并行广义神经网络的交通流预测

Traffic flow prediction based on parallel generalized neural network

实时、准确的交通流预测是智能交通诱导实现的前提和关键。针对BP神经网络学习过程收敛速度慢、容易陷入局部极小的缺点,引入智能神经元组成的广义神经网络建立交通流预测模型,同时给出基于训练集分解和动态通信模式的并行学习算法来提高广义神经网络的收敛速度,并利用大连市的实际交通流数据进行预测分析。实验结果表明,并行广义神经网络能够满足交通流量预测实时性、精确性的要求,具有一定的应用价值。

Real-time and accurate traffic flow prediction is crucial to the development of traffic flow guidance system.In order to solve the problem that BP network has the disadvantages of slow convergence speed and target function getting into local minimal value,this paper employs general neural network composed of intelligent neuron to forecast traffic flow.Meanwhile,a new parallel training algorithm based on training set decomposition is presented.This algorithm uses a dynamic communication profile.Using this algorithm to train general neural network and doing experiments with real traffic flow data of Dalian,experiment results show that the new algorithm improves both convergent speed and forecasting precision,and can meet practical requirement.