In this paper,we present a third-order central weighted essentially nonoscillatory(CWENO)reconstruction for computations of hyperbolic conservation laws in three space dimensions.Simultaneously,as a Godunov-type central scheme,the CWENOtype central-upwind scheme,i.e.,the semi-discrete central-upwind scheme based on our third-order CWENO reconstruction,is developed straightforwardly to solve 3D systems by the so-called componentwise and dimensional-by-dimensional technologies.The high resolution,the efficiency and the nonoscillatory property of the scheme can be verified by solving several numerical experiments.
Van Genuchten模型(简称VG模型)是目前运用最为广泛的土壤水分特征曲线模型,提出适宜的优化算法进行模型参数识别也是一个非常重要的研究方向。针对标准的粒子群算法易陷入局部最优的缺点,给出了一种多邻域粒子群算法,可以有效地克服粒子群算法易陷入局部最优的缺点,并利用该算法对VG模型参数进行识别,最后用所求解的参数对不同类型土壤持水性能进行了试验。数值实验结果表明,多邻域粒子群算法能够有效地应用于VG模型的参数识别,与其它算法相比在性能和精度上都有所提高,而且对参数的取值范围也可以较大地放宽。因此,多邻域粒子群算法可以作为VG模型参数识别的一种新方法。
