The Design And Realization Of The First Control Device For Smart Irrigation

The Internet of Things technology is an important technical support to realize modern agriculture,so the agricultural Internet of Things has been rapidly developed as a strategic industry in China.The use of scientific and technological means to achieve precise,intelligent and scientific management of agricultural production is an important measure to realize the transformation of traditional agriculture to modern agriculture.However,farmland environmental monitoring has the characteristics of wide monitoring area,long period,complex and changeable environmental factors,etc.The farmland environmental monitoring equipment currently put into use generally has problems such as poor stability,low real-time performance,and limited node monitoring parameters.Meet the various needs of accurate farmland environmental monitoring.This thesis designs an intelligent irrigation head control device combining LoRa(Long Rang)communication and NB(Narrow Band)communication,this device performs real-time monitoring of environmental data in various areas of farmland,remote and precise control of farmland irrigation.Moreover,a water demand prediction model based on GA-Elman(Genetic Algorithm Elman)neural network is constructed,and the irrigation water consumption in the farmland in the future period is predicted based on the monitored environmental data to achieve smart irrigation.This thesis first analyzes the domestic and international status of the Internet of Things technology in agricultural environment monitoring and control,and determines the overall scheme design and related technologies of the intelligent irrigation head control device according to the current needs of agricultural monitoring and control equipment.Secondly,the thesis designs and implements the first intelligent irrigation control device,and completes the hardware circuit design of the collection node,the aggregation node and the corresponding embedded software design.Then,using the environmental data collected by the head control device as sample data,a water demand prediction model based on the GA-Elman neural network was constructed,and a good prediction effect was achieved.Through experimental comparison and error analysis,they verified the GA-Elman neural network the prediction accuracy of the water demand prediction model is higher than the traditional BP(Back Propagation)prediction model and E1 man prediction model.Finally,the actual test and application of each functional module of the intelligent irrigation head control device were carried out.The test results show that the head control device has good performance in terms of ad hoc network,packet loss rate,communication distance,etc.And the cloud platform of the Internet of Things was displayed,all meeting the preset requirements.The experimental results show that this device can realize remote monitoring of farmland real-time data and remote intelligent irrigation,and finally this device is applied in practice.Experimental comparison between the water demand predicted by GA-Elman neural network and the water consumption for irrigation in the traditional way,the water-fertilizer integrated drip irrigation method based on GA-Elman neural network water demand prediction was obtained,and the water saving rate reached 82.6%,which validates the prediction model of this design can save water and increase production,and further determines the feasibility of this design content,which can be used in the field of smart agriculture.