پديد آورنده :
رسولي، محسن
عنوان :
كاربرد شبكه عصبي مصنوعي در پايش پروفايل هاي خطي تعميم يافته
مقطع تحصيلي :
كارشناسي ارشد
محل تحصيل :
اصفهان: دانشگاه صنعتي اصفهان، دانشكده صنايع و سيستم ها
صفحه شمار :
يازده،98ص.: مصور،جدول،نمودار
يادداشت :
ص.ع.به فارسي و انگليسي
استاد راهنما :
غلامعلي رئيسي اردلي
استاد مشاور :
اميرحسين اميري
توصيفگر ها :
كنترل فرآيند آماري , پرسپترون چند لايه
تاريخ نمايه سازي :
21/7/92
استاد داور :
علي زينل همداني، رضا حجازي
دانشكده :
مهندسي صنايع و سيستم ها
چكيده فارسي :
به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي
چكيده انگليسي :
Application of Neural Network Artificial in Monitoring Generalized Linear Profiles Mohsen Rasouli Mohsen rasouli@ec iut ac ir Date of Submission 2013 01 05 Department of industrial Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Farsi Supervisor Gholam Ali Raissi Ardali raissi@cc iut ac ir Abstract Statistical process control typically involves monitoring of control charts to detect anormal patterns and perform corrective operations upon detection of an anormality in the system Although control charts could be utilized to determine if a process is in contol or out of control they may not be useful for ordinary and less experienced users due to their complexity and difficulty of interpretation Thus an automated system which could monitor the process detect out of control situations and recommend corrective operations is highly desired The quality of a product or process in some statistical process cotrol methods is characterized by a relationship between a response variable and one or more explanatory variables called profile Profile monitoring is used to understand and check the stability of this relationship over time While a profile could be as simple as a linear regression model for exapmle in calibration a more complex model may be needed for some applications In this project we investigate the application of Artificial Neural Networks for monitoring of generalized linear regression profiles The aim is to monitor a profile and detect out of control situations with minimal delay A feed forward multilayer perceptron artificial neural network is applied to monitor binary and possion profiles The performance of the proposed method is evaluated using simulation and numerical examples The results are compared to those obtained using T 2 control charts It is shown that the proposed method could detect the shifts in the parameters of binary profiles when the process is out of control Keywords Statistical process control generalized linear regression model binary profile artificial neural network average run length PDF created with pdfFactory trial version www pdffactory com
استاد راهنما :
غلامعلي رئيسي اردلي
استاد مشاور :
اميرحسين اميري
استاد داور :
علي زينل همداني، رضا حجازي
