Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1.The importance and research purpose of Atomization of liquid spray solutions through nozzles are introduced in detail, highlighting the innovation of research.

2.The research objectives, questions, and hypotheses are clear and easy to understand.

3.Provide a detailed and accurate description of the experimental method, including experimental materials, steps, conditions.

4.No relevant research cited, no indication that your research is based on existing research.

5.No schematic or flowchart of the experimental method provided.

6.The chart layout is messy and unclear.

7.The conclusion is somewhat innovative but lacks logical coherence.

 

Author Response

Reviewer 1.

 

Comments and Suggestions for Authors

1.The importance and research purpose of Atomization of liquid spray solutions through nozzles are introduced in detail, highlighting the innovation of research.

2.The research objectives, questions, and hypotheses are clear and easy to understand.

3.Provide a detailed and accurate description of the experimental method, including experimental materials, steps, conditions.

 

Added sentences about atomization from a flat fan nozzle and operating conditions used in the analysis early in the methods section.

“Videos of 21 different oil-in-water formulations being sprayed through a conventional nozzle (XR8002, TeeJet) were taken at two different pressures (40 psi, 10 psi) and different oil-phase concentrations (2.5 and 5.0%) were used. Thus, the two-phase fluid being sprayed was mostly water with a small fraction (2.5-5%) oil as representative of typical agricultural sprays. A HELOS Sympatec laser diffraction system (Sympatec GmbH, Clausthal-Zellerfeld Germany) was used to measure the resulting droplet size distribution upon atomization. The laser was aligned at the center of the spray located below the continuous sheet at 330 mm below the nozzle (thus the sheet is fully broken up into droplets. The flat fan nozzle used was an XR8003 (TeeJet Technologies) which is a conventional nozzle with a spray angle of 80°.  Further details of the experimental system can be found elsewhere (9).

 

4.No relevant research cited, no indication that your research is based on existing research.  Not clear what is meant by this comment as a lot of relevant citations, including those from the authors) were used.

Not sure what is meant by this comment since a lot of references are added, including several from the authors discussing earlier work.

5.No schematic or flowchart of the experimental method provided.

Added comments (above) to the Experimental section with details where further information can be found, e.g., “Details of the experimental setup used in the paper can be found elsewhere (9).”

6.The chart layout is messy and unclear.

We believe that referencing the figure for the experimental setup addresses this concern.

7.The conclusion is somewhat innovative but lacks logical coherence

Revised some of the conclusions based upon reviewer 2 feedback (see below).

Reviewer 2 Report

Comments and Suggestions for Authors

The paper deals with the problem of analysis of experimental image data. The topic of the paper is relevant and the methods of analysis used are appropriate. The paper itself has shortcomings that need to be removed before publication in a scientific journal. To help the authors resolve these problems, I am providing a list of questions and comments on the basis of which the authors should make extensive revisions to the manuscript. 

 

Abstract

- The first two sentences provide an introduction to the subject. But the following part is in no way related to the first two sentences and does not logically follow them. Overall, the abstract is unorganized and does not give the reader concise answers to the main questions: what is the focus of the paper, what methods were used, what results were achieved and what novelty does the paper bring.

 

Conclusion

- The first sentence refers to a wide range of experiments and atomizers used, none of which is mentioned in the article. 

- This section should summarize what the authors have achieved, what is new in their findings. 

- I would recommend the authors to focus also on the practical use of their results. 

- Literature citations are not used in this section. The last sentence would belong more in the Introduction section. 

 

Further comments: 

- The article does not specify which atomizer was used, nor its specific dimensions or working parameters (e.g. flow rates, pressures). Details must be provided.

- What fluid was used in the experiment? The Introduction mentions emulsions, but nothing further is given in the description of the experiment. Details must be provided.

- There is only minimal information in the article about the experiment or about the way the experimental data was obtained. If an experiment has been performed. It must be described in sufficient detail to allow it to be replicated. 

- The authors state that they evaluate the droplet size from the image (Table 1). However, Fig. 1 shows that the breakup was not completed in the visualized part of the spray and thus the evaluated sizes may not correspond to the final droplet size further away from the nozzle. It would be advisable to include a more reliable droplet size measurement in the fully formed spray region. 

- The presented methodology is based on the measurement of geometrical parameters (e.g. angles) on the liquid film. However, the methodology does not deal with the fact that the recorded image is a projection of the conical surface onto the plane of the camera sensor. The image of all shapes that are not orthogonally oriented to the camera, i.e. objects further from the axis of the cone, are distorted. It is the angles between the holes, which according to the authors have a significant effect on the variance of the data, that are perhaps most affected by this distortion. How have the authors dealt with this fact? It is necessary to provide a detailed explanation in the text. 

Author Response

Reviewer 2.

 

Comments and Suggestions for Authors

The paper deals with the problem of analysis of experimental image data. The topic of the paper is relevant and the methods of analysis used are appropriate. The paper itself has shortcomings that need to be removed before publication in a scientific journal. To help the authors resolve these problems, I am providing a list of questions and comments on the basis of which the authors should make extensive revisions to the manuscript. 

