【摘要】 阐述为提高短期交通流量数据预测的精度，提出一种基于变分模态分解（VMD）和深度极限学习机（DELM）的组合预测模型，以解决传统预测模型在处理非线性和非平稳时序性质数据时存在的不足。首先，采用VMD方法将原始交通流量数据分解为若干模态分量。然后，针对不同模态分量的特性，分别采用DELM模型进行预测。最后，将各模态分量的预测结果叠加，得到最终的预测值。更多还原

【Abstract】 This paper describes a combined prediction model based on variational mode decomposition(VMD) and deep extreme learning machine(DELM), in order to improve the prediction accuracy of short-term traffic flow time series data, aiming at the shortcomings of traditional prediction models in dealing with nonlinear and non-stationary traffic flow time series data.Firstly, the original traffic flow data is decomposed into several modal components by VMD method. Then, according to the characteristics of different modal components, DELM model is used to predict respectively. Finally, the predicted results of each modal component are superimposed to obtain the final predicted value.更多还原