Research On Grey Modeling Of Regional Agricultural Drought Losses

In recent years,with the global warming,the frequency and scope of drought disasters have increased significantly,and they have been affecting all areas of human life.Among them,the impact on agriculture is the most significant.Henan Province is a major grain producing area in the country.The annual agricultural losses caused by drought far exceed other meteorological disasters,which severely restricts the agricultural development of Henan Province.Therefore,research on drought disasters in Henan Province is of great significance.This paper takes the agricultural drought disaster system in Henan Province as the research object,analyzes the temporal and spatial evolution of drought disaster losses in Henan Province,explores the spatial distribution characteristics and critical period impact characteristics of regional agricultural drought losses,and identifies the main effects of regional agricultural drought losses Factors: Aiming at the problem of mixing characteristics in the main influencing factor data,a gray multivariate forecasting model based on direct modeling of mixing data is proposed to explore the response relationship between regional agricultural drought losses and meteorological and irrigation factors.The main work is as follows:(1)The drought-affected area and the disaster-affected area are selected to characterize the regional agricultural drought losses,and the basic statistical laws of the drought losses in Henan Province in the past 20 years are analyzed.Using MK test,EEMD and other methods to evaluate the temporal and spatial changes of the trend,sudden change and periodicity of drought disaster loss in Henan Province.Research has shown that the drought in Henan has shown a slight downward trend in the past 20 years.There is a 4-6year cycle,and it shows fluctuation characteristics during the cycle.As irrigation and other water conservancy facilities in various cities are gradually improved,the irrigation coefficient has gradually increased.The area of rain-fed agricultural areas is gradually decreasing,and the response of drought disasters to pure meteorological elements shows a weakening trend.(2)Using relevant data such as the actual situation of agricultural drought in Henan,the relationship between the comprehensive yield loss of Henan agricultural drought and the critical period meteorological factors was systematically analyzed.The analysis found that the comprehensive yield loss due to drought in Henan Province is mainly affected by changes in meteorological elements such as precipitation and temperature during the critical period.The reason is that the different growth stages of crops depend on meteorological elements differently and the meteorological elements have uneven seasonal distribution characteristics.The fitting relationship between the comprehensive yield loss of drought in Henan Province and the meteorological elements in the critical period is obviously better than that of the meteorological elements in the whole year.(3)Aiming at the coexistence of high-frequency meteorological elements and low-frequency socio-economic elements in the regional agricultural drought loss system,a gray mixing GM(1,N)model-MFGM(1,N)model is proposed.The model uses a weight function in the form of a Fourier series to aggregate high-frequency data information,uses the moisture sensitivity index and temperature sensitivity index of food crops as initial weights,and uses the simulated annealing algorithm to optimize the parameters to obtain the weight function value of the model,So as to realize the direct modeling of the mixing data.Take the loss of grain production due to drought in Henan Province as an example to predict.The results show that the MFGM(1,N)model proposed in the article has less error in the simulation and prediction of grain yield loss,and its accuracy is higher than that of the classic GM(1,N)model,which verifies the rationality and effectiveness of the model.