 

Abstract

The first two sentences provide an introduction to the subject. But the following part is in no way related to the first two sentences and does not logically follow them. Overall, the abstract is unorganized and does not give the reader concise answers to the main questions: what is the focus of the paper, what methods were used, what results were achieved and what novelty does the paper bring.

We agree that the abstract was confusing when rereading and was based upon truncating down the original abstract of an internal report to meet the constraints for the journal.  The following in red were added to address these concerns.  The added words do slightly exceed the journal guidance but make a much better abstract for the readers.

Atomization of liquid spray solutions through nozzles is the mechanism for delivering many pesticides to the target. The smallest drop sizes (< 150 mm) are known as driftable fines and have the propensity for wind induced convection. Many agricultural applications include oil-in-water formulations, and experimental metrics obtained from spray images of these formulations include the distance from the nozzle origin to a drop centroid once a drop has formed, the hole location and surface area for holes that form in the liquid sheet (all hole areas approximated as polygons), angles formed between polygon segments (whose vertices are represented as boundary points), and ligament dimensions that form from intersecting holes such as the ligament aspect ratio (L/R), ligament length (L), ligament radius (width), along with the number of drops a ligament breaks up into.  These metrices are used in Principal Components Regression (PCR) analysis and illustrate that 99% of the variability in the response variable (DT₁₀) is addressed by 10 Principal Components.  Angles formed by colliding holes, hole distance from the nozzle, drop distance, hole number, ligament number and drop number are negatively correlated to atomization driftable fines fraction, while the hole-area, ligament distance, ligament area, and boundary area are positively correlated.  Thus, to decrease/minimize driftable fines, one needs to increase negatively correlated metrics.

 

Conclusion

- The first sentence refers to a wide range of experiments and atomizers used, none of which is mentioned in the article. 

Added to the end of the first paragraph in the methods section.

Videos of 21 different oil-in-water formulations being sprayed through a conventional flat fan nozzle (XR8002, TeeJet) were taken at two different pressures (40 psi, 10 psi) and different oil-phase concentrations (2.5 and 5.0%). Thus, the two-phase fluid being sprayed was mostly water with a small fraction (2.5-5%) oil as representative of typical agricultural sprays. A HELOS Sympatec laser diffraction system (Sympatec GmbH, Clausthal-Zellerfeld Germany) was used to measure the resulting droplet size distribution upon atomization. The laser was aligned at the center of the spray located below the continuous sheet at 330 mm below the nozzle (thus the sheet is fully broken up into droplets. The nozzle used was an XR8003 (TeeJet Technologies) which is a conventional nozzle with a spray angle of 80°.  Further details of the experimental system can be found elsewhere (9).”

 

- This section should summarize what the authors have achieved, what is new in their findings. 

Discussion section has been slightly adjusted to emphasis the new findings.  The discussion section now reads:

 

Photographs of atomization spray patterns for two-phase systems show rich and diverse patterns that are dependent upon nozzle geometries/types, and nozzle operating conditions, but also the composition of the formulation and formulation attributes such as interfacial tensions between various phases, wettability, etc.  We find, metrics generated from images of spray patterns can be used as dependent parameters in data mining and statistical techniques and in offering insight/quantification for mechanistic modeling of sheet atomization.  PCR analysis showed that 99% of the variability in the response variable for atomization droplet size (DT₁₀) is addressed by 10 Principal Components.  Sensitive parameters making up these Principal Components include the angles formed by colliding holes within the atomization sheet, hole distance from the nozzle, drop distance, hole number, ligament number and drop number are negatively correlated to atomization driftable fines fraction (drops smaller than 150 mm), while the hole area, ligament distance, ligament area, and boundary area are positively correlated.  Image analysis and machine learning techniques suggest the most sensitive metrics include log(hole number²) and log(hole_area) followed by hole_dis and log(ligament number²).  Some variables which have significant impact on the variance of the data are the angles that form when holes collide and the log of hole area [log(hole_area)].  Based on coefficient and significant analysis, and analysis of variance, the variables “angle”, “log(hole_area)”, “drop_dis², hole_dis”, “log(ligament number²)” are the five most important predictors that can be used to quantify the breakup of oil-in-water formulations into individual droplets.  PCA/PCR and linear regression were utilized to substantiate the findings in this paper. 

 

- I would recommend the authors to focus also on the practical use of their results. 

Agree and added the following comment in the Results section.  “Thus, metrics generated from images of spray patterns can be used as dependent parameters in data mining and statistical techniques and in offering insight/quantification for mechanistic modeling of sheet atomization.”

 

- Literature citations are not used in this section. The last sentence would belong more in the Introduction section.

Agree, and we placed a sentence (which was the last sentence in the Discussion) which is now the last sentence in the Results section, e.g., “Details of deep learning and other software approaches are discussed in other publication articles [9-10].

 

Further comments: 

- The article does not specify which atomizer was used, nor its specific dimensions or working parameters (e.g. flow rates, pressures). Details must be provided.

