无检测器道路交通流数据质量检测方法研究--基于灰色系统理论

In the traffic flow field, many quality problems exist inevitably in detected data which has multiple attributes of high dimensional, such as speed, flow, latitude, longitude, errors, and redundancy, etc. Conventional data quality detection methods require large number of initial data while traffic flow data at non-detector road is very limited, in order to deal with this contradiction, a new non-detector roads traffic flow data quality detection method based on grey system theory is put forward. This method provides a new thought to solve the increase in time complexity and separate multiple attribute links which were caused by the traffic flow data multiple attributes. Finally, using the probe car traffic flow data which covers a local road network of Hangzhou to validate the efficiency of the model by adopting the method put forward in the article.