Bilgisayar destekli kalite kontrol sistemleri
Computer aided quality control systems
- Tez No: 46357
- Danışmanlar: PROF.DR. MURAT DİNÇMEN
- Tez Türü: Yüksek Lisans
- Konular: Endüstri ve Endüstri Mühendisliği, Industrial and Industrial Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1995
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 97
Özet
ÖZET Değişen ve koşullan sürekli olarak zorlaşan uluslararası pazarlarda rekabet şansını arttırmak isteyen firmalar giderek artan miktarlarda kalite güvence bölümlerine yatırımda bulunmaktadırlar. Kalite güvencesinin en ucuz ve en etkili olarak sağlamasındaki en etkili araçlardan biri de Bilgisayar Destekli Kalite Kontrol (CAQ) sistemleridir. Teknolojinin son yıllarda gösterdiği inanılmaz gelişme ve özellikle bilgisayar teknolojisinin vardığı nokta bilgisayarın hayatımızın çeşitli aşamalarında kullanılması sonucunu doğurmuştur. Dolayısıyla işletmelerin üretim problemlerini Çözmeye yönelik çeşitli uygulamalarda bilgisayar kullanımı gerçekleştirilmiştir. Üretimde bilgisayar desteğinin sağlandığı son alan ise kalite kontrol alanıdır. Bu çalışmada, bilgisayarın üretim alanında kullanılmasını sağlayan Bilgisayar Destekli Kalite Kontrol (CAQ) sistemleri tanıtılmaya çalışılmış, bu tür sistemlerin değerlendirmesine yönelik kriterler tespit edilmiş ve bir değerlendirme yöntemi oluşturulmuştur. Birinci bölümde üretim sektöründeki genel gelişmelerden bahsedilmiş, kalitenin gelişen önemi ve bilgisayarın üretimde kullanılmasına değinilmiştir. İkinci bölümde bilgisayarın kalite kontrol alanında kullanımından bahsedilmiş ve bilgisayar desteğinin sağlandığı temel istatistiksel kalite kontrol yöntemleri anlatılmıştır. Üçüncü bölümde CAQ sistemleri ayrıntılı olarak incelemiş ve elektrik motorlarının kontrolünde kullanılan bir CAQ sistemi tanıtılmıştır. Dördüncü bölümde CAQ sistemlerinin değerlendirilmesine yönelik derlenen performans kriterleri verilmiş, bir sistem değerlendirme yöntemi sunulmuştur. Beşinci bölümde ise dünya üzerinde bilgisayar destekli kalite kontrol sistemlerine eğilim anlatılmıştır. viii
Özet (Çeviri)
SUMMARY Computer Aided Quality Control Systems Production concept has been always actual since the existence of human beings on the earth. During the centennials technology has shown a rapid improvement and helped human who was trying to make his environment more comfortable by executing necessary production. Today we, as human beings, have reached to manless production techniques with the help of improved technology. In our rapid changing world, hard competition in the world market has forced the companies to check their company organizations and to follow and adept the new technologies in field of production to their organizations and production facilities. Demands of several world markets from production sector can be listed as follows: - High and certified quality - On time supply - Continuity in production - Reasonable prices As can be seen above high and certified quality is a major factor to be competitive and successful in international market. Due to that fact tendency in production sector is in direction of total quality management and ISO 9000 certificates. To ensure the requirements of ISO 9000 standards all companies are highly concentrated on automating their production process by using new production concepts, namely Computer Integrated Manufacturing (CIM) techniques such as CAD/CAM, CAE etc.. Actually computer is used since for a long time to execute statistical quality control methods like control diagrams etc.. As the technology get improved during the years with an incredible acceleration parallel to this development computer systems and their main elements, i.e. both hardware and software has increased their capacity continuously and reached to a very high level. In fact, however, this level is not a final frontier for computer technology. As a result of all these events computers are started to be used widely in quality control area where large amount of data should be processed with complex calculation methods. The last stage in computerized quality control reached is Computer Aided Quality Control (CAQ) system technology. Classical function of quality control is to separate bad from good and to eliminate that the bad products are reached to the end consumer. As mentioned above IXduring the last ten years computers are widely used in production and this has resulted that quality control concept has left its place to quality assurance concept. Since powerful computers enable easy save of process data continuity of high quality in production is ensured. To reach this high quality assurance 100 % product control is necessary. On the other hand, however, zero fault production strategy and batch production capacities of new production lines ( for example 10,000 motors per day are installed on a modern motor assembly line) make the realization of 100 % control of production very difficult. Otherwise the lost of the efficiency in last control of production will cause that the producers will confront with image lost and related to this profit lost. Computer aided quality control systems