农业机械学报2018，Vol.49Issue(4)：232-240,9.DOI:10.6041/j.issn.1000-1298.2018.04.026
基于SCE-UA算法的小麦穗分化期模拟模型参数优化
Parameters Optimization of Wheat Spike Differentiation Stages Model Based on SCE-UA Algorithm
摘要
Abstract
WheatGrow model is a mechanism model for the simulation of growth and development process of wheat spike differentiation,but the crop varietal parameters to drive the model are more difficult to obtain,which greatly limits its application.Shangqiu,which is in Henan Provice was taken as the studying area and the sensitivity of varietal parameters of WheatGrow model was analyzed with the method of one-at-a-time (OAT).On this basis,the cost function was constructed with start date of heading as the constraint condition,and shuffled complex evolution method developed at the University of Arizona (SCE-UA) was applied to search for optimal varietal parameters.At last,a series of experiments on spike differentiation stages were carried out in two years (from 2015 to 2016 and from 2016 to 2017) to verify optimized results and the model.The results showed that intrinsic earliness (IE) had the most significant effect on the simulation results of spike differentiation stages,temperature sensitivity (TS) had higher sensitivity than photoperiod sensitivity (PS) and physiological vernalization time (PVT),and the sensitivity of physiological vernalization time (PVT) was the lowest of all varietal parameters.The mean absolute error (MAE) and root mean square error (RMSE) between the simulated and the observed values of the spike differentiation stages based on the optimized parameters were both less than three days,indicating that the SCE-UA algorithm can effectively obtain the optimal parameters of WheatGrow model.Therefore,the SCE-UA algorithm was a feasible optimization method for WheatGrow calibration and validation.关键词
冬小麦/模型参数优化/SCE-UA算法/WheatGrow模型/穗分化期Key words
winter wheat/parameters optimization/SCE-UA algorithm/WheatGrow model/spike differentiation stages分类
农业科技引用本文复制引用
刘峻明,潘佩珠,王鹏新,崔珍珍,胡新..基于SCE-UA算法的小麦穗分化期模拟模型参数优化[J].农业机械学报,2018,49(4):232-240,9.基金项目
国家自然科学基金项目(41471342) （41471342）
