The effective analysis and utilization of urban traffic big data is the key to the intelligent control of urban traffic.However,there are many problems in traffic data such as the too many kinds of traffic data,low data quality and non-standard output format.It will result in data not being fully and effectively utilized.At the same time,a large number of missing,redundant and abnormal data will also affect the accuracy and reliability of subsequent traffic flow calculation. Based on this problem,a set of multi-source traffic data reconstruction based on cellular automaton model, multi-dimensional quality identification and diagnosis, spatio-temporal restoration,confidence analysis algorithm flow is proposed.T he reasonableness of the algorithm is verified by the actual data in the city.T he algorithm has been effectively applied in Hangzhou"city brain"project,and the social value is significant.%城市交通大数据的有效分析利用,是城市交通智能化管控的关键.交通数据存在种类繁多,质量参差不齐,输出格式不规范等问题,数据并未被充分有效利用.同时,大量的缺失、冗余及异常数据还会影响到后续交通流指标计算的精度与可信度.基于此,提出了一套城市基于元胞自动机模型的多源交通数据重构,基于多维度的质量识别与诊断、时空修复以及置信分析算法流程,并利用城市内实际数据对算法合理性进行验证.该算法已在杭州"城市大脑"项目中得到有效应用,社会价值显著.
展开▼
