Dissertation > Excellent graduate degree dissertation topics show

Studying the Credibility of Micro-Cantilever Beam Based on Neural Network

Author: XuZuo
Tutor: BaiYanPing
School: University of North
Course: Applied Mathematics
Keywords: BP neural network Flexible BP arithmetic Micro-electro-mechanical system Micro-cantilever beam Wavelet multi-scale analysis Failure probability Reliability Weilbull distribution
CLC: TP211.4
Type: Master's thesis
Year: 2011
Downloads: 59
Quote: 0
Read: Download Dissertation

Abstract

MEMS is a microsystem that is comprised of many electronc components.As a new interdiscipline subject.Many newest results of science and technology are used in this subject. And it can combine information,sensitive and actuators.The way how people see and control the external world.In recent years,it is interested by many scholars and companies.It is rapidly developed with the craft improve and perfected and more and more wide application.As a result,the reliability of MEMS must be improved.So reliability research is very important.BP Network is a multilayered feedforward network.It traing network by back-propagation arithmetic.It includes the best prime part in the neural network.Because its structure is simple and easy to improve,it is applied in a wide range of areas.But converqence speed is very slow and it is easy to lost in local minimum in BP network,BP network is made better in this paper.We creat the BP model which studying the relationship of load and deflection in the environment of matlab7.0.Then,we simulate and predict the relationship of load and deflection of different parameters micro-cantilever beam by wavelet multi-scale arithmetic analyzing the experimental data.The experimental results show that the deflection increases with increment of the micro-cantilever beam length,decrease with increment of width and decrease with increment of number of beam.When deflections of specimena is smaller,the relationship of load and deflection is linear.The results is agreeable with the practice.At the same time,we can predicted the relationship of load and deflection of micro-cantilever beam apply BP neural network,and the average effect of prediction is more than 0.997.The experimental results show that the BP neural network is effective in studying the relationship of load and deflection of micro-cantilever beam.As we known ,the micro-cantilever beam is more easy to be broken whit increase in the deflection ,when fixed the load .So, this paper show that the reliability decrease with increment of the micro-cantilever beam length, increases with increment of width,and increases with increment of number of beam.It is very useful for the future design micro-cantilever beam.At last. By applying mathematical model mehted ,this paper investigates the reliability of micro-cantilever beam which are the cases of axial tensile stress or vertical load stress .And acquiring the reliability predictable model of micro-cantilever beam for static load.The model show that failure probability increases with increment of the micro-cantilever beam length, decrease with increment of width and decrease with increment of number of beam.In other words ,the reliability decrease with increment of the micro-cantilever beam length, increases with increment of width,and increases with increment of number of beam. Which is agreeable with the previous results.

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Automation components,parts > General automation components,components > Mechanical and electrical components , parts
© 2012 www.DissertationTopic.Net Mobile