基于VMPSO-BP神经网络的话务量预测

Traffic Forecasting based on VMPSO-BP Neural Network Algorithm

为了更快速、准确地预测移动话务量,提出了速度变异的粒子群算法(VMPSO),并与BP算法相结合,形成速度变异的粒子群—BP(VMPSO-BP)神经网络算法,用以训练神经网络,从而优化了神经网络的参数,最后对移动话务量进行预测。与传统BP神经网络方法和PSO-BP神经网络方法相比较,并且通过实验数据的分析以及对预测结果地比较,速度变异的粒子群—神经网络预测方法精度更高,收敛速度更快,从而更好地实现了对移动话务量地预测。

For rapid and accurate forecast of the traffic,the Velocity Mutation Particle Swarm Optimization（VMPSO） algorithm is presented.It is combined with the back propagation algorithm,thus forming Velocity Mutation Particle Swarm Optimization and Back Propagation（VMPSO-BP） neural network algorithm for the training of neural network and the optimization of neural network parameters.Then the traffic of mobile company is forecasted by the model.The experimental results and also the analysis of experimental data and comparison of predicted results show that its forecast accuracy is more accurate and its speed is faster than that of the traditional BP neural network and the PSO-BP neural network.It is a better realization of mobile traffic forecast.