Tesi sul tema "Industrial applications"

Ce projet s’est focalisé sur l’ajout de nouvelles propriétés à des papiers grâce à l’utilisation de nanocelluloses fonctionnelles. Ces nanocelluloses sont des nanoparticules extraites du bois qui peuvent être divisées en deux catégories : les nanofibrilles de cellulose (CNFs) et les nanocristaux de cellulose (CNCs). Ce travail s’est essentiellement penché sur l’utilisation des CNFs. Leur réactivité chimique a été utilisée afin de les fonctionnaliser avec des organotrialkoxysilanes. C’est aussi leur fort enchevêtrement ainsi que la grande viscosité de ces CNFs en suspension qui ont été utilisés afin de synthétiser des petites particules de silsesquioxane pour rendre le matériau final antimicrobien et (super)hydrophobe. Les connaissances obtenues à travers l’étude sur des films modèle de CNFs ont ensuite été appliquées au couchage du papier. Ces CNFs fonctionnelles ont donc été évaluées pour le développement d’un papier possédant une surface antimicrobienne, anti-adhérente, barrière aux graisses ou superhydrophobe
The aim of this work is to implement new properties to a paper based material via the use of functional nanocelluloses. Nanocelluloses are nanoparticles extracted from wood and distinguished in two categories: Cellulose Nanofibrils (CNFs) and Cellulose Nanocrystals (CNCs). This work has only been carried out with CNFs. The chemical reactivity of CNFs was used to functionalize them with organotrialkoxysilanes. The entangled network and highly viscous suspension of CNFs was also used to synthesize silsesquioxane particles with limited size to impart (super)hydrophobic and antimicrobial properties. Knowledge obtained through the study of model CNFs films was then applied to paper based material coating. The functional CNFs were evaluated for its use in an antimicrobial, anti-adherent, greaseproof or superhydrophobic paper surface

In the manufacturing industry, the request of complex products and the decreasing of production time have led to more and more sophisticated CNC-controlled multi-axis machines. Often their setup process is affected by different mistakes caused by the persons responsible that lead to collisions inside the working area. Those collisions often lead to damage the tool and the work piece. The thesis deals with this problem, providing new insights for a fast and robust collision detection. Imagining to start from scratch, through a dynamic analysis of the impact in a mechanical transmission, we reached to identify the sensors which provide the optimal trade-off between the quality of impact information measured, feasibility and costs. Then, we propose two new collision detection algorithms able to identify the unwanted event as fast as possible, with the goal to reduce the impact force and containing the damage. Furthermore, their performance are compared with the most successful algorithm found in literature on two different mechanical systems: a heavy automatic access gate and the laboratory’s robotic arm.

Since the discovery of vulcanization, rubbers have invaded our life and nowadays occupie a significant place in the industrial world. In fact, in most applications there are no alternative materials to them. Despite what it could be thought, a rubbery object is a very complex system. In fact, it is generally based on only one or more rubbers and on several other additives, such as reinforcing fillers, plasticizers, antidegradants, vulcanizing agents. Different production phases are required to realized a rubbery product. The first is the rubber compounding with the additives, which occurs at a defined temperature for a fixed time. The dispersion degree achieved by the fillers in the elastomeric matrix and consequently the final product properties are strictly dependent on this first step of production. After the mixing, the forming operations follow: the compound will retain the shape imposed. Finally, vulcanization process provides the material its ‘elastic’ recovery behaviour. The final properties of a rubber product depend mainly on the type of rubber chosen, however they can be further manipulated by varying the additives used and their concentrations, and the processing steps. The special character of rubber, being a multicomponent system, and the complexity of the production phases delayed the study and the development of rubber nanocomposites, in respect to the polymeric ones. However, in the ten past years, the reports published that deal with rubber nanocomposites have been raise. The ongoing R&D interest is mostly due to the significant physico-mechanical properties improvement which is observed when the nanoclays are added to a rubber matrix. This enhancement depends on the nanometric-scale dispersion that the nanoclays can achieve in the compound; contrary to the conventional fillers, such as carbon black and silica, which carry out a micrometric-scale dispersion. Nowadays, the nanoclays are the most studied nano-sized fillers because there are easily available in nature and cheap. Several research works have demonstrated that the addition of even low amounts of layered nanofillers (< 10 wt.%) improves the mechanical properties, decreases gas permeability and swelling in solvents, increases thermal endurance and flame resistance. This PhD activity was financed by the company “IVG Colbachini” placed in Cervarese Santa Croce, Padova. The company, for over 40 years, has been involved in the production of industrial rubber hoses to convey powders, granular, liquid or gas materials. The “IVG Colbachini” products are used in different sectors, such as food, chemical, agricultural, construction, rail, naval and steel industries. The thesis work was aimed at the study and the optimization of rubber compounds produced in “IVG Colbachini”. This thesis consists of 6 chapters and subsequently the main topics dealt with in each chapter will be concisely summarized. Chapter 1 highlights the significant differences between a conventional composite and a nanocomposite, giving also a classification of this latter class of materials. In addition, it explains which filler features are the most important to obtain a nanocomposite and how each of these can influence the final product properties. In Chapter 2 nanoclays and rubber nanocomposites loaded with layered fillers are introduced. Particularly, the chemical structure of these latter and the organo-modification importance are described. In addition there is a summary of the rubber/clay nanocomposite synthesis methods and their characteristic properties present in literature, such as mechanical performance, barrier effect and flame resistance. Chapter 3 explains step by step the art of rubber compounding. In particular, the concept of recipe and how it is indicated are introduced. Types, features and functions of recipe diverse ingredients are specified. In addition, the different processing steps are described, starting from the component mixing, proceeding with the forming, until to the vulcanizing. Finally, some applications of rubber products are summarized. Chapter 4 deals with materials used for the formulations subject of this thesis, experimental procedures and characterization techniques applied. Chapter 5 is devoted to tests carried out on an ethylene vinyl acetate based rubber compound, with the aim to improve its flame retardant properties. The obtained results are indicated and some whose interpretations are presented. Chapter 6 surveys the mechanical performances of a natural rubber/polybutadiene blend. Particularly, the experimental data obtained from the rubber compounds filled with conventional fillers, such as silica and carbon black, are compared with the ones found for the rubber compounds loaded with innovative fillers, like the nanoclays.
Dopo la scoperta del processo di vulcanizzazione, le gomme hanno invaso la nostra vita e attualmente occupano un posto significativo nel mondo industriale tanto che per molte applicazioni non ci sono materiali alternativi ad esse. A differenza di quanto si potrebbe pensare, un oggetto di gomma è una sistema piuttosto complesso. Infatti, esso è in genere costituito da uno o più elastomeri e da molti altri additivi, quali ad esempio cariche rinforzanti, plastificanti, antidegradanti, agenti vulcanizzanti, etc. La realizzazione di un prodotto finito in gomma prevede una serie di operazioni. La prima di queste prevede la miscelazione dell’elastomero/i con diversi additivi ad una specifica temperatura per un tempo prefissato. Tale operazione è significativa nel determinare il grado di dispersione degli additivi nella matrice, influenzando quindi le proprietà del prodotto finale. Successivamente si verifica l’operazione di formatura durante la quale viene data una forma definita alla mescola. Infine con il processo di vulcanizzazione l’oggetto acquisisce la caratteristica proprietà di ritorno elastico, tipica delle gomma. Le proprietà finali di un prodotto di gomma dipendono innanzitutto dall’elastomero di partenza, tuttavia possono essere ampiamente manipolate variando la tipologia e la concentrazione degli additivi aggiunti e le fasi di lavorazione. Il fatto di essere un sistema multicomponente e la complessità delle fasi di produzione sono i motivi principali che hanno ritardato lo studio e lo sviluppo dei nanocompositi a base elastomera rispetto a quelli polimerici. Tuttavia, negli ultimi dieci anni il numero dei lavori scientifici sui nanocompositi elastomerici è ampiamente aumentato. Il continuo interesse deriva dal notevole miglioramento delle proprietà fisico-meccaniche che si osserva quando additivi nanodimensionali sono introdotti in una matrice elastomerica. Il miglioramento ottenuto dipende dalla dispersione a livello nanometrico che tali riempitivi possono raggiungere, contrariamente ai più comuni silice e nero fumo che si disperdono su scala micrometrica. Ad oggi, le nanocariche maggiormente studiate per la loro disponibilità in natura e il basso costo sono le nanoargille. Numerosi studi hanno dimostrato che l’aggiunta di piccole quantità di silicati a strati (< 10 wt.%) migliora le proprietà meccaniche, riduce la permeabilità ai gas e il rigonfiamento in solventi, aumenta la stabilità termica e la resistenza alla fiamma. La borsa di studio di questo dottorato è stata finanziata dalla ditta “IVG Colbachini” di Cervarese Santa Croce, Padova. L’azienda, da più di 40 anni, realizza tubi industriali in gomma per la conduzione di polveri, granuli, gas, liquidi. I prodotti di “IVG Colbachini” trovano applicazione nei settori più diversi, tra i quali l’industria chimica e agro-alimentare, l’edilizia, la cantieristica navale e da diporto, le apparecchiature ferroviarie, le lavorazioni dei metalli. Il lavoro di tesi svolto è stato dedicato allo studio e all’ottimizzazione di mescole elastomeriche prodotte in “IVG Colbachini”. Questa tesi consta di 6 capitoli e di seguito saranno riassunti brevemente gli argomenti principali trattati in ciascun capitolo. Il Capitolo 1 evidenzia le differenze sostanziali tra composito convenzionale e nanocomposito, fornendo anche una classificazione di quest’ultima categoria di materiali. Inoltre spiega quali caratteristiche di un filler sono di fondamentale importanza per la realizzazione di un nanocomposito e come ciascuna di esse influenzi le proprietà del materiale finale. Nel Capitolo 2 è contenuta una presentazione delle nanoargille e dei nanocompositi elastomerica additivati con filler a strati. In particolare si descrivono la struttura chimica di quest’ultimi e l’importanza del modificante organico. A questo si aggiunge un quadro dei metodi di sintesi di questi nanocompositi e delle loro proprietà tipiche riportate in letteratura, quali le prestazioni meccaniche, l’effetto barriera ai gas e la resistenza alla fiamma. Il Capitolo 3 illustra passo passo l’arte della lavorazione della gomma. In particolare si introduce il concetto di “ricetta elastomerica” e come viene in genere espressa. Vengono specificate le tipologie, le caratteristiche e le funzioni dei diversi componenti di una “ricetta”. Inoltre si descrivono le varie fasi di produzione di un oggetto in gomma, partendo dalla miscelazione degli ingredienti, passando per la formatura, arrivando fino al processo di vulcanizzazione. In questo capitolo vengono infine riportate alcune possibili applicazioni di prodotti in gomma. Nel Capitolo 4 si introducono i materiali impiegati per la produzione delle formulazioni, oggetto di questo lavoro di tesi, le procedure sperimentali e le tecniche di caratterizzazione utilizzate. Il Capitolo 5 illustra le prove condotte su una mescola elastomerica a base di etilene vinil acetato, con lo scopo di migliorarne le proprietà di resistenza alla fiamma. Vengono quindi riportati i risultati ottenuti e proposte alcune interpretazioni di essi. Nel Capitolo 6 ci si è concentrati sullo studio delle proprietà meccaniche di un blend costituito da gomma naturale e polibutadiene. In particolare, i dati sperimentali ottenuti da mescole contenenti riempitivi tradizionali, come silice e nero fumo, sono stati confrontati con quelli ricavati da compound con filler innovativi, quali le nanoargille.

