Hybrid Digital Twin for Phytotron Microclimate Control: Integrating Physics-Based Modeling and IoT Sensor Networks

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper focuses on the application of the integration of the Internet of Things (IoT) and predictive modeling in the microclimate management of vertical farming in urban agglomerations, showing high innovation and forward - looking nature, which is in line with the development trend of agricultural technology. The proposed hybrid digital twin approach combines a physical microclimate model with a distributed IoT monitoring system. It parameterizes the heat and mass balance equations through a genetic algorithm and optimizes the model using real - time data from multiple sources, with a reasonable and practical design. The experimental results are excellent. The model performs well in terms of temperature deviation, relative humidity error, and energy consumption accuracy, and can reliably track climate and energy usage, strongly demonstrating the effectiveness of the approach. The research findings have clear practical significance, enabling precise environmental control and energy optimization in vertical farming and laying the foundation for scalable digital twins in controlled - environment agriculture.

However, there are some deficiencies in the research. The universality of the model needs to be further verified in different vertical farming scenarios. There is no cost - benefit analysis, which is unfavorable for evaluating the practical feasibility and promotion value. The long - term stability and reliability of the system and corresponding countermeasures are not discussed. There is also a lack of comparative analysis with existing similar methods. Overall, the paper has high research value, and the approach is innovative. If the above - mentioned deficiencies can be improved, it is expected to be widely applied in the field of vertical farming. 

Comments on the Quality of English Language

good

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1: This paper focuses on the application of the integration of the Internet of Things (IoT) and predictive modeling in the microclimate management of vertical farming in urban agglomerations, showing high innovation and forward - looking nature, which is in line with the development trend of agricultural technology. The proposed hybrid digital twin approach combines a physical microclimate model with a distributed IoT monitoring system. It parameterizes the heat and mass balance equations through a genetic algorithm and optimizes the model using real - time data from multiple sources, with a reasonable and practical design. The experimental results are excellent. The model performs well in terms of temperature deviation, relative humidity error, and energy consumption accuracy, and can reliably track climate and energy usage, strongly demonstrating the effectiveness of the approach. The research findings have clear practical significance, enabling precise environmental control and energy optimization in vertical farming and laying the foundation for scalable digital twins in controlled - environment agriculture.
Response 1: Thank you for pointing this out. We agree with this comment. To clarify these points, new sections "3.4. Model Generalizability and Adaptation", "3.5. Economic Analysis", "3.6. Comparison with Existing Microclimate Control Approaches" were added to Chapter 3 Results and Discussion, starting on page 21 of the new version of the manuscript from line 641:

3.4. Model Generalizability and Adaptation

The proposed digital twin framework was developed and validated for a basil phytotron. The following adaptations are possible to improve the model’s versatility:

• The transpiration model (Eq. 9) can be adapted for different crops by varying the values of Itransp, Atransp, and Lcycle based on species-specific physiological data.
• The model’s heat balance equations (1–3) scale linearly with room volume and surface area, making them applicable to a variety of facilities from small research chambers to commercial-scale vertical farms.
• The outside temperature model (Eq. 10) can incorporate regional climate data, with the tdelay value adjusted based on local building thermal mass characteristics.
• The energy models (Eqs. 11–15) use common power ratings and efficiency factors, allowing for substitution of different HVAC systems, lighting technologies, and dehumidification equipment.

3.5. Economic Analysis

The initial cost to build such a monitoring system includes:

• LoRaWAN sensors (8×S2101 + 1×S2103): ~$1,200
• Gateway and weather station: ~$800
• Smart plugs and meters: ~$400
• Orange Pi and software development: ~$500

Total initial cost: ~$2,900

At this stage, our digital twin demonstrates the capabilities of high-precision monitoring and modeling. The economic benefit will be determined only after the implementation and validation of optimization algorithms in future work.

The cost of implementing the monitoring system (~$2,900) represents the minimum investment required to build the digital twin infrastructure. This lays the foundation for future optimization efforts. Future work will focus on developing and testing optimization algorithms for:

• Dynamically adjusting the temperature setpoint depending on the growth stage of plants
• Predictive control to minimize the cycling of HVAC systems
• Optimizing the LED spectrum depending on environmental conditions

Only after implementing and testing these algorithms will it be possible to quantify the real economic benefits.

