Intelligent Control Technology and System of on-Demand Irrigation Based on Multiobjective Optimization
Round 1
Reviewer 1 Report
The new scientific results are not very significant for the article. The research is the result of valuable work. The specialists worked hard to develop a new irrigation efficiency method. In my opinion, the new scientific value of the completed article is low.
The opening or closing of individual valves obviously affects each other, for which compensation methods are already in use in agriculture.
In my opinion, if the article is resubmitted after revision, it could be a valuable work, but it currently has no significant benefit for practical agriculture.
If the authors are able to integrate the results of the research and prove its practical usefulness by supplementing it with new measurements, then it can be a work suitable for a suitable scientific journal.
In my opinion, a valuable addition to the results of the article could be the use of tests on cultivated plants, evaluation of yield and yield parameters, possibly water use, water savings.
The methods used in the article are new and valuable, but they need to be supplemented.
Author Response
Response to Reviewer 1 Comments
Thank you very much for you appreciation.
Point 1: The new scientific results are not very significant for the article. The research is the result of valuable work. The specialists worked hard to develop a new irrigation efficiency method. In my opinion, the new scientific value of the completed article is low.
Response 1: The flow and pressure of irrigation pipeline are two important control variables in the irrigation systems, which determine whether the pump station meet the demand of on-demand irrigation system. In an on-demand irrigation system, the water demand of different hydrants is variable, and the total available water in the irrigation park is uncertain due to factors such as weather, climate, and park planning. The existence of dual uncertainties leads to significant differences in the requirements of different hydrants for irrigation water volume and water pressure. To meet the water needs of different users, improve irrigation water use efficiency, and ensure the safe and stable operation of the on-demand irrigation system, the system should be able to respond quickly and accurately to changes in pressure and flow demand. In order to solve the problem that the parameters of the MIMO irrigation system are difficult to control accurately, the intelligent control technology and system of on-demand irrigation based on multiobjective optimization has been studied.
Point 2: The opening or closing of individual valves obviously affects each other, for which compensation methods are already in use in agriculture.
In my opinion, if the article is resubmitted after revision, it could be a valuable work, but it currently has no significant benefit for practical agriculture.
Response 2: Although there have been many studies on the control technology of irrigation system, traditional irrigation head pumps are mostly constant speed pumps, which cannot meet the constantly changing flow and pressure requirements and can cause significant resource waste. In variable frequency control systems, due to the mutual influence of flow and pressure, it is difficult to adjust one of them separately. The response speed of the first system flow and pressure to changes in demand is another important factor to consider in irrigation systems, which greatly affects the efficiency of irrigation; On the other hand, when there is a sudden change in flow and pressure, the slow adjustment speed of the system can lead to a sudden increase in pipeline pressure, endangering the safety of system operation. The traditional accelerated PID conditional methods have to some extent improved the response time of the control system, but they mostly use PID parameter adjustment as a means, rarely starting from the relationship between the controlled quantity and its control quantity.
This study proposes a fast control system for head flow and pressure based on neural network combined with multiobjective optimization. The system uses a neural network to fit the relationship between flow, pressure, and their control variables, striving to quickly obtain the control values required for the target flow and pressure through the fitting model, directly controlling the corresponding executing components, and then fine tuning the flow and pressure to achieve rapid control of flow and pressure.
Point 3: If the authors are able to integrate the results of the research and prove its practical usefulness by supplementing it with new measurements, then it can be a work suitable for a suitable scientific journal.
Response 3: The irrigation system consists of the water source (tank), a pump station, the irrigation pipe network, and an emitter. The pump station refers to the general term for the equipment and devices used to distribute water from the water source to the pipeline network system, including pumps, filters, valves, pressure gauges, etc. The pipeline network system refers to the various levels of pipelines in the irrigation system, which are divided into tree shaped pipeline networks and circular pipeline networks based on their connection methods. The internal pipeline layout and optimization design of the pipeline network system have a significant impact on the operational performance of the irrigation system, and also affect the investment in the pipeline network.
