Dissertation > Excellent graduate degree dissertation topics show
A Study on Fault Diagnosis Based on Fault-tolerant Neural Network
Author: XuBaiLing
Tutor: LiuQun
School: Harbin Engineering University
Course: Applied Computer Technology
Keywords: Fault Diagnosis Information Fusion Neural Networks Fault-tolerant
CLC: TP183
Type: Master's thesis
Year: 2005
Downloads: 411
Quote: 3
Read: Download Dissertation
Abstract
|
Equipment fault diagnosis technology developed in nearly 40 years to adapt to the engineering needs a multidisciplinary , interdisciplinary inevitable development , and thereby acquire very significant economic benefits . Data fusion as many traditional disciplines and emerging engineering fields combined produce a new frontier disciplines and beyond in the military field , has been widely applied in many fields . A lot of information available equipment fault diagnosis , and only take full advantage of the useful information to diagnose equipment failures in order to improve the accuracy and reliability of the fault diagnosis , fault diagnosis is essentially a multi- information fusion process . In this paper, based on artificial neural network information fusion fault diagnosis , analysis of the list of the neural network fault Jane off characteristics , integrated neural network fault diagnosis model , integrated neural network modeling method established principle and implementation strategy combined diagnosis instance simulation analysis , results show that the integrated neural network information fusion fault diagnosis is an effective method . Equipment fault diagnosis , and how to reduce the rate of misdiagnosis of great significance to improve the reliability of the study . Traditional diagnostic system does not consider the error of the output of the network in the case of the role of various disturbances or failure , the fault tolerance of the system has been largely restricted . Reliability of the neural network becomes more and more important , there is an urgent need to establish the theory and method of neural network analysis and design of high- reliability . This article describes the initial framework of the fault tolerance of neural network theory , fault - tolerant neural networks applied to fault diagnosis and simulation and experimental results show that such a fault tolerant neural network is able to maintain the same performance in the network failure , network failure , also has a good performance . This method greatly improves the fault diagnosis , fault tolerance .
|
Related Dissertations
- Design and Development of Application Programme for Fault Analyzer Based on Wince Platform,TP311.52
- Research on the 6-Dof Fault Tolerant Control of the Vibration Isolation Platform with Eight Actuators,TB535.1
- High Speed Frequency Measurment and Non-Linearity Correction of Frequency Modulated Capacitive Displacement Sensor,TH822
- Tongue Feature Extraction and Research of Fusion Classification,TP391.41
- Multi-Sensor Information Fusion and Its Applications on Wearable Computer,TP202
- Fault Diagnosis Method Based on Support Vector Machine,TP18
- Fault Diagnosis Research on Three-Tank System,TP277
- Research on Anti-periodic Solutions of Delayed Cellular Neural Networks without Assuming Global Lipschitz Conditions,TP183
- Designs and Applications of Fuzzy Synthetic Evaluation Models Based on Parallel Algorithms,TP18
- Research on the Key Technology of Waterborne Transport Security System,U698
- Spectrum Variation of Vegetation in Yanzhou Coal Mine Area and Heavy Metal Stress Characteristic,X173
- Fault Diagnosis on Fuel Control System of a Certain Aero-engine,V263.6
- Remote logging sensor fault diagnosis system,TP212
- Detection and Parameters Estimation of Direct Sequence Spread Spectrum Communication Signals in Non-cooperative Context,TN914.42
- Asymptotic Behavior for Some Kinds of Functional Differential Equations,O175.12
- Research on Tool Wear Monitoring of Milling Cutter Based on Co-integration,TG54
- Researches on the Theories and Algorithms of Earth Orientation Paramters Prediction,P22
- Pb- Al composite electrode materials preparation technology and performance,TB331
- The Degree Assessment of Storm Surge Based on the Land Subsidence and the Storm Surge Disaster Evaluation,P642.26
- Classification of Customer Value Based on Rough Sets - Neural Network,F224
- Network on Chip reliable packet transmission algorithm,TN47
CLC: > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory > Artificial Neural Networks and Computing
© 2012 www.DissertationTopic.Net Mobile
|