3.6. Comparison with Existing Microclimate Control Approaches

Modern digital twin models show a root-mean-square error of temperature forecasting below 1.3 °C in tests [43]. When validating CFD models around 1 °C [44], neural network approaches show accuracy at the level of RMSE = 0.787 °C [45] for temperature and 0.62% for humidity [46].

The results of our economic policy, however, have several advantages:

1. Ensure the accuracy of the calculation optimization parameters at first.
2. Computational requirements are reduced compared to CFD or ML methods.
3. Practical cost is reduced due to the use of commercial Internet sensors.
4. Adaptability at first is possible to take into account the optimization of genetic algorithms.

Unlike CFD models, which require extensive computing resources and detailed 3D geometry, our approach provides increased accuracy with standard computing equipment. Compared to a pure machine learning approach, our physics-based framework provides stable interpretability and requires less training data.

And also a new item in section 3.7 Future Work (line 710)
• Develop and test resilience mechanisms: periodic sensor calibration, LoRaWAN packet loss monitoring, and channel redundancy, as well as a strategy for automatic switching to backup controllers in the event of a node outage.

We hope that our revisions fully address your comments and that the manuscript is now suitable for publication.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This paper is topical, linking the crop sector with the IT sector.

The model in this paper works well using average sensor readings (with a temperature deviation of <0.1 °C and a humidity error of <=2%), but this could hide a heterogeneity of the microclimate, which is very important for plant health. Differences of 1–3 °C from real life observed on the phytotron indicate that localized conditions may not be as well controlled or modeled as claimed. This raises concerns about plant-level accuracy.

The paper talks about energy efficiency with an accuracy of 99.5%, but in my opinion, an accurate energy correlation does not guarantee optimal or sustainable efficiency in the long term (due to the fact that using genetic algorithms on limited time frames may not capture seasonal or operational variations).

The architecture uses LoRaWAN sensors and Wi-Fi-based smart plugs, which are flexible but can be technically difficult to integrate and are not industrial-grade for commercial-scale deployments. I think using Home Assistant may not be viable for large-scale operations that require IT support, redundancy, and security standards typical of agritech enterprises.

We note that while climate variables are modeled, there is little empirical validation of actual crop performance (e.g., yield, biomass, quality) in response to the system. Without this connection, the effectiveness of the digital twin in optimizing agronomic outcomes remains theoretical. Microclimate modeling is different from plant response modeling.

The article does not address the maintainability of the system, i.e. its security and long-term maintenance.

In conclusion, I agree with the publication of this article.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.
Comments 1: 

This paper is topical, linking the crop sector with the IT sector.

The model in this paper works well using average sensor readings (with a temperature deviation of <0.1 °C and a humidity error of <=2%), but this could hide a heterogeneity of the microclimate, which is very important for plant health. Differences of 1–3 °C from real life observed on the phytotron indicate that localized conditions may not be as well controlled or modeled as claimed. This raises concerns about plant-level accuracy.

The paper talks about energy efficiency with an accuracy of 99.5%, but in my opinion, an accurate energy correlation does not guarantee optimal or sustainable efficiency in the long term (due to the fact that using genetic algorithms on limited time frames may not capture seasonal or operational variations).

The architecture uses LoRaWAN sensors and Wi-Fi-based smart plugs, which are flexible but can be technically difficult to integrate and are not industrial-grade for commercial-scale deployments. I think using Home Assistant may not be viable for large-scale operations that require IT support, redundancy, and security standards typical of agritech enterprises.

We note that while climate variables are modeled, there is little empirical validation of actual crop performance (e.g., yield, biomass, quality) in response to the system. Without this connection, the effectiveness of the digital twin in optimizing agronomic outcomes remains theoretical. Microclimate modeling is different from plant response modeling.

The article does not address the maintainability of the system, i.e. its security and long-term maintenance.

In conclusion, I agree with the publication of this article.

