Simulation Analysis Of Highway Abnormal Events' Impact Area Based On DDDAS

Abnormal events of highway(such as vehicle fault, traffic accidents, etc.) will reduce the efficiency of road traffic, and under the case of heavy traffic, it will even cause traffic jam and vehicle waiting in line. The reliable estimation of exceptional events' impact range and development tendency comes to be the precondition and basis of making targeted traffic management and control strategy, moreover it's of great significance in improving service level of freeway as well as the smooth running.For the current estimation models of abnormal incidents of expressway, thery are mostly based on the traffic flow theory. However, there exist great difficulty in engin- ering application, since the existing traffic parameter detection accuracy can't satisfy the input of models. Aiming at this problem, the paper brought in the emulation analysis to study the space-time correlation characteristics of highway traffic flow, meanwhile the paper combined with vehicle detector features and then proposed traffic flow parameters calibration methods as well as driving behavior parameter correction methods based on VISSIM simulation system. In addition that, the paper established the emulation analysis system of highway excepetional evens' impact area based on DDDAS after the in-depth analysis of particle filtering. The major works of the project are listed as follows:(1) Simulation model parameters calibration of traffic flow and driving behavior parameters correction. Based on the characteristics of freeway traffic flow and associated with the analysis, as well as combining with the history of vehicle inspection data, traffic flow simulation model parameters were calibrated. The default value for the parameter calibration of the simulation model driving behavior is inaccurate, the paper combined with poor methods of single factor sensitivity analysis to determine the core parameters for correction. And have parameter correction studied based on genetic algorithm simulation model driving behavior method. Finally, the actual vehicle inspection data were model validation. The simulation results show that the model can accurately establish the trend of traffic flow on the road simulation.(2) Analysing the data assimilation method of traffic simulation models based on particle filter algorithm. Combined with traffic wave theory and threshold theory, it introduces \highway vehicle inspection data preprocessing methods. It studies the particle filter data assimilation method and establishes emulation analysis system of highway excepetional evens based on DDDAS on account of particle filtering.In the meantime, given a numerical example to illustrate the proposed method. The results showed that the traffic simulation model based on particle filter can continue to assimilate real-time data to achieve accurate estimate of the position of the event clogging roads and real-time queue length.Finally the paper introduces the designation and achievement of traffic simulation model and then presents and partition the modules precisely. Combining Beibei tunnel section vehicle detector data of G75 highway, we choose typical and authentic traffic excetional events to build relative simulation scene, and thus effectively vefifying the emulation analysis system of highway excepetional evens' impact area based on DDDAS.