Research And Design Of Smart Irrigation System For Orchards Based On Internet Of Things

The orchard planting area is broad,the coverage scope is big while the orchard irrigation water supply and demand contradiction is day by day outstanding in our country.It has faced a huge test for the traditional extensive irrigation model.With the continuous development of agricultural modernization in our country,solving the problems facing orchard irrigation has become an important research topic in the field of current agricultural production.Combining sensor technology,wireless communication technology,cloud platform technology and Web technology,a set of orchard intelligent irrigation system based on the Internet of Things was researched and designed.Based on the three-layer architecture of the Internet of Things,the whole system was built.Meanwhile,in order to better guide orchard irrigation water,LSTM time series deep learning model is also introduced to forecast orchard irrigation water demand.The main research contents of this paper are as follows:(1)In order to meet the requirements of collection and transmission of field environmental data of orchard irrigation,this paper constructs a wireless communication transmission scheme based on ZigBee+NB-Io T mode.The ZigBee acquisition node is used to collect data of field environmental parameters of orchard irrigation,and the wireless communication technology of NB-Io T is used to realize remote transmission of data.Finally,the data can be sent to the cloud,and the control instructions issued by the cloud can be transmitted to the irrigation execution equipment.In addition,the software and hardware design of ZigBee acquisition node,irrigation execution equipment node and gateway node involved are respectively carried out.(2)The orchard irrigation water requirement prediction study was carried out.By analyzing the characteristics of irrigation water requirement of fruit trees,the revised Penman formula(PAO-56)was used as the real water requirement calculation model to construct the water requirement prediction model based on LSTM.At the same time,in order to improve the accuracy of water requirement prediction,the PSO optimization algorithm was introduced.Two super parameters,the initial learning rate of LSTM and the number of hidden layer neurons,were optimized,and the comparison model prediction experiment was carried out by MATLAB.The experimental results show that the optimized water demand prediction model based on PSO-LSTM has higher accuracy,and the predicted value fits the real water demand better.It can provide theoretical basis for guiding orchard irrigation water better.(3)Taking One NET cloud platform as the transfer hub of irrigation environmental data and control instructions,the BC26 terminal equipment is connected to the cloud platform using MQTT protocol,and the data points are uploaded.Meanwhile,the orchard irrigation trigger warning function is developed based on the uploaded data stream.Finally,the HTTP data push service provided by the cloud platform is used.A visualized third-party Web end orchard intelligent irrigation management platform was developed to realize real-time monitoring of orchard irrigation site environmental parameter data and remote control of irrigation equipment status.(4)System model building,debugging and analysis.Based on the orchard intelligent irrigation system model built in the laboratory,the designed system was tested and analyzed respectively in terms of function and performance,mainly including ZigBee acquisition node data acquisition function test,communication packet loss rate and overall system performance test,gateway node data sending and receiving function test.Web end orchard irrigation management platform data monitoring,historical data query and other functional tests.