The invention provides an urban traffic flow prediction method and system, and the method comprises the steps: collecting target urban traffic flow sequence data, carrying out the wavelet transformation of the sequence data, obtaining a wavelet subsequence, carrying out the modeling prediction of the subsequence according to a GRU neural network, optimizing the hyper-parameters in a GRU network model through a genetic algorithm, and constructing a GA-GRU optimal combination model. And finally, carrying out wavelet reconstruction on the optimal prediction result of each sub-sequence, and outputting a final prediction result. According to the urban traffic flow prediction method provided by the invention, accurate decomposition of sequence data is realized through wavelet decomposition, modeling prediction is carried out on subsequences through a GRU neural network, optimization is carried out on a GRU model through a genetic algorithm so that the subsequences can restore real data better, and finally, each optimal subsequence is reconstructed. And the final accurate prediction of the traffic flow data is realized.
本发明提出一种城市交通流量预测方法及系统,该方法包括:收集目标城市交通流量序列数据,对序列数据进行小波变换,获得小波子序列,根据GRU神经网络对子序列进行建模预测,利用遗传算法优化GRU网络模型中的超参数,构建GA‑GRU最优组合模型,得到子序列最优预测结果,最后对各个子序列的最优预测结果进行小波重构,输出最终预测结果。本发明提出的一种城市交通流量预测方法,通过小波分解实现序列数据的精确分解,通过GRU神经网络对子序列建模预测,通过遗传算法对GRU模型寻优以使子序列更还原真实数据,最后将各最优子序列重构,实现了对交通流量数据的最终精准预测。
Urban traffic flow prediction method and system
一种城市交通流量预测方法及系统
2023-12-05
Patent
Elektronische Ressource
Chinesisch
Europäisches Patentamt | 2023
|2-dimensional flow prediction based intelligent urban traffic-flow prediction method
Europäisches Patentamt | 2016
|