On-demand irrigation refers to the irrigation system where each branch outlet of a pipeline system is treated as a single irrigation unit. Each irrigation unit can take water from the pipeline network at any time. The irrigation time is flexible, reducing the pressure on the water supply requirements of the system's water source at the same time period. It has more practical significance and can meet the water demand of various water consuming units to the maximum extent, maximize engineering benefits, and reduce system investment. On-demand irrigation is suitable for irrigation areas with multiple irrigation units, different crop types with different water requirements, and difficulty in implementing a unified irrigation work system.
Considering the above reasons, for the practical usefulness of the technology and system, it is necessary to combine specific irrigation systems and irrigation systems. Irrigation systems and irrigation systems vary in different regions, the intelligent control technology and system in this study was only developed, and the relevant technologies and systems that need to be combined with specific irrigation systems for practicality.
Point 4: In my opinion, a valuable addition to the results of the article could be the use of tests on cultivated plants, evaluation of yield and yield parameters, possibly water use, water savings. The methods used in the article are new and valuable, but they need to be supplemented.
Response 4: The research purpose of this article is to improve the accuracy of pressure and flow control in on-demand irrigation systems, and accurate control of pressure and flow can ensure the safe operation of irrigation system pipelines. The crop yield, yield parameters, possibly water use, water savings is not the main research content of this study.
Reviewer 2 Report
Authors have done an excellent work by putting this research/manuscript together. It is highly technical but timely and relevant research.
English is fine with minor editorial errors. Please make sure to spell check especially for on-demand
Author Response
Point 1: Authors have done an excellent work by putting this research/manuscript together. It is highly technical but timely and relevant research.
Response 1: Thank you very much for you appreciation.
Point 2: English is fine with minor editorial errors. Please make sure to spell check especially for on-demand
Response 2: We have consulted many literatures, and the term 'on demand' is often used for irrigation in large irrigation systems.
Reviewer 3 Report
The paper ‘Intelligent control technology and system of on-demand irrigation based on multiobjective optimization’ provides an interesting and innovative study. I do not have any major revisions for the paper, so my recommendation was for 'Accept after minor revision'.
Additional comments
1) At the end of the Introduction section address the research objectives and not how the paper is organized. Review that section (lines 110-114).
2) Use units in the form m3 h-1 and not m3/h, for example (Figure 4, 5, 6, lines 329, 347 and others).
3) In the Discussion section there is no citation of other papers. It is important to compare and discuss with other studies already developed in this line of research.
4) In the conclusion section the text of lines 465 to 473 is not necessary.
Author Response
Thank you very much for you appreciation.
Additional comments
Point 1: 1) At the end of the Introduction section address the research objectives and not how the paper is organized. Review that section (lines 110-114).
Response 1: This paper aims to design and develop an intelligent control technology and system using a hardware platform connected to pressure and flow sensors to monitor and control irrigation pump station. And additionally, determine the opening degree of valves in the pipeline and the frequency of pump station using multiobjective optimization.
Highlights:
On-demand irrigation control technologies are needed to optimize crop production.
The opening degree of valves and frequency of pump station are used to control the pressure and flow.
Machine learning and multiobjective optimization methods are used to optimize the on-demand MIMO control system.
Point 2: Use units in the form m3 h-1 and not m3/h, for example (Figure 4, 5, 6, lines 329, 347 and others).
Response 2: The relevant revisions have been completed in the revised draft
Point 3: In the Discussion section there is no citation of other papers. It is important to compare and discuss with other studies already developed in this line of research.
Response 3: Although there have been many studies on the environmental parameters of solar greenhouses, the models of these studies are difficult to apply to the actual solar green-house. Because the internal space of solar greenhouse is small, the changes of its internal environmental parameters is gradual. Therefore, the cost of environmental parameter sensor inside the greenhouse is low, and the monitoring accuracy is higher.
