Tracking Control Method for Greenhouse Environment Prediction Model Based on Real-Time Optimization Error Constraints

Round 1

Reviewer 1 Report

The work is interesting to be read, the research presents some level of novelty and the mathematical model is well defined.

However, the presented models do not present a very high level of novelty.

Similar models of predictive control are already presented in the specialized literature on similar systems.

The bibliographic references are much too outdated.

To generate high interest for readers the research requires some experimental methods for results  validation

Author Response

1 However, the presented models do not present a very high level of novelty.

The work done in this article is to improve the general greenhouse environment prediction model by adding tracking error state variables to form a new multi degree of freedom model. Emphasis is placed on proposing optimization methods to better implement optimized predictive tracking control on this model.

2 Similar models of predictive control are already presented in the specialized literature on similar systems.

The main content of this paper is to propose a method to reduce predictive tracking control time, simplify the predictive tracking control process, save predictive tracking control losses, and improve error convergence speed, that is, to optimize predictive tracking control strategies.

3 The bibliographic references are much too outdated.

Literature revision has been conducted

4 To generate high interest for readers the research requires some experimental methods for results  validation

In order to verify the effectiveness of the optimized predictive tracking control method proposed in this article, it is necessary to validate its use on an effective model. The effective model is selected from a literature that has established an actual predictive model based on experimental conditions and data at the time, which has already proven the effectiveness of the model's prediction. I do not have specific experimental data, so I am unable to conduct experiments with specific experimental data. Please understand.I hope you could understand this.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript addresses a simple and fast method for optimal predictive tracking control based on the discrete state space models of greenhouse microclimate environmental factors. The method shows that this control strategy can increase the convergence speed of performance indicators such as state variable increment, tracking error, control variable increment, and performance indicator function, so that the output variable can quickly reach the tracking set value, and the maximum value of the performance indicator function does not change much, demonstrating the effectiveness and rationality of it.

 

 

More specifically about the content of the paper, these are my concerns:

 

Please summarize the novel contributions of the current work, preferably point-wise.

What are the remaining probable causes of error in the suggested prototype? Please elaborate.

Please, include further information regarding the limits of the proposed design.

The bibliography is not deep enough, and above all, too old.  It would be appropriate to check the most recent progress, to make a comparison with the proposed method.  References must be written according to the same standard. Please correct them.

What is the result obtained by the other methods present in the literature? How to evaluate the results in comparison to real situations?

The article needs a fundamental review. The abstract and conclusion sections should be rewritten. In these two sections, the innovation of the design must be clearly stated. In the introduction section, similar and more up-to-date articles should be examined.

I would recommend completely revising the usage of the English language, terminology, and writing style, as there are some errors in the text.

Author Response

1 Please summarize the novel contributions of the current work, preferably point-wise.

(1) According to the general equation of the discrete state space model of the greenhouse system, the state variables are transformed into their corresponding increments, and the tracking error state equation is increased to form a multi degree of freedom discrete time state space model.

(2) The performance index function with tracking error and control variable increment as variables is established, which is converted into the form of optimal value of quadratic form function.

(3) Introduce the gradient of tracking error into conventional constraint conditions, combine gradient descent learning rate with vector norm, and use gradient descent theory to determine the upper and lower bounds of the constraint conditions.

(4) Substitute the real-time data at the current time into the upper and lower bounds of the constraint conditions. By iteratively updating the upper and lower bounds of the constraint conditions, the rolling optimization of predictive control is used to solve the optimal value of the quadratic form function, and the optimal tracking control rate at the future time is continuously obtained.

(5) In order to verify the effectiveness of the optimized predictive tracking control method proposed in this article, an actual predictive model established in a certain literature was selected. This actual predictive model was established based on experimental conditions and data at the time, and the simulation verified that the control method was correct.

2 What are the remaining probable causes of error in the suggested prototype? Please elaborate.

If the original model does not add real-time constraints, but only simple constraints, the control rate solved will become slow, the convergence speed of tracking error will become very slow, and it cannot quickly track the set value. If the constraint range is not set properly, it may cause the control input increment loss to increase, and the tracking error will increase and then decrease. These are all shown in the simulation in the figure.

