Research On Forecasting System Of Forage Grass Turning Green Period Based On Random Forest Regression Algorithm

Content: The research object of this paper is Stipa,a representative pasture in Yanchi County,Wuzhong City.It uses various meteorological factors where it grows to make regression predictions on the return to green period.Aiming at the limitation that the linear regression defaults to the linear relationship between the greening period and the meteorological factors,this paper uses the non-linear regression algorithm random forest to fit the forage greening period,and constructs the forage greening period forecast model.On the basis of the forecast model,it integrates functions such as weather condition monitoring and pasture growth monitoring,and develops the salt pond pasture meteorological service system to provide technical support for the determination of the start and end periods of the local livestock's full green and seasonal rest grazing.The research content of this paper is as follows:(1)Analyze the correlation between the forage greening period and various meteorological factors,and explore the correlation between the forage greening period and various meteorological factors in Yanchi County.The analysis results show that the forage greening period has the highest correlation with temperature conditions,and sunshine conditions Secondly,the precipitation conditions are the lowest.(2)Taking the selected 11 meteorological feature factors as input values,and the greening period day sequence as the target value,the forage greening period forecast model based on the random forest regression algorithm is constructed,and the final model is compared and evaluated with the traditional multiple linear regression algorithm.It is shown that the coefficient of determination(R2)of the prediction model constructed based on the random forest regression algorithm is 0.780,and the root mean square error(RMSE)is 4.91,which are better than the traditional multiple linear regression algorithm,which proves that the random forest regression algorithm is used in the prediction of forage returning green period.Has good applicability.(3)Develop and form a meteorological service system for forage grass in Yanchi.The system includes a forecasting module for grass turning green,a weather condition monitoring module,a growth monitoring module,and a system management module to realize real-time monitoring and analysis of meteorological conditions in Yanchi,grass resource growth monitoring,and grass turning green forecast,directly Provide guidance and services for the development of grass production.