Research On Prediction Model Of Soil Nutrient Content

The deep integration of new-generation information technologies such as artificial intelligence,big data,and the Internet of things with the agricultural field has laid a strong foundation for the development of precision agriculture and intelligent agriculture,and the results will certainly provide new ways to solve the problems of resource shortage and environmental pollution.Determining the nutrient content of the soil is the key to precision fertilization,because of the determination of soil nutrient content in related agricultural technologies that solve the problem of precise fertilization,sampling method?spectral analysis and other methods are often used,however,these methods will add additional planting costs,if the soil nutrient content,fertilization amount and yield of previous years can be better used to predict the soil nutrient content of the next year,it will greatly reduce the investment of human resources and material costs.Therefore,this paper takes the soil nutrients,fertilization amount and corn yield collected for many years in the test field of No.13 Village,Gong Peng Town,Yu Shu City,Jilin Province as the research objects,and uses the improved gray wolf algorithm to optimize the BP neural network prediction model for the soil nutrients of the next year,so as to provide a decision basis for precise fertilization.The main research work of this article is as follows:(1)Collect and organize the data of the key research project of Jilin Province's science and technology development plan "Research on precision and control techniques for high quality and high efficiency production of main grain crops" and rice demonstration area,obtain data on soil nutrients?fertilization amount and corn yield for five consecutive years,and research idea of soil nutrient prediction model of BP neural network optimized by gray wolf algorithm.(2)A model for predicting soil nutrient content based on the gray wolf algorithm optimized BP neural network was proposed,and the model was used to predict the soil nutrient content in the fifth year by using the soil nutrient,fertilization,and yield data of the previous four years.The research and experimental results show that the combined GWO-BP neural network is better than the original BP neural network model in the prediction of soil nutrient content.(2)The reverse learning mechanism is added to the gray wolf algorithm to enhance the algorithm's ability to search the optimal solution.The improved grey wolf algorithm wad used to optimized the BP neural network model(OBLGWOBP)to predict the soil nutrient content in the fifth year.The research and experimental results show that the improved gray wolf algorithm optimized BP neural network model has better prediction effect than the gray wolf optimized BP neural network model.