Font Size: a A A

Research And Application Of Reliability Technique In CNC Research And Development

Posted on:2006-01-14Degree:MasterType:Thesis
Country:ChinaCandidate:X F LiuFull Text:PDF
GTID:2121360155953294Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
1.Purpose Based on the subject of Hi-Tech Research and Development Program of China --The research and development on the key reliability-technology of Computer Numerical Control(CNC) equipment and system , put forward the scientific research of this thesis . On the foundation of the domestic and international popular Computer Numerical Control system of extensive research, reliability of the domestic popular CNC system is researched . Under current situation of the development of CNC system , we study CNC system reliability prediction , FMECA analysis and the reliability assurance measure in develop course. Intellectual FMECA of the CNC system research has been tried. It establish the foundation for the systematic industrialization of popular CNC system , offer the basis for systematic improvement, develop for follow-up high-grade , the whole function type CNC system reliability study. 2. CNC system reliability prediction Reliability prediction is for estimating products reliability which are under certain work condition. It presumes reliability of system by reliability of system's components, parts and sub-system. Because a certain non-linear mapping relationship exists between reliability index of sub-system and reliability index of whole CNC system, it happens that the neural network has inherent non-linear characteristics , and the characteristics that can study by training. On the basis of these characteristics we attempt to carry on prediction and assessment through unique prediction and learning function of neural network. The main idea of evaluation and prediction of reliability level for NN-based CNC system is : according to structure of CNC system, divides it into hardware part and software part, software part is divided into relatively independent modules according to its function, hardware part is divided into relatively independent units according to its function . Then collects reliability data and operating mode data , influence factors etc. according to relevant units (modules ). (Aiming at different series of products, firstly according to the expertise evaluates equipment comparability). Then after abstracting the characteristics, writes down them as x1, x2, …xn and regards them as information input of neural network (namely input neuron). The index y of CNC system is regarded as output of network, BP network of three layers of structure is set up. Utilizing historical fault data of CNC system , after statistical analyzing and processing, regard them as sample of training and trains he network. Finally using trained network, can carry on reliability assessment to homologous series of CNC systems, and carry on prediction of reliability level to improving serial CNC system through introducing similar factor. Through the argumentation above, the assessment and prediction of reliability based on neural network model is feasible. 3. CNC system FMECA We have discussed the analysis procedure of FMECA of CNC system and taken some type CNC system as an example and carried on FMECA analysis. We have known the bottleneck of the CNC system .By using RNP method and draw the main risk position that influences the system. We have set up intellectual FMECA based on neural network in tentative finally. Establish the foundation of follow-up computer reliazer. 4. CNC system reliability assurance measure in develop course The final purpose of reliability prediction and FMECA is to improve reliability. So we must improve the measures guaranteed of reliability ultimately. As to enterprises, they usually adopt three kinds of ways in order to improve reliability: (1) Choosing the high-quality raw materials; (2) Improvement of designing technique; (3) Improvement on the product development procedure. This chapter puts forward some measures and thinkings on these main respects of two of three. We describe especially the code requirement to the supplier in the course of purchasing parts and auxiliary equipment, purchasing parts. Then we summarize designing technique of reliability and put forward the main point that should be pay attention in the course of developing. 5. Conclusion Through the research of this thesis, develop the working content of...
Keywords/Search Tags:CNC system, Reliability Prediction, BP Neural network, FMECA, Reliability Assurance Measure
PDF Full Text Request
Related items