Research On Remote Monitoring And Intelligent Irrigation System For Greenhouse Crop

In this paper,based on the traditional greenhouse planting,in order to realize remote monitoring of environmental factors affecting crop growth,an intelligent irrigation system for greenhouse crop remote monitoring was developed.At the same time,intelligent algorithm is used to control the terminal irrigation equipment to achieve the effect of intelligent water-saving irrigation.After analyzing the functional requirements of the system,the paper overall design scheme is put forward.The development and construction of three modules of data acquisition,data transmission and remote monitoring are completed,and the hardware design and software development of the system are described,including the main controller,NB-iot module,sensor,CC1110 wireless transmission module,etc.In order to realize the precision irrigation of crops,a crop evapotranspiration prediction model based on XGBoost optimization algorithm was constructed.Principal component analysis(PCA)was used to screen the key influencing factors of greenhouse crop evapotranspiration as system input variables.The bayesian algorithm is used to determine the hyperparameters of XGBoost model and enhance the prediction fitting degree of prediction model.Experimental results show that compared with the traditional grid search,random search and other hyperparameter optimization methods,the XGBoost prediction model after Bayesian optimization improves the prediction accuracy.Compared with other models on the same experimental data and experimental platform,the results show that the proposed prediction model has high prediction precision and can improve the accuracy and feasibility of greenhouse crop evapotranspiration prediction.At the same time,according to the actual situation of greenhouse crop growth,the paper also built an intelligent irrigation model based on fuzzy control.The predicted deviation between crop evapotranspiration and soil moisture was taken as model input,and the irrigation duration was output through a series of fuzzy processes.At the same time,considering the subjectivity existing in the establishment of fuzzy control system,the simulated annealing particle swarm optimization algorithm is introduced,the standard test function is selected to verify the superior performance of the algorithm,and it is applied to the intelligent irrigation fuzzy control system to complete the optimization of membership function.Finally,the paper uses JAVA language,MySQL database and Intelli J IDEA tool to develop a greenhouse crop remote monitoring and intelligent irrigation system based on B/S architecture.The system has user management,data storage,real-time environmental monitoring,historical data query,equipment control and other functions.The prediction of crop evapotranspiration and intelligent irrigation model were applied to cucumber planting site in greenhouse.The intelligent irrigation method was compared with traditional irrigation method in terms of average fruit yield,leaf number and irrigation amount.The results showed that intelligent irrigation could reduce irrigation amount without affecting crop growth.