In recent years there has been a growing attention from the world of research and companies in the field of Machine Learning. This interest, thanks mainly to the increasing availability of large amounts of data, and the respective strengthening of the hardware sector useful for their analysis, has led to the birth of Deep Learning. The growing computing capacity and the use of mathematical optimization techniques, already studied in depth but with few applications due to a low computational power, have then allowed the development of a new approach called Reinforcement Learning. This thesis work is part of an industrial process of selection of fruit for sale, thanks to the identification and classification of any defects present on it. The final objective is to measure the similarity between them, being able to identify and link them together, even if coming from optical acquisitions obtained at different time steps. We therefore studied a class of algorithms characteristic of Reinforcement Learning, the policy gradient methods, in order to train a feedforward neural network to compare possible malformations of the same fruit. Finally, an applicability study was made, based on real data, in which the model was compared on different fruit rolling dynamics and with different versions of the network.

Abstract The amount of digital data surrounding us has exploded within the past years. In industry, data are gathered from different production phases with the intent to use the data to improve the overall manufacturing process. However, management and utilization of these huge data sets is not straightforward. Thus, a computer-driven approach called data mining has become an attractive research area. Using data mining methods, new and useful information can be extracted from enormous data sets. In this thesis, diverse industrial problems are approached using data mining methods based on similarity. Similarity information is shown to give an additional advantage in different phases of manufacturing. Similarity information is utilized with smaller-scale problems, but also in a broader perspective when aiming to improve the whole manufacturing process. Different ways of utilizing similarity are also introduced. Methods are chosen to emphasize the similarity aspect; some of the methods rely entirely on similarity information, while other methods just preserve similarity information as a result. The actual problems covered in this thesis are from quality control, process monitoring, improvement of manufacturing efficiency and model maintenance. They are real-world problems from two different application areas: spot welding and steel manufacturing. Thus, this thesis clearly shows how the industry can benefit from the presented data mining methods.

The use of low-power wireless sensors and actuators with networking support in in-dustry has increased over the past decade. New generations of microcontrollers, new hardware for communication, and the use of standardized protocols such as the Internet Protocol have resulted in more possibilities for interoperability than ever before. This in-creasing interoperability allows sensors and actuator nodes to exchange information with large numbers of peers, which is beneficial for creating advanced, flexible and reusable systems.The increase in interoperability has resulted in an increase in the number of possible attacks from malicious devices or users. For this reason, the use of encryption techniques to protect client and server communications has become mandatory. However, even with state-of-the-art encryption mechanisms, there is no protection that can control access to each particular service with fine-grained precision. The nodes within an industrial network of wireless sensors and actuators are resource-constrained embedded devices, and increasing interoperability therefore requires a higher level of computation capabil-ities. The nodes’ intrinsic limitations of memory and processing exert an adverse effect on power consumption and communication delays, resulting in a shorter battery life-time. Therefore, the standard computing solutions for Internet communications are not directly applicable, and new mechanisms to achieve security, scalability, dependability, interoperability and energy efficiency are needed.Sensor and actuator networks can transmit sensed data, but they also offer access to the actuators. Such accesses, presumably provided via services, require an access protection scheme. For this reason, the use of access control mechanisms is mandatory. Access control assists in the creation of customized services and access policies. These access policies can isolate access permissions to devices with different roles, such as production and maintenance.The main contribution of this thesis is a novel, efficient IoT framework for industrial applications, including design, implementation, and experimental validation. The frame-work includes features for communication protection, authentication, fine-grained access control, zero-configuration networking, and run-time reconfiguration. These technologies and their corresponding energy consumption data clearly demonstrate the feasibility of integrating a battery-operated IoT concept into a functional System of Systems. The provided data also pinpoint the most critical areas for improvement.

Augmented reality (AR) is a way of overlaying digital information onto a picture or a videofeed and has been used in industrial contexts for more than 20 years. This Master's Thesis examines if AR can be used to help maintenance engineers set up and maintain robot environments by visualizing robot movement and safety zones. The main result of the Master's Thesis is a prototype application for a tablet computer. The user points the tablet towards a robot filming it and the video feed is displayed on the screen. This video feed is augmented with a virtual zone displayed around the robot, illustratingthe area where the robot is allowed to move. The application fetches the coordinatesfor the zone from the safety system SafeMove { a system designed by ABB to increase safetyand allow closer human-robot collaboration.The visualization of a SafeMove conguration is currently limited to an image of atwo-dimensional coordinate system showing the zone as a set of dierent coordinates. Thismakes it dicult grasping the full layout of the three-dimensional zone. By using theapplication the user gets a better view of the layout, allowing the user to look at the robot from different sides and see the safety zone projected around the robot. User tests show that people working with SafeMove could benefit from using the application to verify the conguration of SafeMove systems and the conclusion is that AR, if used right, greatly can improve robot interaction and maintenance.

Thesis (Sc. D.)--Massachusetts Institute of Technology, Dept. of Nuclear Engineering, 2000.
Includes bibliographical references (p. 187-198).
Photonuclear resonance excitation refers to a variety of photonuclear interaction processes that lead to the excitation of a nucleus from some initial state to a higher energy nuclear state. Typical excited nuclear state lifetimes are short, ranging from nanoseconds to femtoseconds or less; however, some isotopes have unusually long-lived excited nuclear energy states, or isomers. This dissertation examines the feasibility of using bremsstrahlung irradiation sources to produce isomers for industrial applications. In contrast with charged particle based isomer production, the use of high energy photons allows for the irradiation and production of isomers in bulk materials. The commercial availability of reliable, high power industrial electron accelerators means that isomer activities sufficient for industrial applications may be achieved using bremsstrahlung, in contrast with neutron based approaches where suitable neutron sources of sufficient intensity for these applications are lacking. In order to design a system for creating nuclear isomers using photons, the resonant photon absorption isomeric excitation cross section must be known. Unlike neutron absorption and scattering cross sections, comparatively little information exists for photon isomeric excitation. To address this, a theoretical model based upon statistical probability distributions of nuclear energy levels has been developed for calculating photon excitation cross sections at energies below neutron and proton binding energies; the ideal region of operation for most applications in order to minimize long term activation of materials. Isomeric excitation cross sections calculated using this technique have been compared with experimentally measured values and are found to agree to within a factor of two or better.
(cont.) sing this, a general transition equation suitable for both nuclear resonance fluorescence and isomer excitation has been developed for calculating nuclear level distribution probabilities for materials undergoing photon irradiation. Experiments have been carried out using an industrial 6 MeV electron accelerator to identify obstacles related to nuclear resonance fluorescence measurements as well as measurements of the decay of short-lived isomers using scintillators in the vicinity of high intensity bremsstrahlung sources. Use of a fast switching gating circuit in combination with a pulsed accelerator was found to be a satisfactory solution for dealing with problems related to the performance of a detectors photomultiplier tube as a result of exposure to scattered radiation during the beam pulse. Calculations have been carried out to assess the performance characteristics which could be expected from industrial photonuclear resonance excitation systems, based upon a 10 MeV electron accelerator. For simple isomer production, specific activities on the order of 1 mCi/g/mA can be expected for irradiation periods sufficiently long for equilibrium to be reached. For the analysis of arsenic concentrations in environmental samples, sensitivities of 1 +/- 0.1 ppm could be achieved using accelerator currents of 50 - 100 [mu]A with irradiations times of a few minutes or less. A system designed to analyze ore traveling along a conveyor belt could be used to sort gold ore based upon a lower grade cutoff of 5 ppm using an accelerator of 10 mA ...
by David Lee Chichester.
Sc.D.

