论文部分内容阅读
以实际调查得到的建筑容积率、建筑密度、总建筑面积、交通小区面积以及各类型土地利用面积等14个土地利用指标作为直接变量,建立交通发生量预测模型。相关分析的结果表明土地利用指标间存在着严重的多重相关性,普通的回归分析方法很难有效解决这一问题。采用偏最小二乘回归方法,通过构建主成分,采用交叉有效性检验确定成分个数,最终建立的交通发生量预测模型物理意义明确,精度较好。研究结果表明:居住用地、建筑容积率、总建筑面积等土地利用指标对交通发生量的影响最大。
Based on the actual land-use index of 14 land-use indexes, such as building volume rate, building density, total construction area, traffic area and various types of land use area, a traffic volume forecasting model was established. The results of the correlation analysis show that there is a serious multiple correlation between land use indicators, and ordinary regression analysis method is difficult to effectively solve this problem. By using partial least-squares regression method, the main components were constructed and the cross-validation test was used to determine the number of components. The traffic volume forecast model established finally has clear physical meaning and good accuracy. The result shows that land use index, such as residential land, building plot ratio and gross floor area, has the greatest impact on traffic volume.