Research Of Soil Moisture Content Forecasting Model Of Genetic Neural Network

This paper embarks from soil moisture model's research based on the present research of the domestic and foreign scholars, analyzes each model's realization,forecasting mechanism,advantages,disadvantages and so on,and summarizes the representative research achievements. For some problems of soil moisture forecast model, this paper proposes a new method of soil moisture prediction model implementation, combining neural networks with genetic algorithms: genetic neural network forecast model. And this model has been applied to the test area.The main contents of the research are as follows:According to the different periods of crop root growth and water demands, the crops in different periods should be applied to different soil moisture forecast model. It can objectively reflect the soil moisture situation.According to the research result on root crops built by time and soil level, and soil moisture forecast modelâ… , modelâ…¡and modelâ…¢, which can meet the needs of actual production, these three models are applied to soil moisture forecast.According to the imperfect effects which caused by neural network's shortcomings, genetic neural network forecast model is established, which is used to forecast the soil moisture.The actual application result indicates that genetic neural network soil moisture forecast model is with high precision, good prediction effect, and can accurately reflect the changes of soil moisture in experiment when comparing with soil moisture forecast model of neural network.