This thesis describes the theory, development and applications of laser beam characterisation for industrial laser materials processing systems. Descriptions are given of novel forms of beam diagnostic systems and their integration into highly automated industrial tools. Work is also presented that has contributed to the new ISO standard on beam characterisation. Particular emphasis is given to excimer laser applications and UV micromachining.

Magnetic resonance (MR) is a highly versatile technique with great potential for use in industrial applications; from the in situ study of unit operations to the optimisation of product properties. This thesis, concerned with the latter, is divided into two parts. Firstly, dynamic MR is applied to characterise the flow behaviour, or rheology, of process fluids. Such characterisation is typically performed using conventional rheometry methods operating offline, with an online, or inline, method sought for process control and optimisation. Until recently, MR was an unlikely choice for this application due to the requirement of high-field MR hardware. However, recent developments in low-field MR hardware mean that the potential of MR in such applications can now be realised. Since the implementation of MR flow imaging is challenging on low-field MR hardware, two new approaches to MR rheometry are described using pulsed field gradient (PFG) MR. A cumulant analysis of the PFG MR signal is first used to characterise the rheology of model power-law fluids, namely xanthan gum-in-water solutions, accurate to within 5% of conventional rheometry, the data being acquired in only 6% of the time required when using MR flow imaging. The second approach utilises a Bayesian analysis of the PFG MR signal to characterise the rheology of model Herschel--Bulkley fluids, namely Carbopol 940-in-water solutions; data are acquired in only 12% of the time required for analysis using MR flow imaging. The suitability of the Bayesian MR approach to study process fluids is demonstrated through experimental study on an alumina-in-acetic acid slurry used by Johnson Matthey. Secondly, MR imaging is used to provide insights into the origins and mechanisms of colloidal gel collapse. Many industrial products are colloidal gels, a space-spanning network of attractive particles with a yield stress. Colloidal gels are, however, known to undergo gravitational collapse after a latency period, thus limiting the shelf-life of products. This remains poorly understood, with a more detailed understanding of both fundamental interest and practical importance. To this end, MR imaging is applied offline to investigate the phase behaviour of colloidal gels. In particular, a comparison of the simulated and experimental phase diagrams suggests gravitational gel collapse to be gravity-driven. Furthermore, measurement of the colloid volume fraction using MR imaging indicates the formation of clusters of colloids at the top of the samples. Whether such clusters initiate gravitational gel collapse is yield stress-dependent; the gravitational stress exerted by a cluster must be sufficient to yield the colloidal gel.

The nitrilases are an important class of industrial enzymes that are found in all phyla. These enzymes are expressed widely in prokaryotes and eukaryotes. Nitrilases convert nitriles to corresponding acids and ammonia. They are used in industry as biocatalysts because of their specificity and enantioselectivity. These enzymes belong to the nitrilase superfamily in which members share a common &alpha
&beta
&beta
&alpha
structural fold and a unique cys, glu,lys catalytic triad with divergent N- and C-terminals.

There are four atomic structures of distant homologues in the superfamily, namely 1ems, 1erz, 1f89 and 1j31. All structures have two-fold symmetry which conserves the &alpha
&beta
&beta
&alpha
-&alpha
&beta
&beta
&alpha
fold across the dimer interface known as the A surface. The construction of a 3D model based on the solved structures revealed the enzyme has two significant insertions in its sequence relative to the solved structures, which possibly correspond to the C surface. In addition there are intermolecular interactions in a region of a conserved helix, called the D surface. These surfaces contribute additional interactions responsible for spiral formation and are absent in the atomic resolution homologues.

The recombinant enzyme from R.rhodochrous J1 was expressed in E. coli BL21 cells and eluted by gel filtration chromatography as an active 480 kDa oligomer and an inactive 80 kDa dimer in the absence of benzonitrile. This contradicts previous observations, which reported the native enzyme exists as an inactive dimer and elutes as a decamer in the presence benzonitrile. Reducing SDS-PAGE showed a subunit atomic mass of ~40 kDa. EM and image analysis revealed single particles of various shapes and sizes, including c-shaped particles, which could not form spirals due to steric hindrances in its C terminal.

Chromatographic re-elution of an active fraction of 1-month old J1 nitrilase enabled us to identify an active form with a mass greater than 1.5 MDa. Reducing SDS-PAGE, N-terminal sequencing and mass spectroscopy showed the molecular weight was ~36.5 kDa as result of specific proteolysis in its C terminal. EM revealed the enzyme forms regular long fibres. Micrographs (109) were recorded on film using a JEOL 1200EXII operating at 120 kV at 50K magnification. Two independent 3D reconstructions were generated using the IHRSR algorithm executed in SPIDER. These converged to the same structure and the resolution using the FSC 0.5 criterion was 1.7 nm.

The helix structure has a diameter of 13nm with ~5 dimers per turn in a pitch of 77.23 Å
. Homology modeling and subsequent fitting into the EM map has revealed the helix is built primarily from dimers, which interact via the C and D surfaces. The residues, which potentially interact across the D surface, have been identified and these confer stability to the helix. The conservation of the insertions and the possibility of salt bridge formation on the D surface suggest that spiral formation is common among microbial nitrilases. Furthermore, the presence of the C terminal domain in J1 nitrilase creates a steric hindrance that prevents spiral formation. When this is lost &ndash
either by specific proteolysis or autolysis - an active helix is formed.

Cross-linked polyethylene has higher upper use temperature than normal polyethylene and is used as an insulating material for electricity carrying cables and hot water pipes. The most common method of inducing crosslinks is by reaction with silanes. After incorporation of silanes into polyethylene and upon hydrolysis with ambient moisture or with hot water, Si-O-Si crosslinks are formed between the various linear polyethylene chains. Industrially, this reaction is performed routinely. However, the efficiency of this reaction with respect to the silane is low and control of product distribution is difficult. A precise fundamental understanding is necessary to be able to manipulate the reactions and thus, allow for the facile processing of the polymers. Hydrocarbon models of polymers - heptane, dodecane - are being used to study this reaction in the laboratory. For the reaction, vinyltrimethoxysilane is used as the grafting agent along with di-tert-butyl peroxide as the radical initiator. MALDI, a mass spectrometric technique is used for the analysis of the product distribution after work-up. Advanced NMR techniques (COSY, HSQC, DEPT, APT, HMBC) are being conducted on the grafted hydrocarbon compounds to gain an in-depth understanding of the mechanism and regiochemistry of the grafting reaction. Scalable and cost effective methods to capture CO2 are important to counterbalance some of the global impact of the combustion of fossil fuels on climate change. The main options available now include absorption, adsorption and membrane technology. Amines, especially monoethanolamine, have been the most commercialized technology. However, it is not without disadvantages. House et al have investigated the energy penalty involved in the post-combustion CO2 capture and storage from coal-fired power plants and found that 15-20% reduction in the overall electricity usage is necessary to offset the penalty from capturing and storing 80% of United States coal fleet's CO2 emssions1. Novel non-aqueous amine solvents, developed by the Eckert Liotta group, react with CO2 to form ionic liquids. The ionic liquids readily desorb CO2 upon heating, regenerating the reactive amines and this cycle can be carried out multiple times. An iterative procedure is being adopted to develop amine solvents for CO2 capture. Thermodynamic information like reversal temperature and boiling point of the solvents are collected; they are then used to formulate structure property relationships which allow for new molecules to be engineered. On reaction with CO2, there is a sharp increase in viscosity which is unfavorable from a processing standpoint. Many approaches to mitigate and control viscosity are being studied as well. 1House et al, Energy Environ Sci, 2009, 2, 193-205