Point 4: In the conclusion section the text of lines 465 to 473 is not necessary.
Response 4: On-demand irrigation is to treat each hydrant as an irrigation unit, each hydrant can obtain irrigation water from the irrigation pipe network at any time. This irrigation meth-od can meet the water needs of multiple users to the greatest extent. The pressure and flow requirements of the hydrants, as well as the irrigation pipeline network and layout are constantly changing. Therefore, the pressure and flow at the pump station should also be adjusted as needed. In recent year, many researchers have developed and implemented systems that control irrigation in real-time for quick management decisions, resulting in adequate yields while saving significant amounts of water. Increasingly, farmers world-wide are turning to automated irrigation systems to save them a significant amount of time by remotely turning on and off pumps and valves.
Photovoltaic pumping system (PVPSs) are more competitive for use in irrigation, tra-ditional PVPS configurations is the use of a variable-speed drive (VSD). The performance of this type of configuration was tested both at a pumping test facility and in the field at a variety of sites. VSDs can thus potentially further improve the economic competitiveness of a PVPS, the studies show that the VSDs can thus potentially further improve the eco-nomic competitiveness of a PVPS. This study is only applicable to regulating the operating frequency of pump stations, and further research is needed for the comprehensive regula-tion of the pumps station and valves. In order to improve the management of an on-demand irrigation networks, a system called GreenValve (GVS) was designed by the Politecnico di Milano, the GVS is able to re-cover energy for its operation from the flow, and three-steps general and replicable meth-odological approach for the definition of installation and operating conditions, at the same time, the study show that simple management rules can reduce and even avoid the occurrence of hydrant failure, creating the conditions for more effective use of the re-sources.
In actual on-demand irrigation system, the distribution of pipeline, irrigation fre-quency, and irrigation scheduling will directly affect the performance of the control model. The on-demand irrigation system is a multiple input and multiple output system (MIMO), the real-time energy, the resilience, the pressure head at all nodes, the perturbation of an on-demand irrigation network were studied and optimized using machine learning, deep learning, multi-objective design, and simulation method, respectively. It is undeniable that these studies have provided new technologies of irrigation system, and ways to control the pump station and valves. However, the control parameters of these studies are single, and most of them focus on specific irrigation system networks. These control methods may not necessarily be applicable to different on-demand irrigation networks. Therefore, further research is needed on the comprehensive control methods of pressure and flow in on-demand irrigation systems.
In this paper, the operating frequency of the pump station, the opening degree of the main pipeline valve, and the branch pipeline valve are the input parameters, and the pressure and flow of the main pipe and the branch pipe are the output parameters. there-fore, conventional rotational irrigation control methods (such as stable flow regulation and constant pressure water supply) are difficult to meet the needs of on-demand irrigation. Therefore, in order to improve the applicability of the model, further con-sideration needs to be given to the impact of the pipeline network, irrigation scheduling and management on the on-demand irrigation system in the future.
Due to the low irrigation pressure and flow caused by rotational irrigation, high irri-gation uniformity, and low power and energy consumption of the head pump, and rela-tively low cost and operating expenses of the irrigation control system. Due to the high pressure and flow demand of the user end, as well as the significant fluctuations, it is dif-ficult to control effectively the uniformity of irrigation pipeline. In addition, the power and energy consumption of the pump station are also high, which correspondingly increases the cost and operating expenses of the irrigation control system. Therefore, in subsequent research, objectives such as irrigation uniformity, system cost, and operating cost should also be comprehensively considered.
Reviewer 4 Report
First of all, I want to congratulate the authors for their efforts in this paper. They presented the comparison of different approaches for optimized irrigation. The study is interesting and fits within the journal's scope. There are some issues to be improved to enhance the quality of the paper. Following, I include a series of comments aimed at improving the paper:
In the introduction, there is a lack of contextualization of current efforts for minimizing water requirements beyond the application of irrigation strategies, such as using monitoring technology or species with lower water requirements. Please cite some examples, such as 10.1016/j.agwat.2022.107581 or 10.1002/agj2.21059.
