基于信息粒化的数据拟合模型对城市主干道交通违法数据分析——以南京市中心城区为例

Data fitting model based on information granulation for data analysis of urban trunk road traffic violation -- a case study of downtown Nanjing

摘要:

文章通过对城市交通大数据的分析，研究中心城区主干道路交通违法行为，利用优化的数学模型进行数字化及可视化分析，找出交通违法行为的现状和分布规律。采用粒计算的思想对数据进行信息粒化，在周期性的数据中找到数据的总体波动，剥落Low层数据，筛选出Up层和R层数据。选取不同模型对数据进行线性拟合与对比，显示Fourier级数模型拟合结果更优。对每条路线的车辆交通违法信息逐一分析，结果显示二号路线7-8月份出现车辆违法56.38次，在四号线路线六月份出现8次。文中建立新的数学模型来处理周期性数据，模型精确度高，能明确评估城市交通违法情况。

Abstract:

Based on the analysis of city traffic data, the illegal behavior research center of urban trunk road traffic, analyze the digitalization and visualization by using the optimized mathematical model, to find out the current status and distribution of traffic violations. The idea of granular computing is applied to data granulation, and the overall fluctuation of data is found in the periodic data. The data of Low layer is peeled off and the data of Up layer and R layer are screened out. The linear fitting and contrast of the data by different models show that the Fourier series model is more suitable. For each route of vehicle traffic violations information one by one analysis, the results show that route two, 7-8 months of vehicle violations 56.38 times, line four in June, 8 times. A new mathematical model is proposed to deal with the periodic data. The model has high accuracy and can be used to evaluate the traffic violation in the city.