This thesis presents the design, fabrication and testing of novel grating based Optical Fibre Sensor (OFS) systems being interrogated using “off the shelf” interrogation systems, with the eventual development of marketable commercial systems at the forefront of the research. Both in the industrial weighing and aerospace industries, there has been a drive to investigate the feasibility of using optical fibre sensors being deployed where traditionally their electrical or mechanical counterparts would traditionally have been. Already, in the industrial weighing industry, commercial operators are deploying OFS-based Weigh-In-Motion (WIM) systems. Likewise, in the aerospace industry, OFS have been deployed to monitor such parameters as load history, impact detection, structural damage, overload detection, centre of gravity and the determination of blade shape. Based on the intrinsic properties of fibre Bragg gratings (FBGs) and Long Period Fibre Gratings (LPFGs), a number of novel OFS-based systems have been realised. Experimental work has shown that in the case of static industrial weighing, FBGs can be integrated with current commercial products and used to detect applied loads. The work has also shown that embedding FBGs in e-glass, to form a sensing patch, can result in said patches being bonded to rail track, forming the basis of an FBG-based WIM system. The results obtained have been sufficiently encouraging to the industrial partner that this work will be progressed beyond the scope of the work presented in this thesis. Likewise, and to the best of the author’s knowledge, a novel Bragg grating based systems for aircraft fuel parameter sensing has been presented. FBG-based pressure sensors have been shown to demonstrate good sensitivity, linearity and repeatability, whilst LPFG-based systems have demonstrated a far greater sensitivity when compared to FBGs, as well the advantage of being potentially able to detect causes of fuel adulteration based on their sensitivity to refractive index (RI). In the case of the LPFG-based system, considerable work remains to be done on the mechanical strengthening to improve its survivability in a live aircraft fuel tank environment. The FBG system has already been developed to an aerospace compliant prototype and is due to be tested at the fuel testing facility based at Airbus, Filton, UK. It is envisaged by the author that in both application areas, continued research in this area will lead to the eventual development of marketable commercial products.

Randomness is one of the most intriguing, inspiring and debated topics in the history of the world. It appears every time we wonder about our existence, about the way we are, e.g. Do we have free will? Is evolution a result of chance? It is also present in any attempt to understand our anchoring to the universe, and about the rules behind the universe itself, e.g. Why are we here and when and why did all this start? Is the universe deterministic or does unpredictability exist? Remarkably, randomness also plays a central role in the information era and technology. Random digits are used in communication protocols like Ethernet, in search engines and in processing algorithms as page rank. Randomness is also widely used in so-called Monte Carlo methods in physics, biology, chemistry, finance and mathematics, as well as in many other disciplines. However, the most iconic use of random digits is found in cryptography. Random numbers are used to generate cryptographic keys, which are the most basic element to provide security and privacy to any form of secure communication. This thesis has been carried out with the following questions in mind: Does randomness exist in photonics? If so, how do we mine it and how do we mine it in a massively scalable manner so that everyone can easily use it? Addressing these two questions lead us to combine tools from fundamental physics and engineering. The thesis starts with an in-depth study of the phase diffusion process in semiconductor lasers and its application to random number generation. In contrast to other physical processes based on deterministic laws of nature, the phase diffusion process has a pure quantum mechanical origin, and, as such, is an ideal source for generating truly unpredictable digits. First, we experimentally demonstrated the fastest quantum random number generation scheme ever reported (at the time), using components from the telecommunications industry only. Up to 40 Gb/s were demonstrated to be possible using a pulsed scheme. We then moved towards building prototypes and testing them with partners in supercomputation and fundamental research. In particular, the devices developed during this thesis were used in the landmark loophole- free Bell test experiments of 2015. In the process of building the technology, we started a new research focus as an attempt to answer the following question: How do we know that the digits that we generate are really coming from the phase diffusion process that we trust? As a result, we introduced the randomness metrology methodology, which can be used to derive quantitative bounds on the quality of any physical random number generation device. Finally, we moved towards miniaturisation of the technology by leveraging techniques from the photonic integrated circuits technology industry. The first fully integrated quantum random number generator was demonstrated using a novel two-laser scheme on an Indium Phosphide platform. In addition, we also demonstrated the integration of part of the technology on a Silicon Photonics platform, opening the door towards manufacturing in the most advanced semiconductor industry.
L’aleatorietat és un dels temes més intrigants, inspiradors i debatuts al llarg de la història. És un concepte que sorgeix quan ens preguntem sobre la nostra pròpia existència i de per què som com som. Tenim freewill? És l’evolució resultat de l’atzar? L’aleatorietat és també un tema que sorgeix quan intentem entendre la nostra relació amb l’univers mateix. Per què estem aquí? Quan o com va començar tot això? És l’univers una màquina determinista o hi ha cabuda per a l’atzar? Sorprenentment, l’aleatorietat també juga un paper crucial en l’era de la informació i la tecnologia. Els nombres aleatoris es fan servir en protocols de comunicació com Ethernet, en algoritmes de classificació i processat com Page Rank. També usem l’aleatorietat en els mètodes Monte Carlo, que s’utilitzen en els àmbits de la física, la biologia, la química, les finances o les matemàtiques. Malgrat això, l’aplicació més icònica per als nombres aleatoris la trobem en el camp de la criptografia o ciber-seguretat. Els nombres aleatoris es fan servir per a generar claus criptogràfiques, l’element bàsic que proporciona la seguretat i privacitat a les nostres comunicacions. Aquesta tesi parteix de la següent pregunta fonamental: Existeix l’aleatorietat a la fotònica? En cas afirmatiu, com podem extreure-la i ferla accessible a tothom? Per a afrontar aquestes dues preguntes, s’han combinat eines des de la física fonamental fins a l’enginyeria. La tesi parteix d’un estudi detallat del procés de difusió de fase en làsers semiconductors i de com aplicar aquest procés per a la generació de nombres aleatoris. A diferència d’altres processos físics basats en lleis deterministes de la natura, la difusió de fase té un origen purament quàntic, i per tant, és una font ideal per a generar nombres aleatoris. Primerament, i fent servir aquest procés de difusió de fase, vam crear el generador quàntic de nombres aleatoris més ràpid mai implementat (en aquell moment) fent servir, únicament, components de la indústria de les telecomunicacions. Més de 40 Gb/s van ser demostrats fent servir un esquema de làser polsat. Posteriorment, vam construir diversos prototips que van ser testejats en aplicacions de ciència fonamental i supercomputació. En particular, alguns dels prototips desenvolupats en aquesta tesi van ser claus en els famosos experiments loophole-free Bell tests realitzats l’any 2015. En el procés de construir aquests prototips, vam iniciar una nova línia de recerca per a intentar contestar una nova pregunta: Com sabem si els nombres aleatoris que generem realment sorgeixen del procés de difusió de fase, tal com nosaltres creiem? Com a resultat, vam introduir una nova metodologia, la metrologia de l’aleatorietat. Aquesta es pot fer servir per a derivar límits quantificables sobre la qualitat de qualsevol dispositiu de generació de nombres aleatoris físic. Finalment, ens vam moure en la direcció de la miniaturització de la tecnologia utilitzant tècniques de la indústria de la fotònica integrada. En particular, vam demostrar el primer generador de nombres aleatoris quàntic totalment integrat, fent servir un esquema de dos làsers en un xip de Fosfur d’Indi. En paral·lel, també vam demostrar la integració d’una part del dispositiu emprant tecnologia de Silici, obrint les portes, per tant, a la producció a gran escala a través de la indústria més avançada de semiconductors.
La aleatoriedad es uno de los temas más intrigantes, inspiradores y debatidos a lo largo de la historia. Es un concepto que surge cuando nos preguntamos sobre nuestra propia existencia y de por qué somos como somos. ¿Tenemos libre albedrío? ¿Es la evolución resultado del azar? La aleatoriedad es también un tema que surge cuando intentamos entender nuestra relación con el universo. ¿Por qué estamos aquí? ¿Cuándo y cómo empezó todo esto? ¿Es el universo una máquina determinista o existe espacio para el azar? Sorprendentemente, la aleatoriedad también juega un papel crucial en la era de la información y la tecnología. Los números aleatorios se usan en protocolos de comunicación como Ethernet, y en algoritmos de clasificación y procesado como Page Rank. También la utilizamos en los métodos Monte Carlo, que sirven en los ámbitos de la física, la biología, la química, las finanzas o las matemáticas. Sin embargo, la aplicación más icónica para los números aleatorios la encontramos en el campo de la criptografía y la ciberseguridad. Aquí, los números aleatorios se usan para generar claves criptográficas, proporcionando el elemento básico para dotar a nuestras comunicaciones de seguridad y privacidad. En esta tesis partimos de la siguiente pregunta fundamental: ¿Existe la aleatoriedad en la fotónica? En caso afirmativo, ¿Cómo podemos extraerla y hacerla accesible a todo el mundo? Para afrontar estas dos preguntas, se han combinado herramientas desde la física fundamental hasta la ingeniería. La tesis parte de un estudio detallado del proceso de difusión de fase en láseres semiconductores y de cómo aplicar este proceso para la generación de números aleatorios. A diferencia de otros procesos físicos basados en leyes deterministas de la naturaleza, la difusión de fase tiene un origen puramente cuántico y, por lo tanto, es una fuente ideal para generar números aleatorios. Primeramente, y utilizando este proceso de difusión de fase, creamos el generador cuántico de números aleatorios más rápido nunca implementado (en ese momento) utilizando únicamente componentes de la industria de las telecomunicaciones. Más de 40 Gb/s fueron demostrados utilizando un esquema de láser pulsado. Posteriormente, construimos varios prototipos que fueron testeados en aplicaciones de ciencia fundamental y supercomputación. En particular, algunos de los prototipos desarrollados en esta tesis fueron claves en los famosos experimentos Loophole-free Bell tests realizados en el 2015. En el proceso de construir estos prototipos, iniciamos una nueva línea de investigación para intentar dar respuesta a una nueva pregunta: ¿Cómo sabemos si los números aleatorios que generamos realmente surgen del proceso de difusión de fase, tal y como nosotros creemos? Como resultado introdujimos una nueva metodología, la metrología de la aleatoriedad. Esta se puede usar para derivar límites cuantificables sobre la calidad de cualquier dispositivo de generación de números aleatorios físico. Finalmente, nos movimos en la dirección de la miniaturización de la tecnología utilizando técnicas de la industria de la fotónica integrada. En particular, creamos el primer generador de números aleatorios cuántico totalmente integrado utilizando un esquema de dos láseres en un chip de Fosfuro de Indio. En paralelo, también demostramos la integración de una parte del dispositivo utilizando tecnología de Silicio, abriendo las puertas, por tanto, a la producción a gran escala a través de la industria más avanzada de semiconductores.

