Prediction of Wastewater Treatment Plant Effluent Water Quality Using Recurrent Neural Network (RNN) Models
Round 1
Reviewer 1 Report
The work titled “Prediction of Wastewater Treatment Plant Effluent Water Quality Using Recurrent Neural Networks (RNN) Models” employs LSTM cells for predict the WWTP. This work identifies a suitable model for sequential data in WWTPs. The RMSE value is much smaller than that of the previous models. There are some comments as shown below:
(1) In Figure 3, what is the meaning of the expanded nodes from seven parameters? How to obtain or ensure the nodes?
(2) In data preprocessing part, how to treat the missing values for some data which is not integrity? Is there any standard?
(3) Is the equation 1) which was applied to rescale the data having any physical meaning?
(4) In line 5, Table number is missing.
(5) This work used 80% of the data for training and 20% for testing. How about using other ratio of data for training and testing? Will the model still predict the results well?
(6) As shown in Table 5, the RMSE of TP, TKN and NH3-N is small enough for prediction. Why the RMS value of TSS is much higher than other parameters, indicating relatively poor prediction.
It's ok
Author Response
Thank you for your comments and suggestions. These are very helpful for improving the manuscript. We adjusted and improved the manuscript and attached the responsed file here.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments to the authors:
In this study, the authors performed prediction of wastewater treatment plant effluent water quality using recurrent neural network (RNN) Models. Overall, the study is methodologically sound with promising results. However, prior to further consideration, there are few comments to be addressed:
(i) Absract: Very poorly written. Include a line or two about the research gap and problem statement in a precise manner. Current statements seems to be taken from the introduction directly. Highlight your novelty. Include more numerical results.
(ii) Introduction: Major revision is required. Highlight the research gaps/objectives you are addressing. Also, provide a literature table will be very helpful for readers to understand your work. The story does not flow well
(iii) Performance criteria section. Advice to cite papers instead of website. For instance
(a) Application of artificial intelligence methods for monsoonal river classification in Selangor river basin, Malaysia
(iv) What optimization methods applied?
(v) Novelty of the study is not too significant.
(vi) No in-depth discussion is provided. Seems to be presentation of results only. More discussion is needed. Such as why certain method outperforms another.
(vii) Draw a taylor diagram for graph illustration.
(viii) Provide implications of the study in the conclusion sections. Also, highlight the future studies.
(ix) Conclusion section seems to be repetition of the results section. Huge modifications are required. Please make sure your ‘conclusion’ section underscores the scientific value added of your paper, and/or the applicability of your findings/results, as indicated previously. Please revise your conclusion part into more details. Basically, you should enhance your contributions, limitations, underscore the scientific value added of your paper, and/or the applicability of your findings/results and future study in this section.
Moderate editing of English language required
Author Response
Thank you very much for your comments and suggestions. These are very helpful for improving my manuscript. We adjusted and improved the manuscript. The attached file is the point-by-point response for your review.
Thank you again for your time and consideration.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Comments to authors:
The authors have substantially addressed my comments. However the introduction still requires some modification, particularly on the clarity about the research gap and objectives. There are some relevant references to be included in the manuscript. (i) Comparison among different ASEAN water quality indices for the assessment of the spatial variation of surface water quality in the Selangor river basin, Malaysia. (ii) Application of artificial intelligence methods for monsoonal river classification in Selangor river basin, Malaysia. (iii) Water quality index modeling using random forest and improved SMO algorithm for support vector machine in Saf-Saf river basin
Also, there are still grammatical errors detected in the study. Please send it for proofreading prior to further consideration.
Minor editing of English language required
Author Response
Thank you very much for your comments. Per your suggestions, we have adjusted and addressed the relevant references to indicate the research gap. In addition, grammatical check was also applied throughout the work.
Author Response File: Author Response.docx
