交通流突变辨识组合算法

针对智能交通系统的开发，结合交通流特性，应用小波多分辨分析理论的Mallat 分解算法与RBF 神经网络建立交通流状态辨识组合算法。利用多种小波系数与交通流参数之间的相应变化规律进行RBF 网络输入参数设计，进而通过RBF 网络进行交通流状态突变的辨识。交通流状态的突变多与交通事件直接相关，故采用事件和非事件条件下的模拟数据对算法进行了离线测试。与传统算法的性能比较结果表明：组合算法在交通流状态突变辨识方面具有良好的性能。

Hybrid algorithm for traffic flow breakdown identification

Aiming at developing intelligent transportation system,a hybrid algorithm for traffic flow breakdown identification based on fast Mallat algorithm and RBF network is presented utilizing the characteristic of traffic flow.Utilizing the association between the wavelet coefficients and traffic flow,the set of approximate coefficients and detail coefficients of wavelet decomposition from traffic flow parameters is regarded as the input of the Radial Basis Funtion（RBF） network.And the state of traffic flow breakdown can be extracted from the RBF network.Traffic flow breakdown usually results from traffic incidents,a performance test is carried out using data obtained from the simulation under the condition of incident and non-incident.According to the results of test compared with other algorithms,the hybrid algorithm performs better than the other algorithms in traffic flow breakdown identification.