Centobelli, Piera (2016) Cognitive Sensor Monitoring of Machining Processes for Zero Defect Manufacturing. [Tesi di dottorato]

[img]
Preview
Text
Centobelli_PhD thesis_UniNa_Tecnologie e Sistemi di Produzione.pdf

Download (5MB) | Preview
[error in script] [error in script]
Item Type: Tesi di dottorato
Lingua: English
Title: Cognitive Sensor Monitoring of Machining Processes for Zero Defect Manufacturing
Creators:
CreatorsEmail
Centobelli, Pierapiera.centobelli@unina.it
Date: 31 March 2016
Number of Pages: 138
Institution: Università degli Studi di Napoli Federico II
Department: Ingegneria Chimica, dei Materiali e della Produzione Industriale
Scuola di dottorato: Ingegneria industriale
Dottorato: Tecnologie e sistemi di produzione
Ciclo di dottorato: 28
Coordinatore del Corso di dottorato:
nomeemail
Carrino, Luigiluigi.carrino@unina.it
Tutor:
nomeemail
Teti, RobertoUNSPECIFIED
Date: 31 March 2016
Number of Pages: 138
Uncontrolled Keywords: sensor monitoring, cognitive paradigm, decision making, neural network, surface roughness evaluation in polishing, drilling of carbon fibre reinforced plastic
Settori scientifico-disciplinari del MIUR: Area 09 - Ingegneria industriale e dell'informazione > ING-IND/16 - Tecnologie e sistemi di lavorazione
Date Deposited: 12 Apr 2016 05:33
Last Modified: 21 Apr 2019 01:00
URI: http://www.fedoa.unina.it/id/eprint/10680

Abstract

The topic of this thesis, focused on cognitive sensor monitoring of machining processes for zero defect manufacturing, has been addressed within the framework of the international research project EC FP7 CP-IP “IFaCOM – Intelligent Fault Correction and self Optimizing Manufacturing systems” (2011-2015; FoF NMP – 285489) and the national MIUR PON Project on “Development of eco-compatible materials and technologies for robotised drilling and assembly processes – STEP FAR” (2014-2016). The vision of the IFaCOM project is to achieve near zero defect level of manufacturing with particular emphasis on the production of high value, large variety and high performance products. This goal is achieved through the development of improved methodologies for monitoring and control of the performance of manufacturing processes with the aim to detect abnormal process conditions leading to defects on the produced parts. The overall aim of the STEP FAR project is the study of issues related to drilling and cutting techniques of advanced lightweight components, such as composite material parts, and their relative assembly, using cooperating anthropomorphic robots. The use of innovative materials and processes developed in this research will lead to a reduction in weight and environmental impact in the construction and maintenance of primary aircraft structures. At least a 5% reduction in weight of the structures is foreseen without increase of costs (a possible rise in the cost of raw materials is compensated with the reduction of process costs). In aeronautical industry the reduction of the weight of the aircraft is becoming an increasingly important aim both for environmental requirements (lower emissions) and contraction of the management costs (lower fuel consumption). Therefore new structural architectures through the use of innovative materials and technologies have been developed. One of the innovative processes analysed in this project is the drilling via machining of carbon fibre reinforced plastic (CFRP) stack-ups. In the framework of these projects, this thesis work is focused on the development of cognitive condition monitoring procedures for zero defect machining processes with reference to two different industrial manufacturing applications. The thesis is organized as follows: Chapter 2 reviews the general concept of sensor monitoring of manufacturing processes and provides a comprehensive survey of sensor technologies, advanced signal processing techniques, sensor fusion approach, and cognitive decision making strategies for process monitoring. In Chapter 3, the Strecon industrial case, as a partner of the IFaCOM project, is discussed and analysed. The STRECON end-user case is focused on improving repeatability and predictability of the surface finish produced by a Robot Automated Polishing (RAP) process. In order to establish a robust method for the detection of the polishing process end-point, i.e. the determination of the right moment for tool and abrasive paste change, STRECON sensor system selection focuses on monitoring the progress of the surface quality during the polishing process by means of variation in VQCs (Vital Quality Characteristics), i.e. roughness and gloss of the polished surface. The output data have been used to train a neural network. The employed NN learning procedure was the leave-k-out method where k cases from the training set are put aside in turn, while the other cases are used for NN training. In Chapter 4, the Alenia Aermacchi industrial case, as coordinator and partner of the STEP FAR project, is discussed and analysed. The Alenia Aermacchi user case is based on the analysis of drilling of stacks made of two overlaid carbon fibre reinforced plastic composite laminates. In this case, a neural network based cognitive paradigm based on a bootstrap procedure has been used for the identification of correlations with tool wear development and product hole quality. Finally, Chapter 5 reports the concluding remarks and future developments of this work.

Actions (login required)

View Item View Item