Study And Implementation Of Expressway Traffic Conditions Evaluation Considering Abnormal Data

Expressway traffic conditions evaluation system can provide theoretical support for the control measures and. The evaluation system relies on a reliable source of data, in practice, due to the fault detection system to detect changes in the environment, human improper operation of various factors, traffic information data acquisition will be extraordinary circumstances. These exceptions will directly affect the results of the evaluation data highway efficiency of the system, thus, highway traffic state evaluation method of data under unusual circumstances, can reduce the impact on the results of the evaluation data anomalies, reduce data abnormal situation of a systematic review false positive rate, thus providing more reliable information for the highwa y traffic perception incident management operations.anomaly data for the above evaluation system of highway running status affect the reliability of the results of the evaluation of the problem, starting from system analysis and evaluation of data sources and abnormal factors, fixed to the vehicle detector data and charge data for the typical case, the data from the main pretreatment levels and evaluation methods proposed expressway traffic running state, taking into account the abnormal condition data to improve data evaluation fault tolerance approach system, the main work is as follows:(1) vehicle inspection for missing data, wrong data parameter value(parameter value is 0, speed, high occupancy), the proposed data integrity and correctness of evaluation methods and identification methods, and proposed based on the moving average improved algorithm to fix vehicle inspection data, experimental results show that the proposed method can better approaching real data.(2) charges against loss of data, data errors(mileage and travel time data abnormal value), the proposed charging data completeness and correctness of evaluation methods and identification methods, and charges for different data anomalies, based on statistical method proposed the corresponding evaluation methods, experimental results show that the designed charging data to identify anomalies and effective repair method is available.(3) dynamic timing and differences in data traffic data driven based on the principle of classical fuzzy comprehensive evaluation method was improved by the dynamic changes in the data objectively, dynamic computing real-time weights replace the original method of weight values fixed weight, experiments show that the method can objectively evaluate the traffic running state, and abnormal data more inclusive.Finally, the above data pretreatment methods and improved evaluation methods using VS2010 development tools, design and implementation of the evaluation system, the results show that, compared to no repair in terms of data and evaluation methods to improve highway efficiency evaluation system, this article evaluation system to objectively evaluate traffic conditions, with stronger fault tolerance in case of the presence of abnormal data, affected by the abnormal data smaller, lower false positive rate, so as to provide a more objective basis for the highway support decision- making.