Enzyme catalysis has been scaled up for several industrial sectors during the last decades, including pharmaceutics, food and beverages. This raised the interest of other industries such as energy or paper and pulp sectors, for which biocatalysts need to be improved in order to be economically competitive. Molecular simulations play a key role in finding new applications for enzymes and reducing the costs of experimental work. In this thesis, computational techniques have been developed and applied on oxidoreductases for industrial needs, under the framework of the INDOX project. More specifically, investigations have been focused on lignin degradation and valorization by means of flavoproteins and laccases. Flavoproteins are oxidoreductases containing FAD in most of the cases as a prosthetic group. On the other hand, laccases are multicopper-containing oxidoreductases that can oxidize a large variety of substrates. In both cases, oxygen can act as the final electron acceptor, producing hydrogen peroxide and water as by-products respectively, which makes them suitable for green chemistry applications. Research on flavoproteins - Efforts were made to improve catalytic activity towards 5-hydroxymethylfurfural oxidation and secondary benzyl alcohols. To do so, a Monte Carlo based in-house algorithm was employed to sample the protein-ligand conformational space, revealing two interesting positions for the aryl-alcohol oxidase flavoprotein, 500 and 501. Experimental collaborators found several variants from these positions that improved the activity of this enzyme for secondary benzyl alcohols, which were then computationally characterized with the same methodology. Investigation of the 5-hydroxymethylfurfural biochemical reaction was also performed, revealing that the negative nature of the carboxylic group limits the diffusion of one of the subproducts. This problem was proven to be solved by using another flavoprotein, 5-hydroxymethylfurfural oxidase, whose larger active site allows the diffusion of the negatively charged substrate although with very low activity. The activity of this enzyme increases with the addition of V367/W466F double mutations, and the main reasons have been identified. Research on laccases - The addition of a ruthenium photosensitizer potentially allows the oxidation of challenging substrates, although the electron back-transfer was presented as the main limitation. Electron transfer calculations were performed empirically, instead of quantum calculations, to selectively attach the photosensitizer to the surface of the enzyme, finding that increasing the directionality towards the trinuclear copper cluster was the best strategy. In conclusion, it has been shown how computational studies are a useful complement to experimental work, for the benefit of industries. In silico applications are shown to be beneficial in both ends of the process. On one side, predictions and designs are made to guide experiments, narrowing the spectra of mutations and reducing costs. After the experiments, simulations still provide insights of the reaction mechanism and information about the role of certain amino acids, to better understand the experimental results and, as a final instance, refine future predictions.
La catàlisi enzimàtica a escala industrial s’ha aplicat a diferents sectors durant dècades, entre les quals s’inclouen farmacèutiques i indústries de beure i menjar. Això ha incrementat l’interès d’altres sectors, com ara empreses energètiques i de processat del paper, per les quals els biocatalitzadors encara necessiten millores per tal de poder ser competitius econòmicament. Les simulacions a nivell molecular juguen un paper important en trobar noves aplicacions per als enzims i reduint els costs de la part experimental. En aquesta tesi s’han desenvolupat i aplicat tècniques computacionals en oxidoreductases per a la indústria, sota el marc del projecte INDOX. Concretament, les investigacions s’han centrat en la degradació de la lignina i l’aprofitament dels productes obtinguts mitjançant flavoproteïnes i lacases. Les flavoproteïnes son oxidoreductases que contenen majoritàriament FAD com a cofactor. L’activitat de l’aril alcohol oxidasa s’ha millorat per a alcohols benzílics secundaris, mantenint la selectivitat per l’enantiòmer S. D’altra banda, les lacases son oxidoreductases que contenen múltiples àtoms de coure i que poden oxidar una gran varietat de substrats. En concret, s’han realitzat mesures de transferència electrònica empíriques per tal d’unir un fotosensibilitzador a la superfície d’una lacasa, amb la idea de maximitzar la transferència electrònica al clúster tri-nuclear de coure. En ambdós enzims, una molècula d’oxigen pot ser l’acceptor final d’electrons, produint peròxid d’hidrogen i aigua respectivament com a subproductes, per la qual cosa se’ls considera adequats per a aplicacions de química sostenible. En aquesta tesi es demostra que els estudis computacionals són un bon complement del treball experimental, on hi surt beneficiada la indústria. Les aplicacions dels mètodes in silico son útils per a dissenyar i fer prediccions per tal de guiar els experiments, reduir el número de possibilitats on fer mutacions i tanmateix, els costs. A més a més, les simulacions ofereixen la possibilitat de donar explicacions a posteriori sobre els efectes dels mutants obtinguts, així com informació sobre el mecanisme de reacció i el rol dels aminoàcids.

In this thesis, the automation of the long term and cyclic processes by using a programmable control unit is aimed. To achieve this goal, timing relays and various microcontrollers are investigated. PIC microcontroller is chosen to implement the control unit due to its advantages like high speed, Harvard and RISC architecture, low cost and flexibility for programming. Theory of the PIC microcontrollers is studied and a controller unit to be used in the mentioned processes is designed. Some features are added to the device to widen the application fields and consequently a multi-purpose programmable controller is realized. In the device, Microchip PIC16F877 is used as the microcontroller. The code of the controller is written in Assembly Language and is compiled with MPASM. This controller counts the signals coming from internal Timer 555 or external signals and activates ten different outputs in order. The operating times of the outputs can be changed by a keypad and shown in a display. By keeping the number of the used ports of the microcontroller, as few as possible, room for the future improvements and additions is provided.

The aim of this Master’s Degree thesis project is to design and develop point diffraction interferometer plates. In this project the PDI plates are re-designed, changing the design which was used in previous projects in Halmstad University. The transparency of PDI plates can be controlled by coating them with NiCr film. Firstly, four plates with coating of different thickness of NiCr were developed. The relationship between transmittance and the thickness of NiCr was established by testing these plates for transmittance and reflectance with the help of a laser and an optical power meter. The absorption coefficient of clear substrates and reflection of light is also taken into account to achieve the correct results. The parameters like the diameter of semi-transparent area around the pinholes and the size of pinholes is chosen after fully understanding its application. The lay-out and description of design is also included in the report.

This thesis deals with the industrial application of sliding mode controllers. Sliding mode controllers based on both linear models and nonlinear-models are considered. Special attention is paid to the nonlinear modelling of the systems for sliding mode controller design. The possibility of using neural networks for model generation is explored. Novel schemes for uncertainty bounds estimation are introduced and subsequently used for robust sliding mode controller design. Later, a novel approach for sliding mode based parameter estimation for a nonlinear model with known structure but unknown parameters is introduced. This parameter estimation scheme is integrated with sliding mode controller design to provide an overall controller design framework. The stability of these schemes is proven through quadratic stability concepts. The sliding mode controller design frameworks mentioned above are verified and tested on challenging industrial examples. The temperature control of a high temperature multiburner industrial furnace is a highly coupled and extremely nonlinear problem. A multiburner furnace nonlinear simulation facility is established and used for linear identification and subsequently linear model based sliding mode controller testing. For comparison purposes a two degree of freedom H controller is also designed and tested. Then a nonlinear model based controller is tested on a single burner furnace simulation. Idle speed control of an automobile engine is an extremely difficult control problem characterised with severe nonlinearities, gross disturbances and huge time delays. A sliding mode controller is designed for this problem and successfully implemented on a test rig. Later on, a nonlinear model based sliding mode controller is designed for the same problem and successfully tested.

