| 基于遗传神经网络的地震预测研究 |
| |
| 引用本文： | 陈一超,曾三友,张好春,全芙蓉.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):135-137. |
| |
| 作者姓名： | 陈一超 曾三友 张好春 全芙蓉 |
| |
| 作者单位： | 1. 中国地质大学计算机学院,湖北,武汉,430074
2. 武汉理工大学理学院,湖北,武汉,430070 |
| |
| 摘 要： | 分析遗传算法(GA)及BP神经网络的结构特性,提出利用具有全局搜索能力的遗传算法来弥补BP网络的不足,克服BP(Error Back Propagation)算法收敛速度慢,易陷入局部极小点的缺点,优化神经网络的连接权值和阈值.针对地震预测中,震级预测的困难性等问题,将具有全局搜索能力的遗传算法和具有深度搜索能力的BP算法相结合实现地震震级预测建模.通过实验比较得到了较好的预测结果,该模型是可行、有效的.
|
| 关 键 词： | 神经网络 遗传算法 震级预测 |
| 修稿时间： | 2006年6月26日 |
RESEARCH ON EARTHQUAKE PREDICTION BASED ON GENETIC NEURAL NETWORKS |
| |
| Chen Yichao,Zeng Sanyou,Zhang Haochun,Quan Furong.RESEARCH ON EARTHQUAKE PREDICTION BASED ON GENETIC NEURAL NETWORKS[J].Computer Applications and Software,2008,25(4):135-137. |
| |
| Authors: | Chen Yichao Zeng Sanyou Zhang Haochun Quan Furong |
| |
| Affiliation: | Chen Yichao1 Zeng Sanyou1 Zhang Haochun2 Quan Furong11(Department of Computer Science,China University of Geosciences,Wuhan 430074,Hubei,China)2(School of Natural Sciences,Wuhan University of Technology,Wuhan 430070,China) |
| |
| Abstract: | Theoretical basis and integration methods of genetic algorithm(GA)and BP neural network(NN)are analyzed.In order to accelerate the convergence speed of BP networks and escape local minimum,a genetic algorithm which can carry through global search is introduced to optimize the weights and the threshold.Considering that earthquake magnitude prediction is difficult,a model for earthquake magnitude prediction is presented based on the global search of GA and the in-depth search of NN.The experimental results of... |
| |
| Keywords: | Neural networks Genetic algorithm Earthquake magnitude prediction |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