Response 1: Thank you for pointing this out. We agree with this comment.
To clarify the above points, explanations were added about the system security requirements (line 456):
As a proof‑of‑concept, the current implementation uses SenseCAP sensors integrated with Home Assistant. For commercial‑scale deployments, we recommend industrial‑grade LoRaWAN gateways (e.g., MultiTech, Kerlink) with MQTT over TLS encryption and clustering support, alongside high‑availability brokers such as EMQX or VerneMQ and CI/CD pipelines for automated firmware‑ and container‑image updates. In addition, IT teams should establish clear zones of responsibility, implement role‑based access control, conduct regular access audits, and maintain systematic data backup and recovery procedures.
The need to take into account seasonal fluctuations to validate the flexibility of the model (line 635):
It should be noted that the genetic‑algorithm optimization performed over the three relatively short intervals (72–110 h) provides only a “snapshot” tuning of model parame-ters. To capture seasonal variations and equipment wear, we recommend re‑optimizing over longer time series (at least 1–3 months) and incorporating stability criteria into the objective function (for example, compressor smoothness or minimization of on/off cycling amplitudes).
As well as points in the Future Work section (line 713):

• Extend the current single‑zone model to a spatially resolved digital twin—initially via a simple 2×2 grid of zones and subsequently through CFD‑based approximations—to capture shelf‑ and row‑level microclimatic variations and improve local control fidelity.
• Expand calibration periods to cover longer datasets (minimum 1–3 months) and introduce multi‑objective optimization that balances energy efficiency with equipment longevity and minimizes negative control pulsations (e.g., amplitude of compressor switching cycles).

We hope that our revisions fully address your comments and that the manuscript is now suitable for publication.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

This study presents a hybrid digital-twin approach that integrates a physical climate model with a distributed IoT monitoring system to simulate and control the phytotron environment. The research topic holds significant practical value, particularly for vertical farming applications. However, I have several concerns that should be addressed to strengthen the manuscript:

1. The study focuses on microclimate control, yet no measurements or predictions were taken near the plant canopy. The sensors were installed on the racks rather than in close proximity to the plants, which may limit the accuracy of environmental monitoring.
2. At line 26, the chosen intervals (72 h, 90 h, and 110 h) lack a consistent delta. A justification for selecting these specific time intervals should be provided.
3. The statement at line 189, "It is conventionally referenced to the specific heat capacity of water," requires a supporting reference.
4. In Figure 1 (line 230), Reference 37 was derived from outdoor grapevines. Given that the study focuses on indoor leafy greens, its suitability should be discussed or alternative references should be considered.
5. There appears to be an incorrect equation reference at line 233.
6. The sentence at line 266 repeats the same information as line 265. One of them should be removed or rephrased.
7. At line 267, the authors mention the use of a genetic algorithm, but the results directly obtained from this method are not discussed. A more detailed analysis of these findings would strengthen the paper.
8. A flowchart illustrating the modeling approach would improve clarity and help readers better understand the methodology.
9. It would be beneficial to show the growing racks in Figure 3. Additionally, are there any sensors measuring conditions near the plant canopy?
10. The current versions of Figures 5, 6, 12, and 13 are difficult to interpret. Replotting them with improved labeling, legends, or annotations would enhance readability.

Comments on the Quality of English Language

Minor changes needed.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.
Comments 1: This study presents a hybrid digital-twin approach that integrates a physical climate model with a distributed IoT monitoring system to simulate and control the phytotron environment. The research topic holds significant practical value, particularly for vertical farming applications. However, I have several concerns that should be addressed to strengthen the manuscript:

1. The study focuses on microclimate control, yet no measurements or predictions were taken near the plant canopy. The sensors were installed on the racks rather than in close proximity to the plants, which may limit the accuracy of environmental monitoring.

Response 1: Thank you for pointing this out. We agree with this comment. Added clarifying notes (line 533):

- This arrangement of sensors was chosen for uniform distribution throughout the room, taking into account their number. For ease of installation and quick access, as well as to limit the influence of lighting fixtures on temperature readings, the sensors are mainly located on the stand. Since the automation system and digital twin rely on the average value of indicators throughout the room, the issue of the influence of plant mass and illumination at individual points was not considered, but is planned for study when moving to modeling macro- and microzones.
Comments 2: At line 26, the chosen intervals (72 h, 90 h, and 110 h) lack a consistent delta. A justification for selecting these specific time intervals should be provided.
Response 2: Added clarifying notes (line 581):

Initially, the selected intervals were tied to daily cycles and were 3, 4 and 5 days of one spring season. However, due to several power outages at the facility, such truncated intervals with high-quality data were determined to test the selected hybrid approach, build and refine the model.

