摘要
提出了内外循环集合变分数据融合算法,采用多个具有连续、单极值的局地二次函数序列来逼近实际的非线性目标泛函,使得这些局地二次函数的极值点序列收敛于非线性目标泛函的全局最小点。算法不需要伴随模式及修改原来的非线性数值模式,使用此方法融合不可微过程观测数据时具有方便、程序设计简单等优点。针对云降水中不可微过程的数据融合数值,试验结果表明:内外循环集合变分方法是可行有效的,可以提高不可微过程的融合精度。
This paper proposes an inner/outer loop ensemble-variational algorithm （I/ OLEnVar） to DF. It uses several continuous sequences of local linear quadratic functions with single extreme values to approximate the actual nonlinear CF so as to have extreme point se- quences of these functions converge to the global minimum of the nonlinear CF. This algorithm requires no adjoint model and no modification of the original nonlinear numerical model, so it is convenient and easy to design in fusing the observational data during the non-differentiable process. Numerical experimental results of DF for the non-differentiable problem in moist physical processes indicate that the I/OLEnVar algorithm is feasible and effective. It can in crease the assimilation accuracy and thus obtain satisfactory results.
出处
《金陵科技学院学报》
2016年第1期50-53,共4页
Journal of Jinling Institute of Technology
基金
江苏省自然科学基金(BK20131065)
国家自然科学基金(41175090
41375106)
关键词
不可微过程
内外循环
集合变分数据融合
non-differentiable
inner/outer loop
ensemble-variational data fusion
