论文部分内容阅读
为了定量分析交通流系统的复杂性,引入算法复杂度和近似熵,通过速度时间序列的算法复杂度估计系统周期性成分的比率,在重构序列时通过取多个划分区间来提高算法复杂度的估计能力。计算近似熵时,先由速度序列得到速度变化率序列以去除趋势,然后通过速度变化率序列的近似熵估计系统在结构变化上的复杂性。对实测交通流数据序列的计算表明:在序列长度超过600时可以得到算法复杂度,序列长超过300时可以得到近似熵;交通流的算法复杂度和近似熵在同步状态时较低,拥挤状态时增大,在自由状态时最大。因此,不同的算法复杂度和近似熵对应不同状态下的交通流,算法复杂度能分析较长的交通流序列,近似熵可以分析较短的交通流序列。
In order to quantitatively analyze the complexity of traffic flow system, the algorithm complexity and approximate entropy are introduced, and the ratio of the periodic components of the system is estimated by the algorithm complexity of the velocity time series. When the sequence is reconstructed, the algorithm complexity is improved by taking multiple partitioning intervals Estimated ability. When calculating the approximate entropy, the speed change rate sequence is first obtained to remove the trend, and then the complexity of the system is estimated by the approximate entropy of the rate change rate sequence. The calculation of the measured traffic flow data sequence shows that the algorithm complexity can be obtained when the sequence length exceeds 600, and the approximate entropy can be obtained when the sequence length exceeds 300. The complexity and approximate entropy of the traffic flow are lower when the synchronization state is in the congested state When the increase, the maximum in free state. Therefore, different algorithm complexity and approximate entropy correspond to the traffic flow in different states. The complexity of the algorithm can analyze the longer traffic flow sequence, and the approximate entropy can analyze the shorter traffic flow sequence.