are acting in general like a kind of expert system where the system has to make diagnoses of tested product depending on its data base and knowledge likewise an expert system. They are using different classification methods to enable most precise diagnostic of the tested product. In general these methods are divided into two groups: - Numerical Classification Methods - Neural Networks as Classificator Numerical Classification methods are, as can be understood from the title easily, methods based on mathematical equations. Some of them are easy to calculate where some others contain complex equations. The ease of the calculation is a very important criterion for the speed of the software of system. The easy is the calculation the fast is the diagnoses. The other method group contains neural networks as classificator. Neural networks have the ability of learning the relation between input and output of a system. The network can be trained to mimic this relationship to create a model of the system. And in this way neural network can be used as a classificator. Each method group is briefly described in this study. Conventional quality assurance systems have a loop build from a testing system, an expert for diagnoses, an SPC division to detect the production process and a decision group to examine the SPC reports and to decide on the production status. A total computerized quality assurance system, however, contains only the CAQ system and a production control computer within its control loop. To eliminate of all the elements of a conventional quality assurance loop computer aided quality control systems contain several modules. The major modules of CAQ systems are as listed below: - Sensors - Data processor- Testing environment - Output module - Statistical process control module Each of these modules is designed to fit the requirements of a fully automated quality assurance process. The modules and the principe of CAQ systems are explained and studied on an example. Example CAQ system is used in electric motor production to execute the final tests of electric motors. This motor control system is developed and produced by Carl Schenck AG./Germany. These systems have first entered the market in 1989. Since that time the number of users is rapidly grown. The system examined in this study is settled in Topkapi plant of Türk Elektrik Endüstrisi AŞ., the biggest motor producer of Turkey and one of the biggest in Europe. In Topkapi plant the company has implant four this kind of testing equipment. Three of them work as the last control station on an fully automated production line where the fourth one is used in laboratory. The system acts like a black box in this manner and performs a certain diagnoses for electric motors. Conventional test systems for electric motors are morely based on mechanical components. The main goal by a motor test is to obtain the characteristic curve of the motor and determine several torque values from this curve. Classic systems are using mechanical equipments to measure torque, number of turns per minute and as load. The extensive usage of mechanical equipments in such a systems cause accuracy problems. To overcome any of this kind of problems most components should be checked and calibrated periodically. The new concept in electric motor test equipment is model aided test system. This new generation of test systems are based on a mathematical model of the motor. The model consist of mathematical equations which fully describes the motor. At the beginning phase both motor and model are feed with the same input signal. The response of the motor and model to this input are compared and model is corrected until the response of them are equal enough. Next phase is teaching mode. Motors are tested by using this model and classified by an expert to several classes such as OK., Noisy, Bearing Fault, etc.. The more is the examples taught to the system the more accurate is the diagnoses of the system. Third phase is testing phase. The control system which is built up in the above mentioned manner is placed on the production line as the last station. Already assembled motors coming from the previous station of the line enter in the testing cabin of the system. Necessary measurements are executed in this cabin and the tested motor is classified accordingly as OK. or faulty. If the tested motor is XIdiagnosed as faulty the fault type is also displayed on the screen. Afterwards tested motors are sent to their next destinations, namely if they are OK. to the packaging otherwise to correcting station. The measuring process of described CAQ system consists of four major parts: Parameter Identification PI. Vibroacoustic Diagnoses VAD Technical Fault Diagnoses TFD Threshold Value Diagnoses SDS Necessary data acquisition from the tested motor to the system is realized via parameter identification. The data is acquired from the motor by using a mathematical model. The model of the motor gives the parameters which are necessary for motor diagnoses. Vibroacoustic diagnoses part gets the vibration and acoustic signals of the motor. For that purpose an