本文在综合分析现有各种类型的交通流模型的基础上，提出更为符合实际的微观模型和宏观模型。具体地说，我们提出了一种全速度差车辆跟驰模型，一种考虑前车速度效应的元胞自动机模型，另一种能较好模拟同步流的元胞自动机模型，以及一种能体现交通流各向异性特征的速度梯度宏观连续模型。通过理论分析和数值模拟，研究了交通流中的各种非线性现象，并探讨了各种不同类型的模型之间的相互联系。

This paper presents more realistic models,including the microscopic ones and the macroscopic ones, based on the synthetical analysis of the existing ones. Specifically, we present a full velocity difference(FVD) car-following model, a cellular automata(CA)model that considers the velocity effect of the preceding car, another CA model that can describe the synchronized flow, and a speed gradient (SG) macroscopic model which can describe the anisotropic properties of traffic flow. The various nonlinear phenomena in traffic flow are investigated through theoretical nanlyses and numerical simulations, and the relationship between different kinds of models are discussed.