Design And Implementation Of Sugar Beet Irrigation Control Algorithm Based On Fuzzy Neural Network

As a large agricultural country,China's water resources per capita are less than 25% of the world's per capita.At the same time,human production activities and agricultural development are inseparable from water resources.With the increasing scarcity of water resources,countries around the world pay more and more attention to the development of water-saving and irrigation.Therefore,how to realize agricultural accurate irrigation has become an important problem at this stage.Based on the research of precision irrigation and the development of artificial intelligence at home and abroad,the paper designs the control algorithm of sugar beet irrigation based on fuzzy neural network,and implements it.The main research contents are as follows:First,based on the review of relevant literature and analysis of the research status at home and abroad,the paper constructs the overall framework of it.It introduces the growth environment and water demand of beet,the theory of fuzzy control,the structure of BP neural network,the advantages and disadvantages of BP neural network,the comparison and combination of fuzzy neural network and neural network,and the knowledge of irrigation control.Second,The establishment of beet irrigation control model.The framework of the model analyzes the influence of air temperature and humidity,soil temperature and humidity in different depth,light intensity on the water demand intensity of Sugarbeet irrigation.And then it obtains the actual data fitting diagram and corresponding equation of different influence factors on the water demand of Sugarbeet through MATLAB simulation,and finally designs the overall function model by combining each factor.Thirdly,the fuzzy neural network and its back propagation process are analyzed,and the parameters in the fuzzy neural network are improved by momentum gradient descent and rmsprop optimization algorithm respectively.Then,in the nonlinear function,Mackey glass chaotic time series and the output prediction of water quality grade evaluation,the performance indexes such as error and fitting effect of the optimized fuzzy neural network are analyzed and compared with other improved results.After that,the improved fuzzy neural network is applied to the three periods of beet Growth.And the actual output and prediction output fitting graph of training,prediction output graph before and after optimization,and prediction error graph of different optimization algorithms are obtained.Fourthly,the control model of sugar beet irrigation is realized.The whole development environment is designed.The fuzzy neural network is realized by python.The database is built by mysql.The web page and function structure are designed,and the remote control solenoid valve of sugar beet irrigation is finally realized.