论文部分内容阅读
短时交通流预测是实现交通流诱导的关键技术之一.针对目前短时交通混沌预测模型预测结果差异较大的问题,归纳了4种基于混沌理论的短时交通流预测模型:RBF神经网络模型、最大Lyapunov指数模型、局域线性模型和Volterra滤波器自适应预测模型,并对这4种预测模型进行了比较研究.应用4种预测模型对几个典型的非线性系统进行预测,验证了算法的准确性.然后用这4种预测模型对微观实测交通流的时间序列进行实证分析.仿真结果表明,4种预测模型对典型混沌时间序列具有很好的预测效果;而对实测交通流预测,其预测精度和稳定性较差,但可以满足实时交通流预测的需要.
Short-term traffic flow prediction is one of the key technologies to achieve traffic flow guidance.According to the large difference between the prediction results of short-term traffic chaotic prediction models, four kinds of short-term traffic flow prediction models based on chaos theory are summarized: RBF neural network Model, the largest Lyapunov exponent model, local linear model and Volterra filter adaptive prediction model, and compared the four prediction models.Using four prediction models to predict several typical nonlinear systems, verify And the accuracy of the algorithm.And then, the four kinds of prediction models are used to empirically analyze the time series of the microscopic measured traffic flow.The simulation results show that the four prediction models have a good prediction effect on the typical chaotic time series, , Its prediction accuracy and stability are poor, but it can meet the needs of real-time traffic flow prediction.