CAJ | 학술논문

为提高城市轨道交通车站客流预测模型精度，简化模型数据需求规模，提出基于空间加权的 LS-SVM城市轨道交通车站客流预测模型。基于交通网络距离重新划分车站的影响范围，提出分距离影响带的线型和指数型空间权重系数方程，结合空间权重系数，输入区域特征变量和车站属性变量构建城市轨道交通车站客流 LS-SVM预测模型，运用动态改变惯性权重自适应粒子群优化算法(DCW-APSO)对模型参数进行优化选取。应用模型预测2011年成都市地铁1号线部分车站客流，并与其他模型进行比较，结果表明：模型明显提高客流预测精度，简化数据需求量，作为城市轨道交通客流预测的补充模型可以进一步提高系统的可靠性。
위제고성시궤도교통차참객류예측모형정도，간화모형수거수구규모，제출기우공간가권적 LS-SVM성시궤도교통차참객류예측모형。기우교통망락거리중신화분차참적영향범위，제출분거리영향대적선형화지수형공간권중계수방정，결합공간권중계수，수입구역특정변량화차참속성변량구건성시궤도교통차참객류 LS-SVM예측모형，운용동태개변관성권중자괄응입자군우화산법(DCW-APSO)대모형삼수진행우화선취。응용모형예측2011년성도시지철1호선부분차참객류，병여기타모형진행비교，결과표명：모형명현제고객류예측정도，간화수거수구량，작위성시궤도교통객류예측적보충모형가이진일보제고계통적가고성。
In order to improve the accuracy of the urban rail transit (URRT)station ridership forecast model and to simplify the model data scale,the direct ridership forecast model of urban rail transit stations based on the spatial weighted LS-SVM was proposed.The station influencing area was calculated by transport network distances,and the linear and exponential spatial weight equations associated with different distance bands were put forward.By inputting area characteristics variables and station attribute variables based on spatial weight coefficients,The LS-SVM forecast model of URRT station ridership was built.The dynamic change inertia weight adaptive particle swarm optimization algorithm (DCW-APSO)was applied to optimize selection of mod-el parameters.The proposed model was applied to forecast the ridership of some sations of Chengdu Metro Line in 2011,and the results were compared with other types of models.The comparison show that the pro-posed model improves forecast accuracy,simplifies model data demand and upgrades the model system reliabili-ty as a supplementary model of URRT passenger flow forecast.