基于随机拓扑优化提升回声状态网络的交通流预测

Study on Traffic Flow Prediction Based on Boosting Echo State Networks with Stochastic Topology Optimization

交通流预测是交通智慧化的重要任务。为了提高短时交通流的预测精度,解决单一模型预测精度不足、易受噪声干扰等特点,提出了一种基于随机拓扑优化的提升回声状态网络(echo state networks, ESN)的预测方法。该方法以简单的回声状态网络为基本构成单元,利用随机拓扑优化策略对回声状态网络的拓扑结构进行优化选择,然后利用基于误差补偿的提升算法提高整体模型的预测精度。通过对随机拓扑优化策略和提升算法在实际交通流预测问题中的性能分析,验证了所提方法的可行性和有效性,同时可为其他弱预测学习器的学习性能改进提供参考。

Traffic flow prediction is an important task of traffic intelligence.To increase the prediction accuracy of short-time traffic flow,and address the characteristics of insufficient prediction accuracy of single model and its vulnerability to noise interference,a novel enhanced echo-state network prediction approach based on random topology optimization and boosting strategy is proposed.This method starts with a simple echo state network as the basic unit,whose topology is optimized by using a stochastic topology optimization technique.Then the boosting strategy based on error compensation is used to increase the prediction accuracy of the entire model.We demonstrate the feasibility and effectiveness of the proposed method by demonstrating the performance of the random topology optimization strategy and boosting algorithm in real traffic flow prediction issues,while also offering a reference for enhancing the learning performance of other weak prediction learners.