优化的BP神经网络短时交通流预测方法

Optimized BP neural network method for short-term traffic flow forecasting

针对传统的误差逆向传播(BP)神经网络方法在进行交通流预测时存在网络准确性差,权值设置敏感等缺点,提出一种基于改进蝙蝠算法(BA)优化BP神经网络的交通流短时预测方法。引入自适应惯性权重和加速因子对原始蝙蝠算法进行优化,提高其收敛速度及寻优精度;用改进的BA对BP神经网络的权值和阈值参数优化并构建BA-BP模型进行短时交通流预测。实验结果表明:与传统BP相比,该方法平均绝对误差降低了3.0785,均方误差降低了4.4710。

Aiming at the disadvantages of the traditional error back propagation(BP)neural network method has poor network accuracy and sensitive weight setting in traffic flow prediction,proposes a short-term traffic flow prediction method based on improved bat algorithm(BA)to optimize BP neural network.Firstly,adaptive inertia weight and acceleration factor are introduced to optimize the original bat algorithm,to improve its convergence speed and optimizing precision.Secondly,the improved BA is used to optimize the weight and threshold parameters of BP neutral network and construct BA-BP model for short-term traffic flow prediction The experimental results show that of compared with the traditional BP,the mean absolute error of this method is reduced by 3.0785,and the mean square error is reduced by 4.4710.