Comments 3: The statement at line 189, "It is conventionally referenced to the specific heat capacity of water," requires a supporting reference.
Response 3: Indeed, the wording in the translation is incorrect. The explanation has been corrected as follows (line 187):

- an adjustable model coefficient, initially anchored to the specific heat capacity of water and subsequently determined via an optimization procedure. It takes into account the combined heat retention effect of all structural elements of room, the nutrient solution, etc.
Comments 4: In Figure 1 (line 230), Reference 37 was derived from outdoor grapevines. Given that the study focuses on indoor leafy greens, its suitability should be discussed or alternative references should be considered.
Response 4: Added clarifying notes (line 228):

The transpiration data were taken from open-ground grapevine experiments solely to illustrate the daily cycle and the sinusoidal pattern of variation. During model setup, this coefficient is then optimized, ensuring that the approach remains universal for different crops even when experimental data are lacking.
Comments 5: There appears to be an incorrect equation reference at line 233.
Response 5: Added clarifying notes (line 238)
Comments 6: The sentence at line 266 repeats the same information as line 265. One of them should be removed or rephrased.
Response 6: Added clarifying notes (line 276):

To ensure convergence of the computational model and align it with the real system, we employ a heuristic search method—the genetic algorithm. This algorithm tackles optimization and modeling problems by randomly selecting, recombining, and mutating the target parameters using mechanisms analogous to natural selection.

Comments 7: At line 267, the authors mention the use of a genetic algorithm, but the results directly obtained from this method are not discussed. A more detailed analysis of these findings would strengthen the paper.
Response 7: The current work focuses on the system architecture and the methodology for constructing a digital twin. A detailed study of optimization procedures is planned as part of further research work.

Comments 8: A flowchart illustrating the modeling approach would improve clarity and help readers better understand the methodology.
Response 8: Added flowchart at line 422.

Comments 9: It would be beneficial to show the growing racks in Figure 3. Additionally, are there any sensors measuring conditions near the plant canopy?
Response 9: Unfortunately, the resulting diagram is too cluttered. Therefore, for clarity, additional photos were added with the location of sensors 1-3 and 2-3 (Fig. 6 a and b). Sensors 1-2 and 2-2 are located symmetrically on another rack. 1-2 and 1-3 measure conditions near the plant canopy.

Comments 10: The current versions of Figures 5, 6, 12, and 13 are difficult to interpret. Replotting them with improved labeling, legends, or annotations would enhance readability.

Response 10: The figures shown are screenshots of the actual interface of the manufacturers' web services and the software developed. They have been redrawn to improve readability, new legend elements and axis captions have been added.

We hope that our revisions fully address your comments and that the manuscript is now suitable for publication.

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

The article "Hybrid Digital Twin for Phytotron Microclimate Control: Integrating Physics-Based Modeling and IoT Sensor Networks" considers modeling of microclimate systems in a phytotron room equipped with an air conditioning system, temperature and humidity control.
This manuscript provides methodological solutions to a number of issues related to the uniformity and possibility of reproducing climate conditions in a conditionally closed system.
The work has a number of methodological problems that are not sufficiently clear from the presented materials.
It is well known that the basis of carbohydrate nutrition is CO2, in the absence of its inflow, plant growth is disrupted and can stop, and in some cases can lead to oppression. This parameter depends on pressure and temperature, since this is due to the dissolution of gases in liquid. The graphs provided do not reflect this picture. On the other hand, the phenomenon is widely known that on weekdays the amount of CO2 in buildings increases due to human breathing. This is not observed on weekends. Plants also consume different amounts of this gas as they grow, calculated per biomass, with a constant increase in absolute values. I do not find a correlation with these banal observations. The authors should explain this. The rationality of using a room of such a large area for such non-optimally located vegetation racks raises very big questions.
They are related to the adequacy of temperature distribution, the impossibility of comparing plants from different tiers, and the lack of light characteristics and intensity in different parts of the chamber. Such unevenness is obvious with such a complex device and indicates the impossibility of using this equipment for scientific purposes. Let me give an example: is it possible to compare plants located at the edge of two closely spaced racks with those plants located in the free zone. In the proposed version, the degree of unevenness and the edge effect will have a number of reinforcing factors, and the differences will be significant, which can be seen in Figure 4.
Also, the microclimate will differ at the edges if the plants are close together or freely spaced. And also in terms of uniformity of humidity and temperature, since the air flows will differ in both temperature and speed. This will lead to uneven transpiration.
The authors should eliminate these defects in the description of the system.
Another problem is the lack of dynamics. Obviously, with low plants, the uniformity of adjustments will differ from what we would see in the middle of the plant's development period or with a fully developed plant. I strongly recommend looking at the diagrams and photos of the racks.
Temperature fluctuations of 0.1 degrees in such a system are simply unattainable.
Another problem is the lack of the effect of changing and modulating the spectrum - the effects of sunrise and sunset. The authors should have taken this into account.
Developing such equipment is often a very difficult task not because it is difficult to find a room or difficult to make racks. It is much more difficult to learn how to regulate light, temperature and humidity, and if you try to reproduce the necessary fluctuations in these parameters, then the reproducibility drops sharply. The system proposed by the authors may be interesting from the point of view of control efficiency, but only if the adjustment is carried out independently and over a wider range.

