本文已被：浏览 1721次 下载 3703次
Received:May 31, 2011 Revised:June 26, 2011
Received:May 31, 2011 Revised:June 26, 2011
中文摘要: 为了挖掘隐藏在惯性仪器测试数据背后的信息知识,运用数据挖掘技术,以Clementine12.0 为平台建立模型并实现对惯性仪器故障诊断的过程。提出一种基于两阶段聚类并做改进的BP 算法,与传统BP 算法相比,提高了预测精度和普适能力。
中文关键词: 惯性仪器 两步聚类 k-means 聚类 孤立点检测 BP 神经网络
Abstract:In order to tap the information and knowledge hidden behind the inertial apparatus test data, applying data mining technology and taking Clementine12.0 as platform to establish model and realize the fault diagnosis of inertial instruments. Proposing BP algorithm based on two-stage clustering and its improving, compared with traditional BP algorithm, the prediction accuracy and universal capacity have been improved.
文章编号： 中图分类号： 文献标志码：
基金项目:
引用文本：
李富荣,李笔锋,王玉峰,秦浩.基于改进BP 算法的惯性仪器故障诊断.计算机系统应用,2012,21(2):81-84
LI Fu-Rong,LI Bi-Feng,WANG Yu-Feng,QIN Hao.Fault Diagnosis of Inertial Apparatus Based on Improved BP Algorithm.COMPUTER SYSTEMS APPLICATIONS,2012,21(2):81-84
李富荣,李笔锋,王玉峰,秦浩.基于改进BP 算法的惯性仪器故障诊断.计算机系统应用,2012,21(2):81-84
LI Fu-Rong,LI Bi-Feng,WANG Yu-Feng,QIN Hao.Fault Diagnosis of Inertial Apparatus Based on Improved BP Algorithm.COMPUTER SYSTEMS APPLICATIONS,2012,21(2):81-84
