遗传BP神经网络在网络流量预测中的应用

由于BP神经网络本质上采用的是梯度下降算法,具有收敛速度慢、容易陷入局部极小点等缺陷.针对这种情况,用具有良好全局搜索能力的遗传算法来改进BP神经网络模型,对神经网络的初始权值和阈值进行优化.仿真结果表明,遗传BP神经网络具有良好的预测效果,预测精度比传统的BP神经网络要高,误差更小,说明了遗传BP神经网络对网络流量预测是高效可行的.

Application of Genetic BP Neural Network on Network Traffic Prediction

BP neural network essentially uses gradient descent algorithm, which converges slowly and easily gets into the local extremurrL Global search capabilities of genetic algorithm is used to improve BP neural network and optimize BP network initial weights and thresholds. Simulation results show that the genetic BP neural network has good prediction effect, the prediction accuracy is higher, and errors are smaller compared to traditional BP neural network, which illustrates that the model is feasible and effective to be applied in network traffic prediction.