铁路客运专线模糊k近邻客流预测模型

Fuzzy k-nearest neighbor passenger flow forecasting model of passenger dedicated line

客运专线客运量在短时期内体现准周期的规律性变化,且受多种因素的影响呈现出一种复杂的非线性特点。传统的预测方法不能完全反映客流量准周期性和非线性的特点,预测结果误差相对较大。为更准确地预测铁路客运专线客运量,通过分析客运专线的客流特征,总结相邻时段客流变化规律,在确定相邻时段之间客流变化率的基础上,将客流变化情况划分为8个不同的等级,依据客流变化情况划分的不同等级对客流变化率模糊化,并利用客流变化率模糊值的时序关系,建立客运专线模糊k近邻客流预测模型。通过实例分析,与其他预测方法进行比较,证明该模糊k近邻客流预测结果误差更小,精度更高,为预测铁路客运专线客运量提出一种新思路。

Passenger flow of passenger dedicated line shows the quasi-periodic variations in the short-term forecasts, and also shows complex nonlinear characteristics because of many factors. The traditional prediction model can＇t fully reflect quasi-periodic and nonlinear characteristics of the passenger flow, which result in larger errors in forecast results. In order to forecast the passenger flow more accurately, the passenger flow characteristics of the high-speed railway were analyzed, and variation of passenger flow in the adjacent period was summed up. Passenger flow change rate was divided into different grades and fuzzified on the basis of passenger flow change rate between adjacent periods. Also, fuzzy k-nearest neighbor prediction model was established on the basis of the fuzzy values timing relationship of passenger flow change rate. By comparing it with other predictive methods, the prediction result of fuzzy k-nearest neighbor prediction model is proved to be more accurate and precise, thus providing a new idea for the railway passenger flow forecast.