Dissertation > Excellent graduate degree dissertation topics show

Study on Reliability Test and Evaluation Method of Five-axis Machining Center

Author: YuNaiHui
Tutor: ZhangZhiXiong
School: National University of Defense Science and Technology
Course: Mechanical Engineering
Keywords: Five-axis machining center Reliability test Small sample Weibull distribution Bayesian evaluation
CLC: TG659
Type: Master's thesis
Year: 2011
Downloads: 199
Quote: 0
Read: Download Dissertation

Abstract

The research is studied based on the program“Reliability Design and Performance Test technology of Five-axis Machining Center”, supported by the National Science & Technology Major Project. It focused on the reliability and evaluation technology of five-axis machining center, aiming at improving the mean-time-between-failures(MTBF)of CNC machine tools up to 900h. After a systematic and complete analysis on this kind of products, the method of reliability test and evaluation of five-axis machining center was obtained under the small sample condition. Also, this method can provide other types of CNC machine tools with theories and methodological guidance for reliability study. The thesis made the research on the five-axis machining center reliability test and evaluation, organized as preparation, test and evaluation. The main work done in this thesis is listed as follows:1) Through an extensive investigation, integral aluminum impeller was chose as five-axis machining center’s typical processing object. On this basis, the studies of typical operating conditions analysis and work stress collecting technology were conducted. The impellers were processed by five typical processing parameters. Work stress was collected by a six-component force sensor. By the way of large-scale statistics, process parameters and reliability test’s processing cycles which widely represenedt the five-axis machining center’s processing object were determined. This work could prepare for the following reliability test.2) A set of methods on five-axis machining center’s reliability test under the conditions of small sample was proposed. The methods can give the machining center’s performance of all aspects a comprehensive check and fully expose its failures. In order to conduct reliability test, six kinds of failure criterion of relevance and eight kinds of failure criterion of irrelevance were proposed and they could be the basis of assess the machining center’s reliability index. Besides, in order to determine the fault more accurately and improve the reliabity test’s efficiency, a reliability test system was established. The system mainly included vibration test, noise test, temperature test and accuracy test. They could be used to monitor the machining center’s deviant change.3) A failure data acquisition system was established. Then choosing six kinds of mature products as tracking objects, we spent almost two years gathering and accumulating mass of field test samples.On that basis, two kinds of reliability evaluation methods based on weibull distribution were researched. We first gathered lots of complete failure data from five different types of mass-produced machining centers and made a comparative analysis of their lifespan distribution before and after improvement. Besides, we chose 10 model-6 machining centers coming from a same batch as the field test’s tracking objects and carried out a time censored test. Then, we assessed its test results using maximum likelihood method.4)A study on Bayesian evaluation of machining centers was conducted under the conditions of small sample and high reliability. Using fault information from five mature products, we could choose MTBF as random variable and obtain MTBF discrete distribution of each sample by conducting Bootstrap Sampling. Then, we employed a fusion method based on Kullbank information to get the fusion prior distribution and its weight facts could be determined by the credibility of prior information. In this way, the succession and difference could be reflected in the fusion prior distribution. At last, we combined the fusion prior distribution and test sample according to Bayesian theory to get the posterior distribution of MTBF and reliability inference could be conducted based on it.

Related Dissertations

CLC: > Industrial Technology > Metallurgy and Metal Craft > Metal cutting and machine tools > Program control machine tools, CNC machine tools and machining
© 2012 www.DissertationTopic.Net Mobile