Review Reports

Comments 1: An inconsistency is observed in the formulation of Equation (4). Since the particle passage frequency in each diverging channel is given by f/N, in order to ensure that the sensor operates within its maximum detection capability, the condition f/N≤fmaxmust be satisfied. This leads to the requirement that the number of diverging channels should fulfill N≥f/fmax. However, in the manuscript Equation (4) is presented with the inequality sign reversed （N≤f/fmax）, which contradicts the physical design criterion. Nevertheless, the final selected value of ?(six channels) is consistent with the correct reasoning. It is therefore recommended that Equation (4) be corrected or that the rounding criterion used in the design of the flow divider be explicitly stated. 

Response 1:Thank you for pointing this out. We agree with this comment. Therefore, We revise equation (4) to （N≤f/fmax）.This change can be found – page number 6, paragraph 2, and line 235.

Comments 2: A review of the calculations associated with Equation (5) is recommended. Although the formulation of the equation is conceptually correct, assuming that the sum of the cross-sectional areas of the six diverging tubes is equal to the inlet cross-sectional area, the reported inner diameter of the diverging tube (de = 17.1 mm) is not consistent with the inlet diameter previously defined in the manuscript (D₁ = 32 mm). A direct substitution of these values into Equation (5) would yield de ≈ 13.1 mm. Therefore, it is suggested to verify the value of D₁ actually used in the calculation or to correct the reported result, in order to ensure internal consistency among the adopted geometric parameters and numerical outcomes.

Response 2: Thank you for pointing this out. We agree with this comment. We have found some errors in the calculations.Therefore,We have revised the inner diameter of the diversion pipe de ≈ 13.1 mm.This change can be found – page number 7, paragraph 3, and line 259.

Comments 3:Regarding Section 2.2.3 (Sensing device): the selection of the inclination angle α = 45° is based on preliminary experiments; however, the manuscript does not specify which angles were evaluated or the criteria used for this choice. It is recommended to clarify the procedure or provide an objective justification supporting the selected angle.

Response 3: Thank you for pointing this out. We agree with this comment. Therefore, We separately discussed the working conditions of α when it is too small or too large, and made revisions in the text.This change can be found – page number 8, paragraph 1, and line 284-291.

Comments 4:In Section 2.3, the ranges defined for the flow rate error (e∈ [−60, 60] g·s⁻¹) and its variation (ec∈ [−30, 30] g·s⁻¹) are not justified. Given that the maximum system flow rate is approximately 12.5 g·s⁻¹, these ranges appear to exceed the actual operating conditions. The authors should clarify the adopted criteria and review the dimensional consistency of ec.

Response 4:Thank you for pointing this out. We agree with this comment. Therefore,based on the characteristics of fertilization rate required by agronomy and actual operating conditions,we have revised the definition range of flow rate error (e∈ [−30, 30] g·s⁻¹)and its variation (ec∈ [−15, 15] g·s⁻¹).This modification will be more in line with the actual operating conditions.This change can be found – page number 12, paragraph 1, and line 418.

Comments 5:In Figure 13, the image quality is insufficient, making it difficult to clearly identify the labeled components. In particular, items 1 and 2 cannot be clearly observed. The authors are encouraged to improve the figure quality.

Response 5: Thank you for pointing this out. We agree with this comment. Therefore, We have replaced the images with clearer ones.This change can be found – page number 17, paragraph 1, and line 557.

Comments 6:In Section 3.2.1. Shunt distribution uniformity test, although the fertilizer used in the experiment is specified, the manuscript does not provide a basic physical characterization of the material (e.g., average particle size, bulk density, or moisture content). Since the performance of the dosing, diversion, and detection system directly depends on the physical properties of granular fertilizer, this information is relevant for proper interpretation of the results.

Response 6: Thank you for pointing this out. We agree with this comment. Therefore,We provide the basic physical characteristic parameters of granular fertilizer.This change can be found – page number 17, paragraph 1, and line 561-562.

Comments 1: Discuss extensively the current limitations of the system (e.g., sensitivity of PVDF sensors to humidity or extreme vibrations, long-term mechanical wear, performance at operating speeds >6 km/h). 

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, We discussed in detail the current limitations of the system.This change can be found – page number 24, paragraph 1, and line 725-745.

Comments 2: Enlarge Figure 1 and delete the point at line 113.