For this reason, I believe that the authors should review the manuscript, explain why they have plants of different ages on their shelves, provide measurements of direct and reflected light intensity, and provide data on the reproduction of daily fluctuations. After that, the article can be reviewed.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.
Comments 1:

The article "Hybrid Digital Twin for Phytotron Microclimate Control: Integrating Physics-Based Modeling and IoT Sensor Networks" considers modeling of microclimate systems in a phytotron room equipped with an air conditioning system, temperature and humidity control.
This manuscript provides methodological solutions to a number of issues related to the uniformity and possibility of reproducing climate conditions in a conditionally closed system.
The work has a number of methodological problems that are not sufficiently clear from the presented materials.
It is well known that the basis of carbohydrate nutrition is CO2, in the absence of its inflow, plant growth is disrupted and can stop, and in some cases can lead to oppression. This parameter depends on pressure and temperature, since this is due to the dissolution of gases in liquid. The graphs provided do not reflect this picture. On the other hand, the phenomenon is widely known that on weekdays the amount of CO2 in buildings increases due to human breathing. This is not observed on weekends. Plants also consume different amounts of this gas as they grow, calculated per biomass, with a constant increase in absolute values. I do not find a correlation with these banal observations. The authors should explain this. The rationality of using a room of such a large area for such non-optimally located vegetation racks raises very big questions.
They are related to the adequacy of temperature distribution, the impossibility of comparing plants from different tiers, and the lack of light characteristics and intensity in different parts of the chamber. Such unevenness is obvious with such a complex device and indicates the impossibility of using this equipment for scientific purposes. Let me give an example: is it possible to compare plants located at the edge of two closely spaced racks with those plants located in the free zone. In the proposed version, the degree of unevenness and the edge effect will have a number of reinforcing factors, and the differences will be significant, which can be seen in Figure 4.
Also, the microclimate will differ at the edges if the plants are close together or freely spaced. And also in terms of uniformity of humidity and temperature, since the air flows will differ in both temperature and speed. This will lead to uneven transpiration.
The authors should eliminate these defects in the description of the system.
Another problem is the lack of dynamics. Obviously, with low plants, the uniformity of adjustments will differ from what we would see in the middle of the plant's development period or with a fully developed plant. I strongly recommend looking at the diagrams and photos of the racks.
Temperature fluctuations of 0.1 degrees in such a system are simply unattainable.
Another problem is the lack of the effect of changing and modulating the spectrum - the effects of sunrise and sunset. The authors should have taken this into account.
Developing such equipment is often a very difficult task not because it is difficult to find a room or difficult to make racks. It is much more difficult to learn how to regulate light, temperature and humidity, and if you try to reproduce the necessary fluctuations in these parameters, then the reproducibility drops sharply. The system proposed by the authors may be interesting from the point of view of control efficiency, but only if the adjustment is carried out independently and over a wider range.

For this reason, I believe that the authors should review the manuscript, explain why they have plants of different ages on their shelves, provide measurements of direct and reflected light intensity, and provide data on the reproduction of daily fluctuations. After that, the article can be reviewed.

