基于Kalman滤波的地铁换乘客流预测

Metro transfer passenger forecasting based on Kalman filter

针对地铁换乘通道的换乘客流量,提出了利用Kalman滤波进行短时客流量预测的方法.基于Kalman滤波原理对地铁换乘客流系统构建状态方程,并根据历史数据对状态方程中的状态转移矩阵进行标定,然后运用灰色关联分析的方法来确定该状态转移矩阵在待预测时间序列上的值,进而实现客流量的预测.以北京地铁西单站换乘通道为例,从平日和假日两方面分别对该换乘通道一周内,早高峰时期客流量进行了短时预测.

A short-term forecasting method of passenger flow in the metro transfer channel based on Kalman filter is proposed in this paper. The state equation of the system of metro passenger flow is formulated first. Then the state transition matrix of the state equation is obtained based on historical data. After that, a grey relation analysis method is applied in solving the state transition matrix of the time series to be tested. By this, the problem of the short-term forecasting of metro passenger flow can be solved. At last, the metro passenger flow forecasting of Xidan station in Beijing is taken as an example, and the short-term forecasting of the passenger flow in the morning rush hour in one week is proceeded from the two respects of weekdays and weekends.