神经网络和改进粒子群算法在地震预测中的应用
|
摘要
提出了一种基于神经网络与改进粒子群算法的地震预测方法,该方法采用前向神经网络作为地震震级的预测模型,引入改进的粒子群算法对前向网络的连接权值进行修正。为了设计在全局搜索和局部搜索之间取得最佳平衡的惯性权重,基于粒子动态变异思想对粒子群优化算法进行改进,提出了一种动态变异粒子群优化算法,并将其应用于地震震级预测神经网络模型优化。在仿真实验中,将所提出的方法与另外两个采用不同算法的前向网络预测方法进行了比较。结果表明所提出的优化算法收敛速度最快,所得模型的预测误差最小,泛化能力最强,对地震的中期预测有很好的参考作用。
This paper proposed an earthquake prediction method based on neural networks and an improved particle swarm optimization algorithm.In this method,a feed forward neural network was applied to predict the level of earthquake,and a modified particle swarm optimization algorithm was applied to optimize the neural network model.In order to get weights of the optimal balance between the global search and local search,a Dynamic Mutational Particle Swarm Optimization(DMPSO) algorithm was designed by using the ideology of dynamic mutation.This algorithm was used to adjust weights of the feed forward neural network.The simulation results of the proposed method were compared with the simulation results of two feed forward networks with different training algorithms.The comparison results show that the prediction model with DMPSO has fastest convergence rate,the smallest prediction error and strongest generalization ability.In conclusion,the model with DMPSO is a good reference to the middle earthquake prediction.
This paper proposed an earthquake prediction method based on neural networks and an improved particle swarm optimization algorithm.In this method,a feed forward neural network was applied to predict the level of earthquake,and a modified particle swarm optimization algorithm was applied to optimize the neural network model.In order to get weights of the optimal balance between the global search and local search,a Dynamic Mutational Particle Swarm Optimization(DMPSO) algorithm was designed by using the ideology of dynamic mutation.This algorithm was used to adjust weights of the feed forward neural network.The simulation results of the proposed method were compared with the simulation results of two feed forward networks with different training algorithms.The comparison results show that the prediction model with DMPSO has fastest convergence rate,the smallest prediction error and strongest generalization ability.In conclusion,the model with DMPSO is a good reference to the middle earthquake prediction.
引文
[1]王炜,蒋春曦.BP神经网络在地震综合预报中的应用[J].地震,1999,19(2):118-126.
[2]秦长源,张淑亮.AR模型预测太原盆地地震活动的趋势[J].地震研究,1997,20(4):4-9.
[3]陈一超,曾三友,张好春等.基于遗传神经网络的地震预测研究[J].计算机应用与软件,2008,25(4):77-85.
[4]武安绪,李平安,鲁亚军,等.基于支持向量机的多维地震时间序列建模[J].东北地震研究,2006(4):4.
[5]杨居义,易永红.基于BP神经网络的地震预测研究[J].微电子学与计算机,2008,25(10):11-15.
[6]高海兵,高亮,周驰等.基于粒子群优化的神经网络训练算法研究[J].电子学报,2004,32(9):15-19.
[7]张立,平建军,苏有锦.云南地区综合地震前兆信息量及其短期映震能力分析[J].地震研究,2006,29(10):26-29.
[8]王凤,黄力宇,张宇翔.基于Matlab的BP预测模型在地震前兆预测中的应用研究[J].华北地震科学,2009,27(1):48-52.
[9]王存睿,段晓东.改进的基本粒子群优化算法[J].计算机工程,2004,30(21):55-63.
[10]王晓霞,王涛,谷根代,等.基于改进粒子群优化的神经网络及应用[J].华北电力大学学报,2009,36(5):99-102.
[2]秦长源,张淑亮.AR模型预测太原盆地地震活动的趋势[J].地震研究,1997,20(4):4-9.
[3]陈一超,曾三友,张好春等.基于遗传神经网络的地震预测研究[J].计算机应用与软件,2008,25(4):77-85.
[4]武安绪,李平安,鲁亚军,等.基于支持向量机的多维地震时间序列建模[J].东北地震研究,2006(4):4.
[5]杨居义,易永红.基于BP神经网络的地震预测研究[J].微电子学与计算机,2008,25(10):11-15.
[6]高海兵,高亮,周驰等.基于粒子群优化的神经网络训练算法研究[J].电子学报,2004,32(9):15-19.
[7]张立,平建军,苏有锦.云南地区综合地震前兆信息量及其短期映震能力分析[J].地震研究,2006,29(10):26-29.
[8]王凤,黄力宇,张宇翔.基于Matlab的BP预测模型在地震前兆预测中的应用研究[J].华北地震科学,2009,27(1):48-52.
[9]王存睿,段晓东.改进的基本粒子群优化算法[J].计算机工程,2004,30(21):55-63.
[10]王晓霞,王涛,谷根代,等.基于改进粒子群优化的神经网络及应用[J].华北电力大学学报,2009,36(5):99-102.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心