Working in a commercial research environment, the exploration of novel techniques and adapting methods to the specific needs of the scientific projects running within the organisation, dictate an "out of the box" and creative approach using established techniques from other disciplines (e.g. biomedical and materials science). The company's research projects demand high throughput speed combined with a "simple" but "all-inclusive" answer that correlates compositional information to microstructure from macroscopic down to the nanometre level. The motivation behind this PhD study was the desire to work out the key experimental and practical issues involved in designing and deploying a scanning electron microscope (SEM) combined with Raman spectroscopy, the so-called .Q!YQ. In-SEM Raman system. This cryo In-SEM Raman instrument enables the retrieval of compositional information by Raman spectroscopy at a micrometer-resolution level from a frozen hydrated bulk specimens and correlate this to the observed micro- structure. The effects of cooling on the Raman spectra of model systems were studied. The application section of this thesis shows the added value of the combined technique by analysing the distribution of carotenoids in processed food emulsions. The significance of this combination of techniques is found in the installation of two cryo In-SEM Raman systems in 2011 in the Dutch academia. To produce optimal sample geometry for In-SEM Raman but also for analytical microscopy in general, sample preparation procedures were developed resulting in a grazing angle approach for the Focussed Ion Beam-SEM technique that significantly reduces analysis time while increasing the sample dimensions required for analytical add-ons to the scanning electron microscope. Low temperature resin embedding and cryo preparation methods for the preparation of open and brittle structures packed with organic solvents were developed enabling chemical mapping of the liquid phases of detergent particles by energy dispersive x- ray spectroscopy.

This thesis introduces the circuit and system design of interactive Radio-Frequency Identification (RFID) for Internet of Things (IoT) applications. IoT has the vision of connectivity for anything, at anytime and anywhere. One of the most important characteristics of IoT is the automatic and massive interaction of real physical world (things and human) with the virtual Internet world.RFID tags integrated with sensors have been considered as one suitable technology for realizing the interaction. However, while it is important to have RFID tags with sensors as the input interaction, it is also important to have RFID tags with displays as the output interaction.Display interfaces vary based on the information and application scenarios. On one side, remote and centralized display interface is more suitable for scenarios such as monitoring and localization. On the other side, tag level display interface is more suitable for scenarios such as object identification and online to offline propagation. For tag level display, though a substantial number of researches have focused on introducing sensing functionalities to low power Ultra-High Frequency (UHF) RFID tags, few works address UHF RFID tags with display interfaces. Power consumption and integration with display of rigid substrate are two main challenges.With the recent emerging of Electronic Paper Display (EPD) technologies, it becomes possible to overcome the two challenges. EPD resembles ordinary ink on paper by characteristics of substrate flexibility, pattern printability and material bi-stability. Average power consumption of display is significantly reduced due to bi-stability, the ability to hold color for certain periods without power supplies. Among different EPD types, Electrochromic (EC) display shows advantage of low driving voltage compatible to chip supply voltage.Therefore this thesis designs a low power UHF RFID tag integrated in 180 nm CMOS process with inkjet-printed EC polyimide display. For applications where refresh rate is ultra-low (such as electronic label in retailing and warehouse), the wireless display tag is passive and supplied by the energy harvested from UHF RF wave. For applications where refresh rate is not ultra-low (such as object identification label in mass customized manufacturing), the wireless display tag is semi-passive and supplied by soft battery. It works at low average power consumption and with out-of-battery alert. For remote and centralized display, the limitations of uplink (from tags to reader) capacity and massive-tag information feedback in IoT scenarios is the main challenge. Compared to conventional UHF RFID backscattering whose data rate is limited within hundreds of kb/s, Ultra-wideband (UWB) transmission have been verified with the performance of Mb/s data rate with several tens of pJ/pulse energy consumption.Therefore, a circuit prototype of UHF/UWB RFID tag replacing UHF backscattering with UWB transmitter is implemented. It also consists of Analog-to-Digital Converter (ADC) and Electrocardiogram (ECG) electrodes for healthcare applications of real-time remote monitoring of multiple patients ECG signals. The ECG electrodes are fabricated on paper substrate by inkjet printing to improve patient comfort. Key contribution of the thesis includes: 1) the power management scheme and circuit design of passive UHF/UWB RFID display tag. The tag sensitivity (the input RF power) is -10.5 dBm for EC display driving, comparable to the performance of conventional passive UHF RFID tags without display functions, and -18.5 dBm for UWB transmission, comparable to the state-of-the-art performance of passive UHF RFID tag. 2) communication flow and circuit design of UHF/UWB RFID tag with ECG sensing. The optimum system throughout is 400 tags/second with 1.5 KHz ECG sampling rate and 10 Mb/s UWB pulse rate.

QC 20151012

The goal of this thesis was to investigate improvements that could be made to Acoustic Pulse Reflectometry APR for potential industrial uses with an emphasis on natural gas distribution networks. A typical APR system's range is limited by overlap of acoustic signals that is a result of short source tubes. An experimental rig was constructed to perform a comparative study of wave separation techniques and to develop a novel absorbing termination technique to remove overlap effects in real-time. A measure known as separation index (SI) was used to assess the quality of wave separation for two different multi-microphone algorithms. The results indicated that pair selection frequency domain wave separation algorithms, used with 100 coefficient adaptive least mean square (ALMS) identified inter-microphone transfer functions provide a separation quality of up to -62.2dB. Also, results show that time domain wave separation techniques are as robust as established frequency domain techniques. In addition, a novel method termed active absorbing termination was developed to remove source reflections in real-time. This method was presented and tested using a modified version of SI called the Absorption Index. The results showed that source reflections can be removed in real-time within a single measurement which was not possible in previous works.

The changing manufacturing environment is putting pressure on shop floor operators. To help them cope with the pressures and support them in their tasks an industrial hypermedia application, a computerised information system (IS), has been proposed. However, resistance to IS by target users is a widespread problem. Therefore, companies need guidance on how to augment user acceptance, hence success, of these applications. A literature review on industrial hypermedia applications revealed that five research issues needed to be investigated. Among those issues were the lack of recommendations to hypermedia application developers and managers of manufacturing companies intending to invest in industrial hypermedia applications. This research makes a major contribution by providing a set of recommendations that when adopted may produce successful applications benefiting both the operators and the manufacturing company. This research started by reviewing management of information systems (MIS), human computer interaction (HCI) and hypermedia disciplines to deduce a set of hypothesised factors or constructs that might influence success, measured through intention to use and subjective use. The factors, individual, organisational, task and system, were employed to develop two predictive and explanatory theoretical models, a pre-implementation and a post-implementation model. Empirical data for the constructs in the models were collected in the work place and statistically analysed. Based on the analysis, the ecologically valid findings suggest that shop floor operators can use industrial hypermedia applications and are rational when deciding whether or not to use the applications. The factor having a significant positive influence on user acceptance was an attribute of the application, namely perceived usefulness. The latter is influenced by other factors that can be controlled by developers and facilitated or mitigated by managers. With the proposed recommendations developers and managers may make informed decisions and meet the challenge of developing applications that are successful. Page i Abstract Abstract The changing manufacturing environment is putting pressure on shop floor operators. To help them cope with the pressures and support them in their tasks an industrial hypermedia application, a computerised information system (IS), has been proposed. However, resistance to IS by target users is a widespread problem. Therefore, companies need guidance on how to augment user acceptance, hence success, of these applications. A literature review on industrial hypermedia applications revealed that five research issues needed to be investigated. Among those issues were the lack of recommendations to hypermedia application developers and managers of manufacturing companies intending to invest in industrial hypermedia applications. This research makes a major contribution by providing a set of recommendations that when adopted may produce successful applications benefiting both the operators and the manufacturing company. This research started by reviewing management of information systems (MIS), human computer interaction (HeI) and hypermedia disciplines to deduce a set of hypothesised factors or constructs that might influence success, measured through intention to use and subjective use. The factors, individual, organisational, task and system, were employed to develop two predictive and explanatory theoretical models, a pre-implementation and a post-implementation model. Empirical data for the constructs in the models were collected in the work place and statistically analysed. Based on the analysis, the ecologically valid findings suggest that shop floor operators can use industrial hypermedia applications and are rational when deciding whether or not to use the applications. The factor having a significant positive influence on user acceptance was an attribute of the application, namely perceived usefulness. The latter is influenced by other factors that can be controlled by developers and facilitated or mitigated by managers. With the proposed recommendations developers and managers may make informed decisions and meet the challenge of developing applications that are successful.

Wireless Sensor Networks (WSN) represent a relatively modern concept which has captured the interest of many in the research community. Coupled with appropriate hardware, they offer great flexibility in terms of their applicability to solving real world problems. This can be seen with applications ranging from environmental issues to healthcare and even artificial intelligence. Much of the work relating to WSN has been predominantly in the research domain. and so it is the purpose of this study to investigate ways in which they can be applied to solve industrial issues. This study particularly considers inventory management in the airline and packaged gas industries where there are many common fundamental requirements. A prototype system is presented which includes a database to record and obtain relevant tracking data in order to facilitate asset identification. Information of how this system may be applied within each industry is also included, in addition to how WSN can be utilised to fulfil the specific needs of individual industries through the use of custom built hardware and sensors. Initial experimental results of this system are also given along with experimental results pertaining to the suitability of WSN devices in industry. Despite WSN devices being still relatively new many advances have been made in order to make them more powerful and also smaller. However, as the size of the devices has decreased very been done with regards to critical components such as the antenna. As a result this work looks at the production of an industrially suitable antenna in terms of its design, construction and testing. Finally, wireless sensing in the automotive industry is briefly discussed. The apphcation of WSN in the automotive industry aims to improve recent spot weld monitoring techniques which determine the quality and integrity of a spot weld in real-time.

