论文部分内容阅读
GREENLAB-Maize虚拟模型采用并行模拟植株功能与结构的机制来模拟玉米植株在不同环境条件(目前只考虑气象条件)下的生长。该模型逐时段模拟植株光合产量和光合产物向各个器官分配的数学公式的主要参数不能通过简单的方法确定,称为隐含参数,是通过将该模型与非线性最小二乘法结合,并以田间植株生长的测定值为目标文件而反求出的。本文首先对参数反求方法进行了改进,结果表明,以植株多个生长时段的测定数据为目标文件来反求参数值时,其变异系数明显比只用某一个时段的测定数据时低。从模型隐含参数的稳定性和模型模拟效果两方面对该模型进行了评估。对气象条件不同而其他条件相同的4组玉米试验、玉米不同生长阶段以及不同取样植株个体等情况下所反求的隐含参数值的稳定性统计分析表明,模型隐含参数值因气象条件、生长阶段以及植株个体差异而产生的差异均不显著,可以认为它们是植株生长内在的、稳定的参数。本文还对该模型的植株形态模拟模块进行了改进。模型较好地模拟了植株的生长状况,但还有必要对植株光合生产模块等进行改进。
The GREENLAB-Maize virtual model employs a mechanism that simulates plant function and structure in parallel to simulate the growth of corn plants under different environmental conditions (currently only considering meteorological conditions). The model can simulate the photosynthate production and the distribution of photosynthetic products to various organs on a period-by-period basis. The main parameters of the mathematical formula can not be determined by a simple method, which is called implicit parameter. By combining the model with nonlinear least square method, The measured value of plant growth is obtained from the target file. In this paper, the method of parameter inversion is improved firstly. The results show that when the measured data of multiple growth periods are taken as the target file to reverse the parameter values, the coefficient of variation is obviously lower than that of the measured data of only one period. The model was evaluated from the stability of model implicit parameters and model simulation results. The statistic analysis of the stability of the implicit parameter values under different weather conditions and other conditions of the same four sets of corn experiments, corn different growth stages and different samples of individuals and so on, the implicit parameter values of the stability analysis shows that the model implicit parameter values due to meteorological conditions, Growth stage and plant individual differences are not significant differences, can be considered that they are inherent and stable plant growth parameters. This article also improved the model of plant morphology simulation module. The model better simulates the growth of the plant, but it is still necessary to improve the photosynthetic production module of the plant.