4 The bibliography is not deep enough, and above all, too old.  It would be appropriate to check the most recent progress, to make a comparison with the proposed method.  References must be written according to the same standard. Please correct them.

Literature modifications have been made

5 What is the result obtained by the other methods present in the literature? How to evaluate the results in comparison to real situations?

The methods mentioned in the literature model predictions from different perspectives, and these prediction methods require specific conditional environments. The prediction results have been shown in the literature. Apply the method proposed in this article to the predictive control model, and verify its feasibility through simulation results. The methods in the literature and those proposed in this article are based on different conditions, and can only be compared in terms of the complexity and economic cost of the prediction method.

In order to better evaluate the method proposed in this article, an actual predictive control model established in a literature was selected. Due to the fact that the model in the literature is an actual prediction model that has been validated, applying the proposed method to this model through simulation can prove the actual situation of this method.

6 The article needs a fundamental review. The abstract and conclusion sections should be rewritten. In these two sections, the innovation of the design must be clearly stated. In the introduction section, similar and more up-to-date articles should be examined.

Corresponding modifications have been made

 

Author Response File: Author Response.docx

Reviewer 3 Report

Dear Authors

Some brief suggestion

Title; Even I saw results on climate factors such as humidity or temperature and calefaction, It is no clear if you are using experimental data. So, Pherhaps yiu can add "theorical" study?

Abstract: Same; include limitations of study and be focused in the objective

Introduction: In opinion of this reviewer, literature revistion must be enhancent, at least in the results found by the researchers cited. As wrote in the manuscritp, that will be necessary in the discussion section

M&M: One point basic in this section is the posibility of reproduce your metodology, I think is clear the model description, however may be is convenient add a paragraph respect the initial data and previous treatment

Results: I think, the figures presentation will get better, jus enhalting the most representative results

Discusion: Authors must be expose your main results and comment what abaiut with previoosu and actual reserach in the same topic

Specific comments are in the manuscript

 

 

 

Comments for author File: Comments.pdf

Author Response

1 Title; Even I saw results on climate factors such as humidity or temperature and calefaction, It is no clear if you are using experimental data. So, Pherhaps yiu can add "theorical" study?

Experimental data was not used in this article. This article proposes an optimized predictive control method, which uses an actual predictive model established in a literature to validate the method. Based on the experimental conditions and data at that time, this actual predictive model is established. I do not have specific experimental data and cannot use it for experiments. Although utilizing some theoretical knowledge to achieve optimized predictive control, it seems that it cannot be called theoretical research.

2 Abstract: Same; include limitations of study and be focused in the objective

The main focus of this study is to establish a real-time constraint range based on the known greenhouse system environment prediction model, solve for the optimal control rate, achieve higher prediction accuracy than the greenhouse environment prediction model, and achieve fast tracking results.Corresponding modifications have been made

3 Introduction: In opinion of this reviewer, literature revistion must be enhancent, at least in the results found by the researchers cited. As wrote in the manuscript, that will be necessary in the discussion section

The literature has been revised

4 M&M: One point basic in this section is the posibility of reproduce your metodology, I think is clear the model description, however may be is convenient add a paragraph respect the initial data and previous treatment

The simulation uses a prediction model established based on actual data and has good predictive performance in practice.This prediction model demonstrates the effectiveness and practical value of this method.

5 Results: I think, the figures presentation will get better, jus enhalting the most representative results

I think simulation is easily to see the effectiveness of the method

6 Discusion: Authors must be expose your main results and comment what abaiut with previoosu and actual reserach in the same topic

I have previously studied using neural networks to predict greenhouse environmental parameters and crop growth in greenhouses

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

changes were made to the paper to some extent. the mathematical part presented is well outlined and its application on to the model leads to the expected results. however, the bibliographic part is somewhat deficient and should be updated together with the introduction part. a comparison with other models would be very useful

Author Response

Modifications have been made in the paper

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript must be conforming to the journal's format.

The usage of orthographic punctuation marks must be done with caution.

Author Response

Thank you for your suggestion. It has been modified. If there are any areas that have not been modified, please provide the specific location of the paper.

Author Response File: Author Response.docx