Response 2:Thank you for pointing this out. We agree with this comment. Therefore,We Enlarge Figure 1 and delete the point at line 113.This change can be found – page number 3, paragraph 2, and line 126.

Comments 3:Discuss the state of the art more extensively and compare with other similar studies in the literature.

Response 3: Thank you for pointing this out. We agree with this comment. Therefore,We further discussed the latest developments in this technology and compared it with other similar studies in the literature.This change can be found – page number 24, paragraph 2, and line 746-756.

Comments 4:Expand the references section with studies related to granular fertilizer dosage control.

Response 4:Thank you for pointing this out. We agree with this comment. Therefore, We have expanded the references related to dosage control of granular fertilizers.This change can be found – page number 2, paragraph 2, and line 79-82.

Comments 5:Table 3 (Field Experiment): Valuable data. It is clear that fuzzy PID consistently outperforms traditional PID. Calculating and reporting the standard deviation (or range) for accuracy at each speed (for both systems) would transform these data from demonstrative to statistically convincing.

Response 5: Thank you for pointing this out. We agree with this comment. Therefore,We calculated the standard deviation of accuracy for two systems at different operating speeds.This change can be found – page number 23, paragraph 1, and line 709-716.

Comments 1: The literature coverage is relatively adequate, but the logical organization and problem condensation are insufficient, lacking a problem-oriented convergent structure. For example, under high-throughput conditions, what are the core bottlenecks: signal saturation, particle collision and occlusion, statistical bias, or response delay? Many studies simply conclude that “adaptability is insufficient” or “accuracy needs improvement” without providing quantitative comparisons or mechanism-level analyses of these limitations, resulting in an insufficient justification of the necessity of the proposed method. It is recommended to restructure the review as: high-throughput operating conditions → failure mechanisms of existing detection methods → demand for parallelization and closed-loop control → proposed approach in this study.

Response 1: Thank you for pointing this out. We agree with this comment. Therefore, We will restructure the review into: analyzing existing detection methods and their failure conditions, and then proposing the detection method in this paper to address the problems of insufficient real-time detection accuracy, poor consistency of fertilizer discharge, and imperfect closed-loop control system in existing variable fertilization technologies.This change can be found – page number 2,3, paragraph 1,2, and line 49-56,105-109.

Comments 2: Although a fuzzy PID control system is constructed in the manuscript, the control law is not formally expressed; system stability, convergence, or robustness are not analyzed; and the fuzzy rules are mainly based on empirical descriptions, lacking a systematic design rationale. It is recommended to supplement stability analysis or parameter sensitivity analysis of the control system, as well as a mechanism-based explanation of why fuzzy PID outperforms traditional PID. 

Response 2: Thank you for pointing this out. We agree with this comment. Therefore,we have provided a detailed introduction to the core module composition of the fuzzy controller and the working principles of each module and added sensitivity analysis of fuzzy PID parameters.Compared to traditional PID systems, fuzzy PID systems require shorter response time and stronger stability to reach steady state,therefore, fuzzy PID is superior to traditional PID.This change can be found – page number 10,12,14,paragraph 2,5,1 and line 353-366,427-446,464-470.

Comments 3:Ablation experiments are insufficient (e.g., without diversion vs. with diversion, comparison among different numbers of channels); repeated experiments report only mean values, lacking standard deviations, confidence intervals, or significance tests; and the field experiment sample size is relatively small, with limited variation in operating conditions. Statistical significance indicators should be added, along with sensitivity analyses of key parameters (tilt angle, flow rate, and operating speed).

Response 3: Thank you for pointing this out. We agree with this comment. According to the agronomic requirements for soybean fertilization in the Huang Huai Hai region, the total fertilizer discharge frequency f is approximately 257 Hz based on a soybean fertilization rate of 225 kg/hm², a row spacing of 40 cm, 3 rows of fertilization, a width of 1.2 m for machine operation, and a machine operation speed of 2.0~5.0 km/h. Due to the high frequency, it was found in previous experiments that testing could not be carried out without diversion. At the same time, determine the number of diversion channels based on preliminary experiments, fertilizer application requirements, and reference literature. Regarding the sensitivity analysis of the key parameters you proposed (inclination angle, flow rate, and operating speed), relevant analysis has been conducted in the simulation experiment section mentioned earlier. Finally, the article mainly focuses on the research on the inability to detect fertilizer discharge in real time, achieving real-time monitoring of fertilizer discharge and solving practical production problems. There are indeed many shortcomings in terms of data volume and long-term job performance. Our team will carefully discuss the job performance and optimize it in a targeted manner to further improve performance. Thank you again for the expert's question.This change can be found – page number 16,17, paragraph 1 and line 538-550.