(added the following lines about this in the methods section)

“Videos of 21 different oil-in-water formulations being sprayed through a conventional flat fan nozzle (XR8002, TeeJet) were taken at two different pressures (40 psi, 10 psi) and different oil-phase concentrations (2.5 and 5.0%). Thus, the two-phase fluid being sprayed was mostly water with a small fraction (2.5-5%) oil as representative of typical agricultural sprays. A HELOS Sympatec laser diffraction system (Sympatec GmbH, Clausthal-Zellerfeld Germany) was used to measure the resulting droplet size distribution upon atomization. The laser was aligned at the center of the spray located below the continuous sheet at 330 mm below the nozzle (thus the sheet is fully broken up into droplets. The nozzle used was an XR8003 (TeeJet Technologies) which is a conventional nozzle with a spray angle of 80°.  Further details of the experimental system can be found elsewhere (9).”

- What fluid was used in the experiment? The Introduction mentions emulsions, but nothing further is given in the description of the experiment. Details must be provided.

(see new additions that are added to the methods section above)

- There is only minimal information in the article about the experiment or about the way the experimental data was obtained. If an experiment has been performed. It must be described in sufficient detail to allow it to be replicated. 

Agree, and added more comments in the methods section and referenced the location (9) where further details on the experimental apparatus (and thus experimental details of how the videos were obtained) can be found.

- The authors state that they evaluate the droplet size from the image (Table 1). However, Fig. 1 shows that the breakup was not completed in the visualized part of the spray and thus the evaluated sizes may not correspond to the final droplet size further away from the nozzle. It would be advisable to include a more reliable droplet size measurement in the fully formed spray region. 

Added to experimental section about some details on the experimental apparatus with reference where further details can be found (if warranted). Drop size was measured after sheet breakup at a standardized distance of 330 mm below the nozzle (now stated in the methods section).

- The presented methodology is based on the measurement of geometrical parameters (e.g. angles) on the liquid film. However, the methodology does not deal with the fact that the recorded image is a projection of the conical surface onto the plane of the camera sensor. The image of all shapes that are not orthogonally oriented to the camera, i.e. objects further from the axis of the cone, are distorted. It is the angles between the holes, which according to the authors have a significant effect on the variance of the data, that are perhaps most affected by this distortion. How have the authors dealt with this fact? It is necessary to provide a detailed explanation in the text. 

All images used were from flat fan nozzles and not conical. Also, if a conical spray nozzle was used, those projected distortions would be visibly noticeable in the images; the ‘holes’ would appear smaller the farther away from center you look as they would be subject to the most distortion and occlusion. That is not the case for flat-fan nozzles.  In addition, even if these were conical sprays and the variance in hole size was due to the nonlinear projection onto the 2D camera detector, it would indeed help explain the noted variance in hole size. But it would not preclude the fact that the computed features were sufficient to reach the stated objective which was to classify spray formulations based on computed image features. Our objective was a practical aid to formulation design, and thus this question is irrelevant.  We do add “flat-fan nozzles” throughout the paper to avoid other readers going down this path incorrectly.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

1）There is a lack of recent literatures.

2）How to derive the values of those variables from the images, such as ligament length, radius, and area? The calculation method should be clearly explained in the manuscript.

3）The flow chat should be presented, for the algorithm of image process.

Author Response

1）There is a lack of recent literatures.

Added three new references [7-9] to address this concern.

 

7.  Kluza, P.A., Kuna-Broniowska, I, and Parafiniuk, S.   Modeling and Prediction of the Uniformity of Spray Liquid Coverage from Flat Fan Spray Nozzles. Sustainability 2019, 11(23), 6716.

 

8.            Sijs,R.  Kooij, S. Holterman, H. J., van de Zande, J., Bonn, D.  Drop size measurement techniques for sprays: Comparison of image analysis, phase Doppler particle analysis, and laser diffraction.  AIP Advances. 11, 015315 (2021)

 

9. Xiao, L., Liu, M., Zhu, H., Cai, J., Lin, J. Spray Droplet Size Characteristics of Different Biological Pesticides with Different Hydraulic Nozzle. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(2):100-106.

 

2）How to derive the values of those variables from the images, such as ligament length, radius, and area? The calculation method should be clearly explained in the manuscript.

Added to Methods and Materials:

“The nozzle in the images has a known size (as measured by a micrometer) and is used as the reference for all measurements taken from the image.  The diameter of a ligament is assumed cylindrical with the diameter given by the measured ligament width.  Holes within the sheet are approximated as polygons. The detected features are average angle of the polygons representing holes (angle), average distance from holes’ centroids to nozzle (hole_dis), average area of holes (hole_area), average distance from ligaments’ centroids to nozzle (liga_dis), average area of ligaments (liga_area), aspect ratio (average radius over length of ligaments R/L), average distance from droplets’ centroids to nozzle (drop_dis), average area of droplets (drop_area), average radius of droplets (drop_radius), average distance from continuous water sheet’s centroid to nozzle (bdr_dis), average area of continuous water sheets (bdr_area) and average number of holes, ligaments and droplets (hole_n, liga_n and drop_n).”

 

3）The flow chat should be presented, for the algorithm of image process.

Added a flowchart which is now Figure 3. (Figure 3.  Flowchart representing image processing procedure.)

Author Response File: Author Response.pdf