Response 1: Thank you for pointing this out. We agree with this comment.
To correct the above points, clarifications have been added regarding the lack of consideration of CO2 circulation in line 259:

CO₂ circulation within the system is not accounted for at this stage of our model because, in closed-environment agriculture, agronomists monitor the phytotron remotely and on-site visits are limited and infrequent. A dedicated CO₂ dosing system continuously supplements the chamber atmosphere, ensuring that diurnal concentration swings remain negligible. Consequently, even minimal human presence (typically a few staff per day) does not produce significant CO₂ fluctuations over a 24-hour cycle.
The issue of optimal placement of racks is considered in line 518:

It is worth noting that the two seven-tier racks were installed by the site contractor, and their placement was not determined by the authors. During construction, the equipment and crop supplier provided recommendations on the optimal spatial arrangement of racks, instruments, and sensors to ensure uniform airflow and favorable plant growth.

The issue of different ages of crops being grown is explained in line 520:

At the time of study, the phytotron was an active testbed performing concurrent cultivar trials. As a result, plants on different shelves varied in age. This mosaic of developmental stages reflects typical research-scale operations—larger production farms would zone by crop age. Assessment of the model’s sensitivity to varying biomass loads, however, is beyond the scope of the present work and will be addressed in subsequent studies.

Regarding the daily cycles and lighting, a clarification is given in line 546:

Because the chamber is fully enclosed and relies exclusively on LED fixtures, external solar-day effects (sunrise and sunset) on light intensity and spectral composition are effectively zero. In practice, light intensity remains constant throughout each 16 h photoperiod, simplifying model inputs and obviating the need to simulate natural dawn/dusk transitions.

Regarding the relevance of the obtained indicators and the formulation of the problem in the discussion of the results, explanations are given in line 632:

The primary concept and original goal of the virtual system was to accurately describe the real processes occurring within the existing physical facility, a target which has now been achieved. The digital twin models temperature dynamics with a resolution of ±0.1 °C. Although individual sensor readings may deviate by up to 1.5 °C due to localized hot and cold spots—as noted by the reviewer—these variations do not compromise the model’s effectiveness for the studied vertical farm. To address such spatial heterogeneity and ensure reliable averaging both system-wide and within local zones, we propose installing forced-ventilation ducts at strategic points on each level, which will homogenize the identified fluctuations.

We hope that our revisions fully address your comments and that the manuscript is now suitable for publication.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

I appreciate that most of my concerns have been resolved. Before finalizing the manuscript for publication, I have two minor points to address:

 

1. Regarding Comment 8: The newly included figure serves as a schematic overview of the experimental chamber, not a flowchart. Please revise this accordingly.
2. Line 567: The numeral ‘2’ in ‘CO2’ should be formatted as a subscript (CO₂).

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1:

Regarding Comment 8: The newly included figure serves as a schematic overview of the experimental chamber, not a flowchart. Please revise this accordingly.
Response 1:

Indeed, the figure was duplicated, a new diagram was added in line 437 (Figure 2).

Comments 2:

Line 567: The numeral ‘2’ in ‘CO2’ should be formatted as a subscript (CO2).

Response 2: The CO₂ notation has been brought to a uniform style using a subscript (lines 338, 448, 572, 583, 585, 586).

We hope that our revisions fully address your comments and that the manuscript is now suitable for publication.

Reviewer 4 Report

Comments and Suggestions for Authors

Article Hybrid Digital Twin for Phytotron Microclimate Control: Integrating Physics-Based Modeling and IoT Sensor Networks by
authors Vladimir Viktorovich Bukhtoyarov, Ivan Sergeevich Nekrasov, Ivan Alekseevich Timofeenko, Alexey Aleksandrovich Gorodov, Stanislav Alekseevich Kartushinskii, Yury Vasilevich Trofimov, Sergey Ivanovich Lishik modified the manuscript and made changes.
This manuscript is an important development of technical solutions in the field of plant cultivation in closed systems. Since this topic is relevant in the context of changing climatic vicissitudes, the work is relevant.
A little advice from practice - you should not try to eliminate unevenness locally, on the contrary, some manufacturers have now come to the conclusion that this issue is resolved by installing a perforated or lattice surface before entering the chamber, which allows you to reduce unevenness.
Please note that photo three and photo two are indistinguishable. Please do not take pictures of the installations at different angles, this looks untidy and does not improve the information content.