Sensing and site data acquisition are active areas of research for industrial applications, particularly, for construction engineering and management. A number of research initiatives around the globe are focused on noble sensing applications for managing site safety, productivity improvements, progress monitoring, site lay out planning, and for innovative approaches to supply chain management. Time of Flight (TOF) cameras and laser scanners are the tools of choice for real time and near real time decision making on jobsites. However, most of these applications are limited to academic research and limited field trials have been carried out. A number of operational decisions are necessary before sensing equipment can be deployed. These decisions are challenging for companies and researchers alike as there is limited test data available regarding the performance characteristics for the various equipment options. Recently with technological advances, Microsoft introduced a very affordable new TOF sensor, Xbox Kinect for video games. The objective of this research is to investigate potential industrial applications of this device to monitor industrial environment for safety and modeling purposes. First we conducted designed experiments and characterized the accuracy and resolution of Xbox Kinect sensors as well as the interference between multiple Xbox Kinect sensors. The experiments provided quantitative knowledge about the performance of XBOX Kinect sensors in terms of spatial modeling. Second, using the experimental results as the baseline for Kinect performance, we developed optimal placement methods of Kinect sensors to maximize detection area or performance using fuzzy and heuristic algorithms. Third, we investigated the application of Kinect sensors for 3D reconstruction and rapid modeling and reported its performance in comparison with two other methods..

The use of enzymes as industrial catalysts is a very promising alternative to conventional synthetic chemistry since enzymes are very specific, selective and display very high activity under mild experimental conditions. This research is focused on lipases, which are widely used in several industrial sectors and especially in the laundry industry. Enzymes in laundry formulation are exposed to alkaline pH, high concentrations of detergents and the presence of proteases. These harsh conditions have a negative effect on the stability and activity of the enzymes and can eliminate the benefits of adding enzymes to formulation entirely. This study aimed to investigate the stabilisation of existing commercial lipases and the characterisation of a novel cold-adapted lipase for industrial applications. The commercial lipase Lipex 16L is a variant of the Thermomyces lanuginosus lipase and is the current benchmark lipase for application in laundry products. Lipex 16L has been used to develop an improved method for carrier-free immobilisation, using Cross-Linked Enzyme Aggregates (CLEAs). The CLEAs production protocol has been modified by introduction of an activator step to obtain a higher number of individual lipase molecules in the "open lid" conformation and by the introduction of a terminator step to quench the cross-linking reaction at an optimal time to obtain smaller and more homogenous cross-linked particles. This improved immobilisation method has been compared to a commercially available immobilised enzyme and has been shown to be made up of smaller and more homogenous particles with higher activity than the Lipex 16L. The CLEAs produced show improved features for commercial applications such as an enhanced wash performance comparable with the free enzyme, improved stability to proteolysis and a higher activity after long-term storage. The stabilisation of Lipex 16L has also been investigated through the introduction of two additional glycosylation sites on the protein surface. The commercial Lipex 16L has only one glycosylation site on asparagine 33. A tri-glycosylated mutant has been generated with the introduction of two further glycosylation sites on asparagine 37 and asparagine 99. This recombinant enzyme and a mono-glycosylated wild-type enzyme have been cloned and expressed in Pichia pastoris while a non-glycosylated variant has been expressed in Escherichia coli. The enzymatic activities of the glycosylated and non-glycosylated lipases have been compared under various conditions such as temperature, pH, detergents, and incubation with proteases. The results have demonstrated that while the additional glycans do not affect the lipase activity and cleaning performance, they do improve its resistance to proteases and its overall stability with an increase of the melting temperature of + 4 °C. A novel lipase from the psychrophilic bacteria Psychromonas ingrahamii (PinLip) has been biochemically characterised. The enzyme shows activity towards short and medium chain fatty acids and has a good fat stain cleaning performance which makes it attractive for industrial applications. Structural characterisation of the PinLip has been attempted by using crystallisation trials for X-ray crystallography and NMR spectroscopy with limited success. A 3D homology model has been generated using the server I-TASSER using the most closely related known structures, Gibberella zeae, Rhizomucor miehei and Rhizopus microspores lipases, all with a sequence identity to PinLip between 20 and 24%. The different lipases studied in this thesis have been tested for their stability in the presence of traditional laundry formulation ingredients and new novel biosurfactants using differential scanning fluorimetry (DSF). The results have shown an improved stability of all the Lipex variants in presence of mono-rhamnolipids based biosurfactant, while the cold-adapted PinLip was stabilised by a small concentration of a polymer (EPEI) and few other compounds (Tinopal CBS-CL, and Triethylamine). The improved CLEAs method and the use of the PinLip enzyme have been patented (Patent no: WO2017/036901, WO2017/036902, WO2017/036915, WO2017/036916, and WO2017/036917).

La Computational Fluid-Dynamics (CFD) si trova sempre maggiormente coinvolta nello studio di sistemi energetici innovativi. Quindi è logico pensare a filoni di ricerca in cui si sviluppano metodi numerici efficienti, robusti ed accurati per la soluzione di tali problemi. In questa tesi si affronta la soluzione numerica di alcuni problemi di interesse industriale sia con tecniche risolutive standard che innovative. In particolare sono stati sviluppati solutori ad elavato ordine di accuratezza per flussi incomprimibili basati sul metodo ad elementi finiti discontinui di Galerkin (DG). Il metodo DG è basato su approssimazioni polinomiali all’interno del singolo elemento computazionale senza richiesta di continuità globale della soluzione. Recentemente tale metodo sta ricevendo particolare interesse per l’applicazione a problemi di CFD. Partendo da un codice presistente 2D viscoso, basato su un flusso a comprimibilità artificiale, è stata sviluppata una versione 3D che si sta dimostrando capace di essere utilizata come solutore DNS. Quindi è stato aggiunto il modello di turbolenza di Spalart-Allmaras (SA) sia sulla versione 2D che 3D. Vale la pena notare che la soluzione DG delle equazioni RANS è molto complessa per via dell’enorme rigidezza numerica del problema. In questo lavoro viene proposta un’implementazione innovativa del modello SA che modifica opportunamente i termini sorgenti e diffusivi dell’equazione di evoluzione SA quando la variabile di lavoro, o una delle funzioni di chiusura del modello, diventano negative. E’ importante notare che ad oggi in letteratura non sono presenti lavori che trattano della soluzione DG del sistema di equazioni RANS-SA incomprimibili. L’approccio proposto è stato quindi testato su un’ampia gamma di problemi. Parallelamente è stata studiata sia l’Aerodinamica instazionaria dei rotori eolici di tipo Savonius che il campo di moto interno al tubo vortice ad effetto Ranque-Hilsch (RHVT) attraverso solutori standard a volumi finiti. Nonostante durante questo lavoro di tesi siano stati sviluppati solutori DG per un ampio range di numeri di Reynolds, ad oggi essi non sono stati ancora applicati a problemi come il Savonius o RHVT in quanto al tempo di quelle analisi non si disponeva dei codici allo stato di sviluppo attuale.
The study of innovative energy systems often involves complex fluid flows problems and the Computational Fluid Dynamics (CFD) is one of the main tools of analysis. It is very easy to understand as developing new high-accuracy solution techniques for the fluid flow governing equations is of an extreme interesting research area. This work is aimed in the field of numerical solution of turbulent flows problems in industrial configurations with standard and innovative discretization techquines. In this thesis great efforts were addressed in to develop of a high-order Discontinuous Galerkin (DG) solver for incompressible flows in order to enjoy its accuracy in a wide class of industrial problems. DG methods are based on polynomial approximations inside the computational elements with no global continuity requirement and they are receiving an increasing interest in CFD community. features. Starting from a 2D viscous version of a code, based on the artificial compressibility flux DG method [1], in this thesis a 3D version is presented and its suitability for DNS computations is demonstrated. Moreover the Spalart-Allmaras (SA) turbulence model has been implemented in both the 2D and 3D solvers.It is worth noting that DG space discretization of RANS equations is a difficult task due the numerical stiffness of the equations. In this work the SA model is modified in source and diffusion terms to deal with numerical instabilities coming-up when the working variable, or one of the model closure functions, become negative thus unphysical. It is important to remark that in the present literature are not reported others DG solvers for the incompressible RANSSA equations. The realiability, accuracy and robustness of the solution method was assessed computing several test-cases in simple and real-life configurations. Simultaneously unsteady Aerodynamics of the Savonius wind rotor and the flow field inside a Ranque-Hilsch vortex tube (RHVT) were extensively studied with standard finite volume solvers obtaining innovative results. Neverthless in this moment our DG solvers can cover a wide range of Reynolds numbers, they have not still found application to analyze problems as Savonius rotors or RHVT since at the time of those analysis our codes can not deal with that kind of flows.

