The invention discloses an LSTM short-time traffic flow prediction method based on an improved PSO algorithm. The method comprises the following steps: (1) population initialization; (2) parameter setting; (3) establishment of an input and output relational expression according to LSTM; (4) adaptive functions; and (5) termination conditions. the LSTM method is applied to short-term traffic flow prediction, the PSO algorithm prediction results adopting different inertia weights are as follows: the convergence of the segmented inertia weight is faster than the progressive increase inertia weight, the precision is higher than the progressive decrease inertia weight, and the problems that the convergence speed of the progressive increase inertia weight is low, the prediction precision is high, the convergence speed of the progressive decrease inertia weight is high and the prediction precision is low are solved.
一种基于改进PSO算法的LSTM短时交通流预测方法,包括以下步骤:(1)种群初始化,(2)参数设定,(3)根据LSTM建立输入输出关系式,(4)适应函数,(5)终止条件。应用LSTM方法进行短时交通流预测,采用不同惯性权重的PSO算法预测结果如下,可见分段惯性权重收敛快于递增惯性权重,精度高于递减惯性权重,解决了递增惯性权重收敛速度慢,预测精度高,递减惯性权重收敛速度快,预测精度低的问题。
LSTM short-term traffic flow prediction method based on improved PSO algorithm
一种基于改进PSO算法的LSTM短时交通流预测方法
2021-04-27
Patent
Electronic Resource
Chinese
Short-term traffic flow prediction method based on CASSA-LSTM algorithm
European Patent Office | 2022
|Short-term traffic flow prediction method based on improved LSTM
European Patent Office | 2021
|Short-term traffic flow prediction method based on Conv1D-LSTM model
European Patent Office | 2023
|