Signal control strategy based on deep time division method

In order to make up for the shortcomings of the existing time division methods in complexity and time division effect, this paper proposes a signal control strategy based on deep time division method. Firstly, the ordered clustering algorithm based on Fisher optimal segmentation is used to segment the period. Secondly, to compensate for the dense distribution of segmentation points in the period with low saturation caused by data volatility, deep segmentation is performed in the period with the largest segmentation result. At last, the state critical points of traffic data are obtained by K-means algorithm, and the segmentation points in the same state are merged. Finally, the division of time periods and the corresponding traffic status of each period can be obtained. The VISSIM simulation results show that the proposed method can effectively reduce the queue length and average delay.