| 基于PCA—改进BP算法的软测量技术 |
| |
| 引用本文： | 臧春华,郭小萍,王秀丽.基于PCA—改进BP算法的软测量技术[J].仪表技术与传感器,2001(2):26-29. |
| |
| 作者姓名： | 臧春华 郭小萍 王秀丽 |
| |
| 作者单位： | 1. 沈阳化工学院自动化系统工程中心沈阳市 110021
2. 辽宁省石油化工规划设计院沈阳市 110015 |
| |
| 摘 要： | 文中针对基本BP网络建立软测量模型时所存在的一些问题，例如：输入变量之间可能存在的线性相关等冗余性、基本BP算法收敛速度较慢而且易于限于局部最优等，本文尝试将主元分析与变尺度改进BP算法相结合，以提高软测量模型的训练速度和外推能力，为软测量技术的在线应用提供更大的方便。实验结果表明PCA方法与变尺度的改进BP神经元网络相结合的软测量建模方法在训练速度和外推能力方面有较大的改善。
|
| 关 键 词： | 神经元网络 软测量技术 主元分析 变尺度 |
| 修稿时间： | 1999年11月23 |
Soft - Sensing Teehnique Based on PCA - Improved BP Neural Network |
| |
| Zang Chunhua Guo Xiaoping Automatic System Engineering Center,Shenyang Institute of Chemical Technology,Shenyang.Soft - Sensing Teehnique Based on PCA - Improved BP Neural Network[J].Instrument Technique and Sensor,2001(2):26-29. |
| |
| Authors: | Zang Chunhua Guo Xiaoping Automatic System Engineering Center Shenyang Institute of Chemical Technology Shenyang |
| |
| Affiliation: | Zang Chunhua Guo Xiaoping Automatic System Engineering Center,Shenyang Institute of Chemical Technology,Shenyang 110021 |
| |
| Abstract: | In this paper,PCA(principal component analysis) method is incorporated into BP network improved,aiming at several questions of basal BP algorithm: training speed of network is slower and generalization of network is wealcer. This method not only can solve the linear conelation of inputs,but also can simplify the network structure and improve the training speed and generalization. This brings great convenience to on-line application. Experimental results demonstrate that this technique improvement is very valid. |
| |
| Keywords: | Neural Metwork Soft - sensing Technique PCA(principal component analysis) Varying - scale |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