Add the aim of the paper and the main novelty in an independent paragraph at the end of the introductions before the structure of the paper. Consider using bullet points to highlight the novelty/contributions of the paper.
Section 3 should be divided into two sections. In the first one, the authors have to provide the details about the experiment setup (first paragraphs and pictures). Then, the results have to be presented in the second section, named results.
The discussion must be extended and used to compare the obtained results with similar studies. I suggest splitting the discussion into different sections. In the first section, a comparison of results with similar works, including a comparative table, has to be provided. If no similar studies (about irrigation and ML can be found, use papers in which MIMO and ML are used for other purposes). Additional subsections might include the limitations of the study and their impact.
General comments:
Consider reducing the extension of the abstract to 250 words maximum.
In the abstract, the authors should avoid using acronyms; if an acronym must be used, it has to be defined. Check MIMO, BP, SVR, MAE…
In keywords, avoid using the terms already used in the title.
Use and uniform description of the acronyms. Check the differences between MIMO and SVR. I suggest using the structure followed in MIMO acronym in the entire paper.
Author Response
Thank you very much for you appreciation.
Point 1: First of all, I want to congratulate the authors for their efforts in this paper. They presented the comparison of different approaches for optimized irrigation. The study is interesting and fits within the journal's scope. There are some issues to be improved to enhance the quality of the paper. Following, I include a series of comments aimed at improving the paper:
In the introduction, there is a lack of contextualization of current efforts for minimizing water requirements beyond the application of irrigation strategies, such as using monitoring technology or species with lower water requirements. Please cite some examples, such as 10.1016/j.agwat.2022.107581 or 10.1002/agj2.21059.
Response 1: Better irrigation practices, smart irrigation management in agriculture is essential for increasing crop yield. A closed-loop irrigation system for sugarcane farms using the In-ternet of Thins was developed, the solution seeks to improve irrigation management by seamlessly integrating the WiSA automated irrigation system with the IrriWeb irrigation decision support tool. In order to extend the findings of previous studies investigating the issue of proper positioning of water content sensors, the representativeness of soil water content sensors’ readings and the existence of Time Stable Representative Positions are investigated using a specially developed mathematical model. Salima evaluated differences between six mixtures of C3-C4 turfgrass grown under two water regimes (limited and high irrigation), the regression and conceptual model us-ing remote sensing parameters revealed the most adequate criteria to detect turfgrass var-iability under each growing condition. In order to evaluate turfgrass performance, temporal and spatial soil moisture and salinity dynamics, four irrigation scheduling approaches were compared, the results pro-vide important information to guide adoption of data-driven approaches to irrigation scheduling.
Point 2: Add the aim of the paper and the main novelty in an independent paragraph at the end of the introductions before the structure of the paper. Consider using bullet points to highlight the novelty/contributions of the paper.
response 2: This paper aims to design and develop an intelligent control technology and system using a hardware platform connected to pressure and flow sensors to monitor and control irrigation pump station. And additionally, determine the opening degree of valves in the pipeline and the frequency of pump station using multiobjective optimization.
Highlights:
On-demand irrigation control technologies are needed to optimize crop production.
The opening degree of valves and frequency of pump station are used to control the pressure and flow.
Machine learning and multiobjective optimization methods are used to optimize the on-demand MIMO control system.
Point 3: Section 3 should be divided into two sections. In the first one, the authors have to provide the details about the experiment setup (first paragraphs and pictures). Then, the results have to be presented in the second section, named results.
Response 3: The relevant revisions have been completed in the revised draft.
Point 4: The discussion must be extended and used to compare the obtained results with similar studies. I suggest splitting the discussion into different sections. In the first section, a comparison of results with similar works, including a comparative table, has to be provided. If no similar studies (about irrigation and ML can be found, use papers in which MIMO and ML are used for other purposes). Additional subsections might include the limitations of the study and their impact.
