, methane purity, methane yield of AD of aerobic process and result showed that COD inlet is the most influential input parameters affecting the methane yield, while for aerobic process COD removal was most affected by MLSS.  The idea was good. However, the manuscript is simple with lots of linguistic concerns, and the scope of the paper is too narrow and suffers from a lack of novelty, and all the significant findings and conclusions are straight forward and very simple to be a complete research work. The presentation of this work is terribly below standard. Graphs are poor. Results are disorganized. What is new in this work? ANN has been applied to AD in several works. Results are similar to some other previous works. Therefore, I suggest rejection or extensive revision before considering this work for publication. Some senior researchers from the unit has to at least see the work before submission. More specific comments…….

• The abstract is too long. Summarize only the key findings and add some numerical results.
• Why do you prefer ANN? It is black box model complex to understand the process in the six months period.
• Common meaningless phrases ….”….with severe inaccuracies that are not…..”
• There is no relevant literature review in the introduction part. The introduction part is inflated with various sentences not essential at all. The first three paragraphs can be merged and summerized in to one great introduction.
• Customize Fig.1 and Include all the IAAB parameters and results in Figure 1 or delete it because you have Fig. 3
• Table 1 is unnecessary
• Figures 2, 5-8 is not professional, carless graphs … learn to use some softwares like OriginPro,SigmaPlot etc
• Section 3.1 and 3.2 are not results … and some other sections too
• I couldn’t follow the results because there is no flow ….. results are poorly organized

Comments for author File: Comments.docx

Author Response

Please refer to the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The quality of this manuscript is poor for me. Specially the presentation is difficult to follow. 

Author Response

Thanks for the suggestions. The abstract is shortened. The problem statement has been refined in the Introduction part. To make the presentation clearer, some subtitles have been re-arranged and renamed. Some sections are combined and re-organised. A concluding remark is added in Section 3.5.