基于VMPSO-RBF神经网络的话务量预测

Traffic forecasting based on VMPSO-RBF neural network algorithm

为了更快速、准确地预测话务量,提出了速度变异的粒子群算法(VMPSO),并与RBF算法相结合,形成速度变异的粒子群—RBF(VMPSO-RBF)神经网络算法,并且来训练神经网络,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RBF神经网络方法和PSO-RBF神经网络方法相比较,该文提出的方法预测精度更高,收敛速度更快。

Aiming at improving the traffic forecast accuracy and speed,the Velocity Mutation Particle Swarm Optimization（VMPSO） algorithm is presented.It is combined with the Radial Basic Function algorithm to form Velocity Mutation Particle Swarm Optimization and Radial Basic Function（VMPSO-RBF） neural network algorithm,and the network is trained by the VMPSO-RBF algorithm.Then it can determine the parameters of the network from the data.Then we can forecast the traffic of mobile company by the model.The experimental results show that the forecast accuracy of the method is more accurate and it is faster than that of the RBF neural network and the PSO-RBF neural network.