|
|||||
|
|
| 粒子群神经网络用于电磁兼容预测 | |
| 引用本文： | 陈书文,张煜东,张斌,王水花.粒子群神经网络用于电磁兼容预测[J].无线电工程,2010,40(3):39-41. |
| 作者姓名： | 陈书文 张煜东 张斌 王水花 |
| 作者单位： | 1. 江苏省辐射环境保护咨询中心,江苏,南京,210096 2. 东南大学,信息科学与工程学院,江苏,南京,210096 3. 江苏省辐射监测管理站,江苏,南京,210096 |
| 基金项目： | 东南大学优秀博士学位论文基金，河海大学青年科技基金 |
| 摘 要： | 为了更好地对电磁兼容进行预测,提出采用人工神经网络的方法。传统的BP神经网络易于陷入局部最优,因此采用粒子群算法对网络权值进行优化。以平行线间电磁耦合干扰为具体算例,证明本算法的预测结果的均方误差仅有10-4数量级。因此,使用PSO优化网络权值的方法有效,且神经网络模型能准确预测电磁兼容。 |
| 关 键 词： | 电磁兼容 预测 神经网络 |
PSO-based Neural Network Used for Prediction of EMC |
|
| CHEN Shu-wen,ZHANG Yu-dong,ZHANG Bin,WANG Shui-hua.PSO-based Neural Network Used for Prediction of EMC[J].Radio Engineering of China,2010,40(3):39-41. | |
| Authors: | CHEN Shu-wen ZHANG Yu-dong ZHANG Bin WANG Shui-hua |
| Affiliation: | CHEN Shu-wen1,ZHANG Yu-dong2,ZHANG Bin3,WANG Shui-hua2(1.Radiation Environmental Protection Consultation Center of Jiangsu Province,Nanjing Jiangsu 210096,China,2.School of Information Science & Engineering,Southeast University,3.Radiation Monitoring Station of Jiangsu Province,China) |
| Abstract: | In order to predict the electromagnetic compatibility more effectively, an improved method based on artificial neural network was proposed. Due to the fact that BP neural network was inclined to be trapped in local extrama, a novel network--particle swarm optimization based neural network--was proposed in this paper to solve the above shortcoming. The specific example on electromagnetie coupling interference between two parallel wires demonstrates the median square error of the prediction is more or less only 10^-4 order of magnitude. Thus, this PSO-based neural network is effective, and our model can predict electromagnetic compatibility accurately. |
| Keywords: | electromagnetic compatibility prediction neural network |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |
