Design Of The Decision System For Corn Precision Seeding Based On Cloud Platform

Precision agriculture is an agricultural production management model,which is the main trend of agricultural development in the world now and in the future.It can obtain information about the actual spatial and temporal differences in crop yields and environmental factors affecting crop production in farmland communities,analyze the reasons that affect yield differences,adopt technically feasible and economically effective control measures,change the traditional method of wasteful use of resources in large areas and large samples,implement positioning for crop cultivation management,and input variables on demand.Precision seeding decision-making technology is an important part of precision agriculture.According to the established production and planting model,it can realize the optimal utilization of farmland-related resources such as soil fertility,planting space,and air light,so as to reach the most suitable seeding amount of current farmland.As for precision seeding technology,most of the domestic researches are still precision seeders.Compared with foreign countries that have achieved the commercial scale of seeding decision technology,domestic research and development of overall seeding decision technology is less.Aiming at the lack of accurate seeding decision-making technology in China,taking corn as the research object,this paper designs an online decision-making system for precision seeding of corn.The methods for obtaining decision data,the optimization of precision corn seeding models,the construction of an online decision system,and the format of prescription charts were studied.The main research contents and results of this paper can be summarized as follows:(1)The corn sowing planting model was optimized.The previous method of fuzzy value of soil water and fertilizer content in previous corn planting and planting models was replaced with the using soil nutrient content classification index to quantize and subdivide the content of nitrogen,phosphorus,potassium,organic matter and other contents of soil water and fertilizer according to the first-class proportion of classification index,so as to make corn precise planting decision more accurate.The kriging interpolation algorithm in the spatial interpolation algorithm was used to interpolate and obtain important data required for decision-making such as soil water and fertilizer content,historical yield in farmland plots,and provide relevant decision data for each grid for a precision seeding decision model.It ensures the realization of variable seeding in the precision seeding decision of corn,and can be combined with the corn seeding decision model of a single plot to build a corn precision seeding decision model.The data acquisition technology required for the seeding decision was optimized and improved.Using the Scrapy crawler framework,a corn variety information web crawler was built,which achieved fast,accurate,and directional access to agricultural network information,providing a new channel and method for agricultural data acquisition.By setting up a portable small weather station in the field combined with a 4G module,long-distance wireless transmission of farmland environmental information data was achieved.A temperature prediction model based on empirical modal decomposition-long and short-term memory network(EMD-LSTM)was built,which realized the prediction of weather and temperature after sowing corn,and improved the accuracy of the seeding decision model.(2)An accurate cloud seeding online decision cloud platform based on Alibaba Cloud Server was constructed.By combining traditional field sampling technology,GPS coordinate acquisition technology,Scrapy web crawler technology,EMD-LSTM temperature prediction model,precision corn seeding decision model,and MySQL database technology,an online decision-making system for corn precision sowing was constructed,which realizes the monitoring and control of corn sowing decision-making anytime and anywhere.In order to match with the domestic mainstream precision corn seeders,the standard Shapefile format was selected as the file format of the standard decision prescription chart.In order to facilitate farmers to have an intuitive understanding of the whole farmland plot,the Api interface of Baidu chart was selected and used as a container to display the seeding prescription chart.Constructed a MySQL-based two terminal system database to ensure the real-time consistency of the dual-end data.The database information mainly includes:user information,farm/field information,farm/field boundary point information,water and fertilizer data information,corn variety information,historical yield information,historical prescription information,and weather information.(3)An online decision-making system for precision corn seeding on the mobile Android terminal was constructed.In order to facilitate the unification of two terminal data and the rapid generation of prescription chart,the mobile terminal system interacts with the seeding decision model and database of cloud platform through Java HTTP communication mechanism.In order to increase the convenience of mobile terminal operation and ensure the diversity of data collection,the mobile terminal has set up a function for farmers to upload data by themselves,which is convenient for farmers to disseminate data such as seed quantity information,fertilization information,and pictures of crop growth.(4)Experiments were conducted in Xuchang farmland using this corn seeding decision system.The experimental results show that the relative error between the predicted yield and the target yield of all grids is less than 4%,and the maximum absolute error is 342.2kg/ha.On the whole,the predicted output of all grids is basically consistent with the output trend of the target output,and the overall fitting effect is good.The accuracy of corn seeding decision-making can be verified indirectly from the side by combining the predicted yield with the good vegetation coverage of multispectral monitoring.