Comments 4:The results show that fuzzy PID outperforms traditional PID in response time and accuracy, which is an expected outcome; however, the physical or statistical reasons why consistency can still be maintained under high-throughput and inclined operating conditions are insufficiently explained. Although many visualizations are provided, in-depth interpretation remains limited.

Response 4:Thank you for pointing this out.We agree with this comment. Therefore,we introduced the reasons why fuzzy PID systems can still maintain consistency under high-throughput and tilt operating conditions.In the simulation experiment and analysis of the diversion device in the paper, simulation analysis methods were applied to conduct experiments on the diversion device under inclined states of 1 °, 3 °, and 5 °. It was shown that the diversion device can still maintain good diversion effect when the fertilizer flow rate is small and the inclination angle is large. When the tilt of the equipment is large, the ground surface is relatively uneven, and the operating speed will decrease. Therefore, the particle flow rate will decrease, and the detection effect can be guaranteed when the flow rate is low.

This change can be found – page number 14, paragraph 2, and line 472-478.

Comments 5:The discussion section does not systematically address the reliability of sensors under strong vibration, dust, and humid environments; the scalability of the system (more channels and higher throughput); or the adaptability to different fertilizer particle shapes and material properties.

Response 5: Thank you for pointing this out. We agree with this comment. Therefore, We discussed in detail the current limitations of the system.This change can be found – page number 24, paragraph 1, and line 719-740.

Response to Reviewer X Comments

1. Summary

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.

2. Questions for General Evaluation

Reviewer’s Evaluation

Response and Revisions

Is the work a significant contribution to the field?

[Please give your response if necessary. Or you can also give your corresponding response in the point-by-point response letter. The same as below]

Is the work well organized and comprehensively described?

Is the work scientifically sound and not misleading?

Are there appropriate and adequate references to related and previous work? 

Is the English used correct and readable?        

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: The paper needs to discuss the relationship between the variation range of fertilizer flow rate and the parameter selection of the designed device. It is better to consider the differences in flow rate calculations for fertilizers with different densities and particle sizes, and to take into account the generality of the device; How to ensure measurement accuracy when a large amount of fertilizer is needed; How to modify the device when more branches are needed, i.e. when the N_s result of equation (4) changes?

Response 1:Thank you for pointing this out. We agree with this comment.The fertilizer dosage in this study ranges from 5.0 to 12.5 g/s, representing a relatively high flow rate. Therefore, a diversion structure was designed. Based on calculations, the number of diversion pipes was determined to be 6.N_s ≤f/f_max.This change can be found – page number 6,7, paragraph 3, and line 241-251.

Comments 2: The paper should explain the relationship between the number of diversion devices and the number of operating rows. It is not clear whether there is one device for one row. In the experimental plan, L582: "including 3 rows of operations" indicates the setting of 3 rows, but from Figure 15, there is only 1 row.

Response 2: Thank you for pointing this out. We agree with this comment.The design specifies three rows of operations. For the sake of testing and analysis, one row was selected for experimentation.

Comments 3:The parameters A and N₁ of formula (7) need to be explained.

Response 3: Thank you for pointing this out. We agree with this comment. Therefore,we have explained the parameters A and N₁ in equation (7).This change can be found – page number 11, paragraph 3, and line 380-381.

Comments 4:In Figure 18, the target fertilizer discharge amount is not displayed.

Response 4:Thank you for pointing this out. We agree with this comment. The target fertilizer application rate for the field plot being 225 kg/hm².This change can be found – page number 20, paragraph 3, and line 650-653.

Comments 5:There are some language errors in the paper. For example, Uniform is written as' unfirming '(L549, L552, etc.). Is the vertical axis title of Figure 14 right? granular fertilizer quality？

Response 5: Thank you for pointing this out. We agree with this comment. Therefore,we corrected the incorrect words.The vertical axis label in Figure 14 is labeled as “Particle Fertilizer Quality.”This change can be found – page number 18,19, paragraph 2, and line 589-600.