Development And Application Of Intelligent Irrigation System In Large-scale Irrigation District

Large-scale irrigation districts are characterized by diversified information,large spatial variability,and complicated water management scheduling,which lead to the difficulty in the development of intelligent irrigation systems.The development of intelligent irrigation systems in large-scale irrigation districts has important practical significance for achieving efficient water management and water dispatching decisions in irrigation districts,improving the level of irrigation management intelligence,irrigation water use efficiency and benefits.This paper took the Qifu Plain Irrigation District as a typical irrigation area,used real-time information monitoring technology to realize real-time monitoring of irrigation area information,developed real-time irrigation forecasting model and canal dynamic optimization water distribution model to realize irrigation water management.At the same time,this paper used B/S architecture,MVC design mode,Web software development technology such as Think PHP framework to develop and implements intelligent irrigation system software.The main research contents and conclusions were as follows:（1）The overall architecture,hardware architecture and software business of the large-scale irrigation district intelligent irrigation system were designed in view of the characteristics of large-scale irrigation districts,the functional requirements,performance requirements and environmental requirements of the intelligent irrigation system.Then,how to carry out large-scale irrigation district intelligent irrigation system development work were described.（2）The specific location of the typical field and monitoring points was determined according to the monitoring method of monitoring points and the field investigation of the irrigation area.Real-time information monitoring equipment and gate remote control equipment were developed based on the real-time information monitoring technology and control principle.A total monitoring points of 26 field water layer,14 soil moisture and 19 channel water level/flow were set up in the study area including 10 water level/flow monitoring point data access points in the irrigation area,and the one gate remote control point to achieve comprehensive monitoring of real-time information in the irrigation district.（3）The real-time irrigation forecasting model of the irrigation area in the Qifu Plain Irrigation District was constructed,and the relevant model parameters were determined.The test results showed that the model was suitable for real-time irrigation forecasting of rice in the irrigation area of the Qifu Plain Irrigation District.（4）A multi-objective large-scale system decomposition coordination model for dynamic optimization of water distribution in large-scale irrigation districts was constructed.The flow of main canal was used as a coordination variable to coordinate the main-branch subsystem layer and the total system coordination layer model.The main-branch canal subsystem aimed to minimum number of main canal gate operations,minimum channel loss of water,and minimum abandoned water in the canal system,which used the weighting method to transforms multi-objective function into a single objective function.And then the flow of main canal and the optimal solution of the subsystem were obtained using genetic algorithm.The total system coordination layer made minimum number of general main canal gate operations as the objective function,and used the water distribution strategy to adjust the main channel flow and time.The results of this model showed that the number of operations of the main channel and the above canal gates was significantly reduced compared with the traditional scheme,the number of reductions was 3 to 18 times,the water loss of the canal system was reduced,and the loss reduction ratio was 3 to 17%.This model reduced irrigation management costs and canal water loss,improved the effective utilization of irrigation water.（5）Software system design and implementation has used B/S architecture,MVC design pattern and Think PHP development framework for large systems;The system function was divided into six parts:real-time information monitoring,information communication,real-time irrigation forecasting and dynamic water decision,gate monitoring control,file management,which based on overall architecture design and software business design;The system database was divided into real-time database,decision database,basic database,and managed by My SQL;The function module and user interface implementation was used PHP,HTML,CSS,Java Script;The simple intuitive user interface was used Bootstrap front-end framework,and to to implement custom data visualization charts using ECHARTS,DTGrid chart plug-in.（6）The practical application of the intelligent irrigation system in the west general main canal of the Qifu Plain Irrigation District in 2017 and 2018 showed that the ET₀ forecast and irrigation forecast were accurate,which had considered the future rainfall,leading to reduction of net irrigation in the field.The use of the canal dynamic optimization water distribution model reduced the leakage loss of the canal system,and realized the irrigation district management water saving.The total water saving was128.84 million m³ and the annual average water saving rate was 14%in 2017（late rice）and 2018（the early and late rice）.