基于多策略粒子群优化RBF的云资源预测模型

Cloud resource prediction model based on multi-strategy particle swarm optimization RBF

针对云计算资源利用率低等问题,构建基于多策略粒子群优化RBF神经网络的云资源预测模型(MPSO-RBF)。采用改进的粒子群算法对RBF神经网络参数进行优化,避免随机初始化参数引起的预测精度低等问题;对于粒子群容易陷入局部最优解等问题,采用动态惯性权重、自适应学习因子和变异粒子位置3种策略对粒子群进行改进,提高算法的寻优能力。基于云计算资源负载数据,将该模型与BP、RBF和PSO-RBF模型进行对比实验,验证了该模型具有良好的性能。

Aiming at the problem of low utilization of cloud computing resources,a cloud resource prediction model based on multi-strategy particle swarm optimization RBF neural network(MPSO-RBF)was constructed.The improved particle swarm optimization algorithm was used to optimize the parameters of RBF neural network,which effectively avoided the problem of low prediction accuracy caused by random initialization parameters.For the problem that particle swarm optimization is easy to fall into local optimal solution,three strategies of dynamic inertia weight,adaptive learning factor and mutation particle position were used to improve the particle swarm optimization,which improved the optimization ability of the algorithm.Based on cloud computing resource load data,the model was compared with BP,RBF and PSO-RBF models to verify the better performance of the proposed model.