数据挖掘技术在固定型交通检测器配置优化中的应用

结合固定型交通检测器空间配置的四条原则和配置密度优化步骤，提出固定型交通检测器配置优化的数据挖掘方法。设计6种高速公路出口匝道的固定型交通检测器配置密度方案作为实例研究对像，运用数据挖掘技术的时间序列指数平滑方法、ARIMA方法和神经网络方法分别建立高速公路出口匝道小时交通量Winters预测模型，ARIMA预测模型及神经网络预测模型。采用网络搜索技术确定Winters模型参数，设计一种比传统ARIMA模型参数估计方法更精确的算法程序估计ARIMA模型参数，采用三项误差指标评价模型预测效果，根据预测结果及高速公路事件管理交通参数精度要求确定可行方案及最佳方案，研究结果表明：在保证满足ITS对交通参数精度要求的同时，通过数据挖掘技术降低了交通流信息采集固定型检测器的配置密度及成本，为固定型交通检测器配置密度的优化提供了一种简单可行的新方法。

Application of data mining technology in the optimization of fixed traffic detector deployment

Designed six project of fixed traffic detector deployment for freeway off-ramp according to four rules. Establishing Winters, ARIMA and feed-forward nerual network hourly traffic volumes forecasting model for freeway off-ramp by applying data mining tim series technology, namely exponential smoothing, ARIMA and feed-forward neural network method. Applying grid searching techniques to find exponential smoothing parameters. Proposing an algorithm of finding ARIMA parameters,which is more accurate than traditional method. Applying three error index to evaluate accuracy of forecasting model, The results indicate that data mining technology is a simple and feasible method in the application of opimization of fixed traffic detector deployment, which decreases the detector density ,cost and meets the request for ITS application,