I strongly recommend replacing the word in line L738
"A GENETIC algorithm was applied to optimize the model". Obviously, this is acceptable in the conversational genre, but genetic mechanisms are still not related to the area under study. If you want to use philological techniques in a scientific publication to emphasize some logic of decisions, the word should be put in quotation marks.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.
Comments 1:

This manuscript is an important development of technical solutions in the field of plant cultivation in closed systems. Since this topic is relevant in the context of changing climatic vicissitudes, the work is relevant.

A little advice from practice - you should not try to eliminate unevenness locally, on the contrary, some manufacturers have now come to the conclusion that this issue is resolved by installing a perforated or lattice surface before entering the chamber, which allows you to reduce unevenness.

Response 1:

We sincerely thank the reviewer for this valuable practical recommendation. Indeed, perforated or lattice inlet surfaces are a well-established engineering solution to reduce airflow and microclimate unevenness. In response, we have revised the manuscript by explicitly noting this approach in the Discussion section and by adding a statement in the Future Work section. The revised text highlights that passive homogenization solutions can complement the proposed digital-twin-based strategy and should be comparatively assessed against active ventilation methods.

Line 659:
In addition to active solutions such as forced-ventilation ducts, passive homoge-nization techniques are also applied in practice. Specifically, several manufacturers employ perforated or lattice inlet surfaces that diffuse the incoming airflow before it enters the chamber, thereby reducing spatial gradients of temperature and humidity. Although such solutions were not implemented in the present study, they represent a proven engineering approach that can complement digital-twin-based control strate-gies..
Line 747:
•          Investigating the comparative effectiveness of passive homogenization methods (e.g., perforated or lattice inlet surfaces) versus active ventilation strategies. A systematic evaluation of these approaches in terms of airflow uniformity, energy efficiency, and cost will provide further guidance for practical deployment in large-scale vertical farming facilities.

Comments 2:

Please note that photo three and photo two are indistinguishable. Please do not take pictures of the installations at different angles, this looks untidy and does not improve the information content.

We hope that our revisions fully address your comments and that the manuscript is now suitable for publication.

Response 2:

Indeed, the figure was duplicated, a new diagram was added in line 437 (Figure 2).

Comments 3:

I strongly recommend replacing the word in line L738 "A GENETIC algorithm was applied to optimize the model". Obviously, this is acceptable in the conversational genre, but genetic mechanisms are still not related to the area under study. If you want to use philological techniques in a scientific publication to emphasize some logic of decisions, the word should be put in quotation marks.

Response 3:

We thank the reviewer for this comment. We have clarified our terminology throughout the manuscript. The term "genetic algorithm" is used in its established technical sense as a specific optimization methodology, not as a biological metaphor. To address any potential confusion, we have:

1. Changed the wording in the abstract (line 23)

A set of heat- and mass-balance equations governing the dynamics of temperature, humidity, and transpiration was implemented and parameterized using a genetic al-gorithm (GA) - an evolutionary optimization method - with real-time data collected over three intervals (72 h, 90 h, and 110 h) from LoRaWAN sensors (temperature, hu-midity, CO₂) and Wi-Fi-connected power meters managed by Home Assistant.

2. Added a more detailed description of the genetic algorithm methodology in line 284 (Section 2.1):

To ensure convergence of the computational model and align it with the real sys-tem, we employ a heuristic evolutionary optimization method—a genetic algorithm (GA). The GA maintains a population of candidate solutions (chromosomes), each representing a complete set of 20 model parameters. During each iteration, the popu-lation is evolved through three main operators:

• Selection, in which the fittest individuals are retained based on the objective function;
• Crossover, which recombines parameter values from selected individuals;
• Mutation, which introduces random variations to maintain diversity and avoid premature convergence.

The objective function minimizes the sum of squared errors between simulated and measured values of temperature, humidity, and energy consumption.

3. 3. Expanded the description at line 626 to: "optimization was performed using a genetic algorithm - a population-based metaheuristic that iteratively evolves a set of candidate solutions through selection, crossover, and mutation operations"
4. In line 769, the wording has also been corrected:

A genetic algorithm (GA) - an evolutionary optimization method - was employed to optimize the model parameters against real monitoring data, achieving high predictive accuracy (0.1 °C temperature deviation, 2% humidity deviation)

We hope that our revisions fully address your comments and that the manuscript is now suitable for publication.