Multivariate Copula and Dynamic Copula are two important directions in the Copula theory and application. A graphical tool, vine, with cascading structure in the graphic modeling theory is introduced. And we take the bivariate conditional Copulas, instead of usual static Copulas, as the nodes of the vine, combine the multivariate copula and conditional copula, construct a multivariate dynamic copula based on vine structures, named multivariate dynamic vine copula. Empirical results demonstrate that, the multivariate dynamic copula gets higher likelihood than that of the multivariate static copula.