Fiber lasers can be considered a revolutionary technology in the laser field thanks to their unique properties, such as high efficiency, simplicity, compactness and robustness. These features have allowed in the last ten years their outstanding growth both in scientific and industrial applications, eroding the market share of traditional laser sources like solid-state and gas lasers. Fiber lasers power scaling to the kilowatt range is now well established and, thanks to the fiber confinement, excellent output beam quality can be obtained, with a remarkable benefit for applications. Today, high power laser sources are based on ytterbium doped, large mode area fibers because ytterbium can be efficiently pumped in the range 915nm − 975nm (where pump sources are widely available), generating laser action at 1060nm − 1090nm. With this configuration, impressive power scaling has been demonstrated in the last few years. This Ph.D. thesis has been focused on the design and development of high power fiber lasers for a wide range of industrial applications, like cutting, wending, drilling and micro-machining. Both continuous and pulse wave fiber lasers have been demonstrated and particular attention has been devoted to the development of critical technological de-vices like fused fibers combiner, strategic components either for pump light coupling into the laser active fiber (pump combiner) and for power scaling through the beam combining of several fibers lasers (signal combiner). Ytterbium doped fiber lasers have been developed during the Ph.D activity and, in particular, after a theoretical analysis devoted to the modeling of fiber laser cavities and amplifiers, a continuous wave fiber laser and two pulsed laser systems have been demonstrated. The CW fiber laser has been developed with a modular approach: 7 laser modules, capable of emitting hundreds of Watts each, have been coupled together thanks to a fused fiber combiner. A multi-kilowatts output power has been demonstrated. The photo-darkening effect in the active fiber of the laser modules has also been exper-imentally investigated. The pulsed architectures are instead a Q-switched MOPA and a Seed MOPA fiber lasers. The first system is based on a fiber laser oscillator operating in the Q-switching regime, followed by a power amplifier. This laser is capable of delivering 100ns pulses with 10W average power (2kW maximum peak power). The Seed MOPA consists instead of a current modulated laser diode followed by two amplification stages; 2W output average power with adjustable pulse widths from 10 to 100ns has been demonstrated. In the last part of the activity, a preliminary version of a thulium doped fiber laser emitting at about 2000nm (i.e. in the so-called eye-safe region) has been developed. The laser is a Seed MOPA system that has been tested in cw regime but in the near future the pulsed behavior will be investigated.

Enzymes are proteins necessary to maintain essential biochemical processes taking place in all living organisms. Many of them have found applications in industrial sectors such as biotechnology, production of pharmaceuticals and other fine chemicals. The search for new and efficient biocatalysts requires specialised methodology designed for biochemical characterisation and selection of those with potential applications.

Robots are sophisticated machines which are specially designed to have the capabilities to handle operations, on behalf of human, in many different scenarios. In the past decades, the design of robot systems has been evolving and there are increasing numbers of possible applications of robot. Some systems can even be able to overcome the individual limitations and handle complex problems by combining the strengths of multiple robots. To reduce the risk of human life, robots are now being put into missions under extremely dangerous or hazardous environment where human intervention is not tolerable , such as search-and-rescue missions inside damaged buildings after natural disasters and cleaning up of radioactive materials in nuclear accidence. Even though robots are dispensable, if they are damaged, disabled or trapped, the mission would not be accomplished. Therefore, the longevity of a robot system is always a challenge for robotic operations in such difficult environments. To tackle this challenge, many studies focus on improving the design of individual robot, minimizing the chance of robot failure; or the way that how functioning robots may share the job of the failed robots. The way that how other robots can help failed robots recover, however, has yet to be widely discussed. This thesis studies the feasibility of having multi-robot system with different automatic cooperative recovery abilities on top of its primary functions. A novel cooperative recovery framework is proposed for generic control among system primary functions and recovery behaviours. A number of experiments have been done to study the influence of cooperative recovery on a multi-robot system and how it can affect the system in terms of system performance, sustainability and overhead. An Immunity-based cooperative recovery model has also been created to overcome the drawback introduced by cooperative recovery, finding a balance between the two system objective among system productivity and longevity. Two modified versions of cooperative recovery model are also included in this study to further maximize the system potential.
published_or_final_version
Industrial and Manufacturing Systems Engineering
Doctoral
Doctor of Philosophy

This thesis describes a project developed throughout a six months internship in the Machine Vision Laboratory of Datalogic based in Pasadena, California. The project aims to develop a deep learning system as a possible solution for industrial optical character recognition applications. In particular, the focus falls on a specific algorithm called You Only Look Once (YOLO), which is a general-purpose object detector based on convolutional neural networks that currently offers state-of-the-art performances in terms of trade-off between speed and accuracy. This algorithm is indeed well known for reaching impressive processing speeds, but its intrinsic structure makes it struggle in detecting small objects clustered together, which unfortunately matches our scenario: we are trying to read alphanumerical codes by detecting each single character and then reconstructing the final string. The final goal of this thesis is to overcome this drawback and push the accuracy performances of a general object detector convolutional neural network to its limits, in order to meet the demanding requirements of industrial OCR applications. To accomplish this, first YOLO's unique detecting approach was mastered in its original framework called Darknet, written in C and CUDA, then all the code was translated into Python programming language for a better flexibility, which also allowed the deployment of a custom architecture. Four different datasets with increasing complexity were used as case-studies and the final performances reached were surprising: the accuracy varies between 99.75\% and 99.97\% with a processing time of 15 ms for images $1000\times1000$ big, largely outperforming in speed the current deep learning solution deployed by Datalogic. On the downsides, the training phase usually requires a very large amount of data and time and YOLO also showed some memorization behaviours if not enough variability is given at training time.

Turning and milling are the main manufacturing techniques in industry. A great deal of time and money is spent for machining operations. Although most of the time, the tool path of a CNC machine tool is generated by a Computer Aided Manufacturing (CAM) software package, to be sure that the result of the machining operation will be as required, it is necessary to use a simulation software. There are various machining simulation software packages available in the market. However, they are not only expensive, but also specialized for only one of the before mentioned machining techniques. Most of the companies in the industry are small or medium scale ones and, it is not so easy for them to afford a specialized simulation software for that purpose. In this thesis, it is aimed to develop a software package, which will be used to simulate advanced industrial machining processes, including turning and milling. Dexelmodeling, which is generated by ray casting, and sweep plane algorithm withpolygon clipping technique are used forvisualization. For polygon clipping technique, outer surfaces of the resultant workpiece are generated from planar contours. The software is developed in C# programming language and DirectX libraries are utilizedfor visualization purposes. With the aid of this software, it is also aimed to visually confirm the validity of both mill and lathe NC-code, by representing highly accurate 3D displayed results of these simulations.

This thesis describes research in the area of automated industrial inspection using machine vision systems. It is anticipated that the algorithms described will contribute to the design of a machine vision system for the automatic surface inspection of cylindrical pellets. Firstly, the acquisition and segmentation of pellet tray images using area capture is described. Individual pellets are segmented from a pellet tray image by a novel system using the Radon transform coupled with data clustering. Subsequent to the segmentation, the linking of four pellet views depicting the entire circumferential area of the pellet is described along with a simple technique to compensate for intensity variations brought about by imaging the three-dimensional cylindrical surface of the pellet. The image processing techniques of filtering, edge detection, thresholding and morphology are used in the segmentation of grey level pellet defect images. The grey level pellet images are low-pass filtered and binary images formed using edge detection with thresholding. Binary morphology operators are then used in conjunction with a termination condition based on the number of objects in the image to ensure homogenous defect representations. The problem of overlapping defects is addressed, resulting in a second algorithm using the Radon transform coupled with data clustering. Prior to classification salient features are extracted from a set of synthetic binary defect images to form feature vectors. The novel idea of image object classification using 100% fuzzy inference is described, and results are shown to be superior to results obtained by feature space classifiers. The sub-classification of crack defects is carried out using a heuristic classifier, and the parameterisation of pellet defects is described.

Whether it is a car, a mobile phone, or a computer, we are noticing how automation and production with robots plays an important role in the industry of our modern world. We find it in factories, manufacturing products, automotive cruise control, construction equipment, autopilot on airplanes, and countless other industrial applications.         Automation technology can vary greatly depending on the field of application. On one end, we have systems that are operated by the user and rely fully on human ability. Examples of these are heavy-mobile equipment, remote controlled systems, helicopters, and many more. On the other end, we have autonomous systems that are able to make algorithmic decisions independently of the user.         Society has always envisioned robots with the full capabilities of humans. However, we should envision applications that will help us increase productivity and improve our quality of life through human-robot collaboration. The questions we should be asking are: “What tasks should be automated?'', and “How can we combine the best of both humans and automation?”. This thinking leads to the idea of developing systems with some level of autonomy, where the intelligence is shared between the user and the system. Reasonably, the computerized intelligence and decision making would be designed according to mathematical algorithms and control rules.         This thesis considers these topics and shows the importance of fundamental mathematics and control design to develop automated systems that can execute desired tasks. All of this work is based on some of the most modern concepts in the subjects of robotics and control, which are synthesized by a method known as the Virtual Holonomic Constraints Approach. This method has been useful to tackle some of the most complex problems of nonlinear control, and has enabled the possibility to approach challenging academic and industrial problems. This thesis shows concepts of system modeling, control design, motion analysis, motion planning, and many other interesting subjects, which can be treated effectively through analytical methods. The use of mathematical approaches allows performing computer simulations that also lead to direct practical implementations.