通过分析交通流量时间序列的特点，引入BP神经网络进行短时交通流预测。首先，分析了短时交通流量预测的意义及研究背景；然后，介绍了BP神经网络的结构模型、学习规则以及BP算法的改进算法；最后，通过BP神经网络对短时交通流进行预测，并分析了在各种不同条件下的预测情况。-through the analysis of the characteristics of traffic flow time series, introduces BP neural network for short-term traffic flow prediction. First of all, it analyzes the significance of short-term traffic flow prediction and research background. And then, it has discussed the model and structure, the study rule and the improved algorithm of BP neural network. Finally, by BP neural network to forecast the short-term traffic flow, it analyzes under different conditions of the neural network prediction.