Research And Application Of Crop Life Cycle Growth Model Based On Logistic And Gompertz Algorithms

Based on the single Logistic and Gompertz curves,the relationship between the growth regularity of quinoa in the whole life cycle and meteorological factors were studied.The requirements and control points of 6 meteorological factors,including relative humidity,sunshine,accumulated temperature,evaporation,water pressure and precipitation,were observed during the four stages of sowing,germination,development and maturity of quinoa.And the single Logistic and Gompertz regression equations were modified by the multi-factor model.Multivariate Logistic and Gompertz regression growth models being used to predict the relationship between each meteorological factor and quinoa biomass at different growth stages.The effects of predicting the growth regularity of quinoa biomass at different stages and controlling inflection point factors by items were comprehensively realized.This study provided a theoretical basis for the growth prediction model of quinoa planting and cultivation techniques in the experimental area.The specific work of this study was as follows.According to the characteristics of different varieties,the geographical and meteorological conditions of different planting areas,the main meteorological factors affecting the growth law of quinoa in the experimental area and the corresponding theoretical algorithm of growth prediction model were selected.The time series and characteristic information of six meteorological factors were collected,the exploration of natural environmental factors affecting quinoa biomass and the filling of database were realized.The single factor Logistic regression and Gompertz curve regression were used to explore the relationship between each meteorological factor and quinoa biomass in the planting area,the classification and correlation of individual control factors were achieved.Based on the selected single factor Logistic and Gompertz regression curves,the least square method and logarithm method were used to modify the single factor growth model equation.The data heteroscedasticity produced by time series data was solved,and the single factor Logistic and Gompertz regression prediction models were modified.SPSS statistical software was used to pre-process the data collected from the initial database and explored the correlation between independent variables and strain variables,screening and optimizing abnormal data,such as missing data and duplicate data being achieved pure data purification.With the help of program language and MATLAB software,the time-series quinoa biomass data of different stages and six different meteorological factors data of the same period were substituted into the revised multi-factor logistic and Gompertz regression prediction curves.The different coefficient parameter values,regression curve equation and fitting function images of the two curve models were obtained.In view of the six meteorological factors affecting the quinoa biomass in the quinoa planting experimental area of WanChang Town,Jilin Province in the same period,the established multifactor logistic and Gompertz regression growth models were used to predict the extent that the quinoa biomass was affected by meteorological factors in different periods.Grasping the maximum value and inflection point range of quinoa biomass growth rate under specific factors,the demand degree of quinoa for meteorological and other natural environment at different stages was calculated.Theoretical model support for planters was provided,and the planting technology system combining data rule and experience were achieved.The research results of this paper had been validated in the quinoa planting experimental area of WanChang Town,Jilin Province.