Napolitano, Francesco (2018) Smart Sensor Monitoring in Machining of Difficult-to-cut Materials. [Tesi di dottorato]


Download (22MB) | Preview
[error in script] [error in script]
Item Type: Tesi di dottorato
Lingua: English
Title: Smart Sensor Monitoring in Machining of Difficult-to-cut Materials
Date: 11 December 2018
Number of Pages: 112
Institution: Università degli Studi di Napoli Federico II
Department: Ingegneria Chimica, dei Materiali e della Produzione Industriale
Dottorato: Ingegneria dei prodotti e dei processi industriali
Ciclo di dottorato: 31
Coordinatore del Corso di dottorato:
Date: 11 December 2018
Number of Pages: 112
Uncontrolled Keywords: sensor monitoring, machining, cognitive systems
Settori scientifico-disciplinari del MIUR: Area 09 - Ingegneria industriale e dell'informazione > ING-IND/16 - Tecnologie e sistemi di lavorazione
Date Deposited: 07 Jan 2019 23:53
Last Modified: 22 Jun 2020 09:23


The research activities presented in this thesis are focused on the development of smart sensor monitoring procedures applied to diverse machining processes with particular reference to the machining of difficult-to-cut materials. This work will describe the whole smart sensor monitoring procedure starting from the configuration of the multiple sensor monitoring system for each specific application and proceeding with the methodologies for sensor signal detection and analysis aimed at the extraction of signal features to feed to intelligent decision-making systems based on artificial neural networks. The final aim is to perform tool condition monitoring in advanced machining processes in terms of tool wear diagnosis and forecast, in the perspective of zero defect manufacturing and green technologies. The work has been addressed within the framework of the national MIUR PON research project CAPRI, acronym for “Carrello per atterraggio con attuazione intelligente” (Landing Gear with Intelligent Actuation), and the research project STEP FAR, acronym for “Sviluppo di materiali e Tecnologie Ecocompatibili, di Processi di Foratura, taglio e di Assemblaggio Robotizzato” (Development of eco-compatible materials and technologies for robotised drilling and assembly processes). Both projects are sponsored by DAC, the Campania Technological Aerospace District, and involve two aerospace industries, Magnaghi Aeronautica S.p.A. and Leonardo S.p.A., respectively. Due to the industrial framework in which the projects were developed and taking advantage of the support from the industrial partners, the project activities have been carried out with the aim to contribute to the scientific research in the field of machining process monitoring as well as to promote the industrial applicability of the results. The thesis was structured in order to illustrate all the methodologies, the experimental tests and the results obtained from the research activities. It begins with an introduction to “Sensor monitoring of machining processes” (Chapter 2) with particular attention to the main sensor monitoring applications and the types of sensors which are employed in machining. The key methods for advanced sensor signal processing, including the implementation of sensor fusion technology, are discussed in details as they represent the basic input for cognitive decision-making systems construction. The chapter finally presents a brief discussion on cloud-based manufacturing which will represent one of the future developments of this research work. Chapters 3 and 4 illustrate the case studies of machining process sensor monitoring investigated in the research work. Within the CAPRI project, the feasibility of the dry turning process of Ti6Al4V alloy (Chapter 3) was studied with particular attention to the optimization of the machining parameters avoiding the use of coolant fluids. Since very rapid tool wear is experienced during dry machining of Titanium alloys, the multiple sensor monitoring system was used in order to develop a methodology based on a smart system for on line tool wear detection in terms of maximum flank wear land. Within the STEP FAR project, the drilling process of carbon fibre reinforced (CFRP) composite materials was studied using diverse experimental set-ups. Regarding the tools, three different types of drill bit were employed, including traditional as well as innovative geometry ones. Concerning the investigated materials, two different types of stack configurations were employed, namely CFRP/CFRP stacks and hybrid Al/CFRP stacks. Consequently, the machining parameters for each experimental campaign were varied, and also the methods for signal analysis were changed to verify the performance of the different methodologies. Finally, for each case different neural network configurations were investigated for cognitive-based decision making. First of all, the applicability of the system was tested in order to perform tool wear diagnosis and forecast. Then, the discussion proceeds with a further aim of the research work, which is the reduction of the number of selected sensor signal features, in order to improve the performance of the cognitive decision-making system, simplify modelling and facilitate the implementation of these methodologies in a cloud manufacturing approach to tool condition monitoring. Sensor fusion methodologies were applied to the extracted and selected sensor signal features in the perspective of feature reduction with the purpose to implement these procedures for big data analytics within the Industry 4.0 framework. In conclusion, the positive impact of the proposed tool condition monitoring methodologies based on multiple sensor signal acquisition and processing is illustrated, with particular reference to the reliable assessment of tool state in order to avoid too early or too late cutting tool substitution that negatively affect machining time and cost.


Downloads per month over past year

Actions (login required)

View Item View Item