Response 4: On-demand irrigation is to treat each hydrant as an irrigation unit, each hydrant can obtain irrigation water from the irrigation pipe network at any time. This irrigation meth-od can meet the water needs of multiple users to the greatest extent. The pressure and flow requirements of the hydrants, as well as the irrigation pipeline network and layout are constantly changing. Therefore, the pressure and flow at the pump station should also be adjusted as needed. In recent year, many researchers have developed and implemented systems that control irrigation in real-time for quick management decisions, resulting in adequate yields while saving significant amounts of water. Increasingly, farmers world-wide are turning to automated irrigation systems to save them a significant amount of time by remotely turning on and off pumps and valves.
Photovoltaic pumping system (PVPSs) are more competitive for use in irrigation, tra-ditional PVPS configurations is the use of a variable-speed drive (VSD). The performance of this type of configuration was tested both at a pumping test facility and in the field at a variety of sites. VSDs can thus potentially further improve the economic competitiveness of a PVPS, the studies show that the VSDs can thus potentially further improve the eco-nomic competitiveness of a PVPS. This study is only applicable to regulating the operating frequency of pump stations, and further research is needed for the comprehensive regula-tion of the pumps station and valves. In order to improve the management of an on-demand irrigation networks, a system called GreenValve (GVS) was designed by the Politecnico di Milano, the GVS is able to re-cover energy for its operation from the flow, and three-steps general and replicable meth-odological approach for the definition of installation and operating conditions, at the same time, the study show that simple management rules can reduce and even avoid the occurrence of hydrant failure, creating the conditions for more effective use of the re-sources.
In actual on-demand irrigation system, the distribution of pipeline, irrigation fre-quency, and irrigation scheduling will directly affect the performance of the control model. The on-demand irrigation system is a multiple input and multiple output system (MIMO), the real-time energy, the resilience, the pressure head at all nodes, the perturbation of an on-demand irrigation network were studied and optimized using machine learning, deep learning, multi-objective design, and simulation method, respectively. It is undeniable that these studies have provided new technologies of irrigation system, and ways to control the pump station and valves. However, the control parameters of these studies are single, and most of them focus on specific irrigation system networks. These control methods may not necessarily be applicable to different on-demand irrigation networks. Therefore, further research is needed on the comprehensive control methods of pressure and flow in on-demand irrigation systems.
In this paper, the operating frequency of the pump station, the opening degree of the main pipeline valve, and the branch pipeline valve are the input parameters, and the pressure and flow of the main pipe and the branch pipe are the output parameters. there-fore, conventional rotational irrigation control methods (such as stable flow regulation and constant pressure water supply) are difficult to meet the needs of on-demand irrigation. Therefore, in order to improve the applicability of the model, further con-sideration needs to be given to the impact of the pipeline network, irrigation scheduling and management on the on-demand irrigation system in the future.
Due to the low irrigation pressure and flow caused by rotational irrigation, high irri-gation uniformity, and low power and energy consumption of the head pump, and rela-tively low cost and operating expenses of the irrigation control system. Due to the high pressure and flow demand of the user end, as well as the significant fluctuations, it is dif-ficult to control effectively the uniformity of irrigation pipeline. In addition, the power and energy consumption of the pump station are also high, which correspondingly increases the cost and operating expenses of the irrigation control system. Therefore, in subsequent research, objectives such as irrigation uniformity, system cost, and operating cost should also be comprehensively considered.
General comments:
Consider reducing the extension of the abstract to 250 words maximum.
In the abstract, the authors should avoid using acronyms; if an acronym must be used, it has to be defined. Check MIMO, BP, SVR, MAE…
In keywords, avoid using the terms already used in the title.
Use and uniform description of the acronyms. Check the differences between MIMO and SVR. I suggest using the structure followed in MIMO acronym in the entire paper.
The relevant revisions have been completed in the revised draft.
Round 2
Reviewer 4 Report
The authors have addressed the comments correctly.