accelerometer and a special microphone are used. All signals and parameters received from the motor via these two methods are processed in the diagnoses parts of the system. In Technical Fault Diagnoses, one of two responsible parts for the diagnoses, a numeric classification method is used to arrange the tested motor to a certain class. Used classification method is simple and accurate enough and is similar to fuzzy logic. During the teaching phase in data base of the system, clouds from n-dimensional specification vectors are built. Each of these cloud is a motor identity class. During the testing phase first the specifications vector of the motor is built from the measurement data. Then the vectorial distance of specification vector to the center of the clouds (or clusters). Tested motor is then classified in the closest class. If all distances are unacceptably large then this motor belongs to a new, undefined fault class. If this occurs often for a certain fault, after an expert control, the class can be named accordingly. The Threshold Diagnoses, on the other hand, is not as much complicated as Technical Fault Diagnoses method. The user sets some threshold (or limit) values. After building up the specifications vector related components are compared with the threshold values. If one of these components exceeds the threshold values than the tested motor is labeled as faulty. Example CAQ system has a powerful statistical module to execute major statistical process control (SPC) techniques. All SPC charts like X and R charts, s and c charts etc. are both visually and printed available. Moreover the system is able to send process control signals to a production control computer net in order to stop the production in case of an“out of control”process. The production XIIcontrol computer will be then able to decide on the status of production depending on the data coming from CAQ system. The hardware of the system is considerably more plain than conventional motor testing systems. Since the system use a model for necessary data acquisition from the motor the system does not contain complex mechanical components. The result is less maintenance and calibration and nigh accuracy. Due to the fact that computer quality control systems are still in developing phase the end users have almost no experience in this field. Another fact is that the companies should invest huge amounts to implant a CAQ system to their production processes. The lack of experience makes it difficult to choose the right CAQ system for the production process. Considering all this problems to help the end users of this kind of systems some performance criteria are selected to evaluate CAQ systems. These criteria are divided in five major groups: - Sensors - Measurement Techniques/Software - Usage - Service - General Added to these criteria an evaluation method based on these criteria is also developed. In spite of the actual status of computer aided quality control systems the tendency towards CAQ' s is relatively high. High competition in market, strict quality standards and high production rates make it very difficult to obtain a high level of quality. End users define computer aided quality control systems as key factor to reach their quality targets. Mil
Benzer Tezler
- Development of a computer aided quality control software for the integration of CMM in meticum
Metucım bilgisayar tümleşik üretim sisteminde koordinat ölçme tezgahının entegrasyonu için bilgisayar destekli kalite kontrol yazılım geliştirilmesi
YUSUF BAŞIBÜYÜK
Yüksek Lisans
İngilizce
1999
Makine MühendisliğiOrta Doğu Teknik ÜniversitesiMakine Mühendisliği Ana Bilim Dalı
PROF. DR. ÖMER ANLAĞAN
- MPS modüler üretim sisteminin bilgisayar destekli gerçek zamanlı kontrolü ve teknik eğitime uygulanması
Computer aided real time control of MPS modular production system and technical education application
SEZAİ TAŞKIN
Doktora
Türkçe
2007
Elektrik ve Elektronik MühendisliğiMarmara ÜniversitesiElektrik Eğitimi Ana Bilim Dalı
DOÇ. DR. KORAY TUNÇALP
YRD. DOÇ. DR. HASAN ERDAL
- Nazofarenks, prostat, akciğer kanseri tedavilerinde volumetrik ark tekniğiyle yapılan planlamaların dozimetrik kalite kontrollerinin karşılaştırılması
Comparison of dozimetric quality controls of volumetric arc technique plans in nazopharense, prostate, lung cancer treatments
CANER ARSLAN
Yüksek Lisans
Türkçe
2016
OnkolojiAcıbadem ÜniversitesiRadyasyon Onkolojisi Ana Bilim Dalı
PROF. DR. KAZIM MERİÇ ŞENGÖZ
- Endüstriyel üretim hatlarında yapay zekâ destekli kalite kontrol sisteminin geliştirilmesi
Development of artificial intelligence-supported quality control system in industrial production lines
ELİF AYDAN BİKE
Yüksek Lisans
Türkçe
2024
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolBursa Teknik ÜniversitesiAkıllı Sistemler Mühendisliği Ana Bilim Dalı
DOÇ. DR. GÖKAY BAYRAK
- Modern imalat süreçlerinde bilgisayar destekli kalite yönetimi
Computer aided quality management at modern manufacturing processes
EMRE KIRIMTAY
Yüksek Lisans
Türkçe
2010
Makine MühendisliğiYıldız Teknik ÜniversitesiMakine Mühendisliği Ana Bilim Dalı
PROF. DR. M. NUMAN DURAKBAŞA