Development Of Jujube Intelligent Irrigation System Based On Stacking Ensemble Learning

The shortage of water resources in the southern Xinjiang region of China has had an impact on agricultural development,and farmers have insufficient understanding of water-saving irrigation technology.They believe that drip irrigation,sprinkler irrigation,and other irrigation methods have limited irrigation volume and cannot meet the growth needs of fruit trees.Most of them still use traditional flood irrigation methods.Overall,there is a serious waste of water resources in the southern Xinjiang region.This article utilizes sensor technology,wireless transmission technology,Internet of Things technology,machine learning technology,and other technologies to design and develop an intelligent irrigation equipment based on Stacking integrated learning prediction model.It achieves functions such as data collection and transmission,water demand prediction,and intelligent irrigation.The main research findings of the paper are as follows:(1)In order to increase the accuracy of the prediction model of crop water demand,the Stacking integrated learning prediction model was constructed with BP neural network and random forest as primary learners and ridge regression as secondary learners.The model chooses to add water demand factors from seven days ago and meteorological factors from one day ago based on six factors such as average temperature,atmospheric pressure,and solar radiation in the P-M calculation formula.The prediction results indicate that the fitting coefficient R2 between the predicted water demand of the Stacking ensemble learning prediction model and the actual crop water demand is 0.972.The Stacking ensemble learning water demand prediction model can better meet the actual demand.(2)The system adopts "LoRa+4G" communication technology to build a wireless sensor network for field environment,and establishes a MySQL cloud database to store relevant data.The cloud database contains sensor data information tables and irrigation information tables.And utilizing Alibaba Cloud Io T application development,corresponding web and mobile application platforms have been built to achieve real-time display of field environment and device information.(3)The system conducts communication and irrigation testing.The communication test includes data acquisition and transmission testing,remote control platform function testing,and the application platform can display data normally during the test,and the irrigation switch control function is normal.In the irrigation testing,all modules of the system can operate normally and well.