Research on oilseed yield forecasting in Henan Province based on SR_GA_BP neural networks
Efficient prediction of oilseed yield is important for making precise management decisions during the growth period of oilseeds.In order to improve the efficiency and accuracy of oilseed yield prediction in Henan Province,we collected data on an-nual oilseed yield and oilseed planting area in Henan Province,selected nine indicators as influencing factors,and used stepwise regression analysis(SR)to screen out the influencing factors with significant and independent effects as the main influencing fac-tors of oilseed yield in Henan Province.Aiming at the shortcomings of BP neural network model,which is slow to converge and easy to fall into the local optimal solution,genetic algorithm(GA)is introduced to optimize its weights and thresholds,so as to better fit the complex nonlinear relationship between oilseed production and its influencing factors in Henan Province.The simulation re-sults show that compared with the single BP neural network model and the SR_BP neural network model,the SR_GA_BP neural network model has higher prediction accuracy,with a MAPE of only 1.58%,so the SR_GA_BP prediction model can predict the oil-seed production in Henan Province more accurately.
NETL
NSTL
万方数据
