中文摘要：

为了制定合理的不利天气下交通管理方案，采用离散化动态交通路网容量描述道路网容量变化情况．定义了不利天气下离散化动态路网容量，提出了与时空消耗概念相类似的基于通行能力的离散化动态路网容量计算模型，并对模型中的通行能力和平均出行距离两个变量进行了说明：采用径向基函数神经网络模型处理不利天气下通行能力与众多非线性影响因素；用概率统计理论推导出行距离的概率密度函数，概率密度函数的均值即为交通平均出行距离．通过对哈尔滨市暴雨天气下道路实际数据的分析，验证了模型的可行性和有效性．

英文摘要：

For the sake of establishing a reasonable traffic management scheme under adverse weather, the discrete dynamic traffic network capacity was adopted to describe the changes of road network capacity. The concept of discrete dynamic road network capacity under adverse weather was defined and a discrete dynamic mathematical model was put forward based on traffic capacity, which is similar with the concept of spatio-temporal dissipation. The traffic capacity and average trip distance in this model were explained. The radial basis function neural network was applied to dispose the relationship between traffic capacity and many non-linear influencing factors under adverse weather. Based on probability and statistics theory, the probability distribution function of trip distance was deduced, and the expectation of this function is the average trip distance. Through the analysis of actual road data under storm weather in Harbin, the feasibility and validity of this model were validated.