| Since the founding of the People’s Republic of China,my country has attached great importance to disaster prevention and mitigation.Through the construction of large-scale water conservancy projects,the ability to resist drought has been significantly improved.However,due to the intensification of human activities such as the construction of farmland water conservancy projects and the process of urbanization,my country’s drought resistance and disaster reduction work is facing a series of new challenges.Henan Province is a central grain-producing province and an important guarantee for national food security.Due to frequent droughts caused by human activities and climate change,agriculture has become one of the areas most affected by droughts.To ensure stable and increased crop production,we should actively change traditional concepts and emergency management methods,and actively and scientifically combat drought.How to construct a technical system of drought assessment and regulation,improve crop yield and efficiency of drought resistance and disaster reduction in Henan Province,and enhance the comprehensive response capability of drought disasters is a scientific problem that needs to be solved urgently.Therefore,this paper takes crop drought losses in Henan Province as the analysis object,combined with the grey uncertainty analysis model,and conducts related research on the identifying of factors affecting crop drought losses,time-delay correlation analysis,and crop drought loss prediction in Henan Province.The main research contents and results are as follows.(1)In order to identify the influencing factors of crop drought loss in Henan Province,the existing models are mostly based on subjective factors,and the index system has overlapping and redundant problems.In this paper,a grey-rough set model is constructed to screen and optimize the indicators.Firstly,based on the principles of purpose,feasibility,comprehensiveness,and no redundancy,this chapter builds a hierarchical system consisting of categories,variables,and indicators,and initially selects 22 influencing factors.Secondly,the grey-rough set model is used to reduce the initially selected indicators.This paper establishes twelve indicators of cultivated land scale,land adaptability,soil consumption,labor force scale,labor base,labor quality,investment in irrigation facilities,mechanization input,agricultural input,water inflow,temperature,and water resources storage capacity in Henan Province.Correlation and prediction analysis of the influencing factors of crop drought losses were carried out,and the modified method was used to verify further the rationality and effectiveness of the grey-rough set model in the selection of indicators.(2)Aiming at the time-lag effect between the influencing factors of crop drought loss in Henan Province,the classical grey relational model has limitations in dealing with the time-lag relationship between indicators.In this section,a time-lag relational model is constructed.Based on the relevant content identified by the influencing factors,the twelve indicators selected from the four aspects of land,labor,capital,and nature are modeled and analyzed using the constructed time-delay correlation model.Firstly,based on the principle of new information priority,the optimal time delay length of time series is determined by using a dynamic time-delay window.Then,the optimal timedelay term is determined by the algorithm of solving the maximum average value,and the time-delay correlation model is constructed.Finally,the possible leading,contemporaneous and lagging relationships between the influencing factors and crop drought losses in Henan Province are analyzed.The results show that under a reasonable length of the time-delay window,the method of this paper is used to analyze the time-lag values of the factors affecting crop drought losses in Henan Province,which are pretty representative.The effectiveness of this method is verified by further selecting different time-delay window lengths.(3)Aiming at the interaction effect relationship between the influencing factors of crop drought loss in Henan Province,the traditional multivariate prediction model believes that the system influencing factors are independent of each other and affect the system characteristic sequence,ignoring the interaction effect between the system influencing factors.Based on the results of factor identification and correlation analysis in the above two chapters,this section constructs an interactive effect EIGM(1,N)model to predict and analyze crop drought losses in Henan Province.Firstly,with the help of the idea of a simple linear relationship,the interaction effect term is introduced into the grey GM(1,N)model.Secondly,the parameters are optimized by the leastsquares method to improve the accuracy of the model,and the interaction effect EIGM(1,N)model is constructed.Finally,the prediction model of crop drought loss in Henan Province is carried out.The results show that the method has high accuracy in the simulation and prediction of crop drought losses,which further verifies the reasonable validity of the grey prediction model in predicting crop drought losses. |
