Neural Network Model for Greenhouse Microclimate Predictions

Round 1

Reviewer 1 Report

In the study, the authors present an ANN-based application to better manage the efficiency and energy consumption of a greenhouse, which will ensure that the greenhouse climate is stable without being affected by changing outdoor parameters. The study contains an application for an important need. However, some improvements are needed.

1) Detail the importance of computer-aided systems for greenhouses. Knowledge of precision farming is required. You can use the article below.
A Comprehensive Survey of the Recent Studies with UAV for Precision Agriculture in Open Fields and Greenhouses
2) Dots should be used after references.
3) The resolution of all images should be increased.
4) Equations 1 and 2 should be explained in more detail. Information should be given about all parameters.
5) With Figure 7, is Figure 8 necessary?
6) The difference in the estimated value in Fig 10 and Fig 11 is not clear. For this, zoom the graph.
7) Discussion Section required. For example, can ANN model parameters be determined by different optimization techniques? Discuss.
8) Use and compare different methods such as SVM, KNN, decision tree, etc. instead of just one ANN. More methods are needed. This provides more comprehensive results. For example, the following study is an example. You can cite.
Human action recognition with bag of visual words using different machine learning methods and hyperparameter optimization

Author Response

Dear reviewer,

1. The reference has been accepted.
2. Corrected.
3. It was corrected, with the maximum resolution provided by MATLAB.
4. Corrected.
5. We think Figure 8 has a higher representation of our model and must remain.
6. Accepted.
7. Section "Results and Discussion" has been expanded according to the reviewer's comments.
8. The reference has been added to the references section. Using a variety of modeling methods is beyond the scope of this research and will be future work.

Author Response File: Author Response.pdf

Reviewer 2 Report

All comments and corrections are inserted in text.

Please pay attention to References which must be edited.

Comments for author File: Comments.pdf

Author Response

Dear reviewer,

all your comments have been accepted.

Author Response File: Author Response.pdf

Reviewer 3 Report

The study may be interesting in order to model the internal temperature and relative humidity of an agricultural greenhouse. The authors designed Multi-Layer Perceptron Neural Network. The results of such investigations have practical value.

However, I have a number of suggestions:

S1. The introduction should include some facts and figures about the current situation with greenhouses production.

S2. The authors should add a strong Related works section on existing methods for solving the stated task.

S3. Authors should provide a link to an open access repository with the dataset used for modelling.

S4. Authors should argue the chosen pre-processing method. Why didn't they choose another method?

S5. The conclusion section should be extended using: the limitations of the proposed approach; prospects for future research.

S6. Discussion may need to focus on what was achieved from this research and compare it with the literature.

S7. A lot of references are outdated. Please fix it by using 3-5 years old papers in high-impact journals

Author Response

Dear reviewer,

1. Our research has been enriched with facts from relative publications.
2. A strong Related works section on existing methods for solving the stated task has been added.
3. Done.
4. This has been further analyzed in the sub-section "Data and Methodology"
5.  An analysis of this research's limitations has been added.
6. Accepted as per your comments.
7. Corrected.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors made the necessary revisions. The manuscript is acceptable.

Reviewer 3 Report

The authors could add more references