摘要
根据公交换乘枢纽换乘量的生成特点和影响交通换乘量的主要组成因素,研究发现交通换乘量具有"小样本、贫信息"的灰色特征,为此提出了一种基于灰色软计算的换乘需求量预测方法。该方法利用灰色系统原理建立灰色神经网络系统预测模型,通过采用遗传算法改进神经网络的性能,提高系统预测的精度。以兰州市市区公共交通枢纽规划为例,结合实际的道路交通调查数据,运用该方法对提出的交通枢纽方案进行了实证分析与评价。结果表明:改进的灰色神经网络能有效地改善预测精度;运用GA-GNN模型求解道路交通中的非线性问题对提高决策水平具有较大的现实意义。
Abstract： According to the characteristics of generating transfer volume at public transport transfer hub and the major factor influencing the transfer volume, the gray characteristics of the transfer volume, i. e. , ＂small sample, poor information＂, is discovered, thus, a novel method based on gray soft computing employed to transfer volume forecast is put forward. The method uses grey system theory to establish a grey neural network forecasting model, and uses genetic algorithm to improve the performance of the neural network and the precision of prediction system. Taking the planning of location for Lanzhou urban public transport transfer hub for example, this method is applied to analyse and evaluate all the proposed schemes in combination with the actual road traffic survey data. The result shows that （1） the improved gray neural network can effectively improve forecast precision; （2） using the GA-GNN model to solve the nonlinear problem of road traffic has obvious practical significance for improving decision-making level.
出处
《公路交通科技》
CAS
CSCD
北大核心
2013年第6期115-119,126,共6页
Journal of Highway and Transportation Research and Development
基金
甘肃省自然科学研究基金项目(0803RJZA020
096RJZA084)
关键词
交通工程
换乘枢纽选址
遗传灰色神经网络
公交换乘枢纽
traffic engineering
location for public transport transfer hub
genetic gray neural network
public transport transfer hub
