Neuartige Verfahren für die Überwachung und Diagnose von aktiv magnetgelagerten rotierenden Maschinen

Mit aktiven Magnetlagern ist es möglich, rotierende Körper durch magnetische Felder berührungsfrei zu lagern. Systembedingt sind bei aktiv magnetgelagerten Maschinen wesentliche Signale ohne zusätzlichen Aufwand an Messtechnik für Diagnoseaufgaben verfügbar. In der Arbeit wird ein Konzept entwickelt, das durch Verwendung der systeminhärenten Signale eine Diagnose magnetgelagerter rotierender Maschinen ermöglicht und somit neben einer kontinuierlichen Anlagenüberwachung eine schnelle Bewertung des Anlagenzustandes gestattet. Fehler können rechtzeitig und ursächlich in Art und Größe erkannt und entsprechende Gegenmaßnahmen eingeleitet werden. Anhand der erfassten Signale geschieht die Gewinnung von Merkmalen mit signal- und modellgestützten Verfahren. Für den Magnetlagerregelkreis erfolgen Untersuchungen zum Einsatz modellgestützter Parameteridentifikationsverfahren, deren Verwendbarkeit wird bei der Diagnose am Regler und Leistungsverstärker nachgewiesen. Unter Nutzung von Simulationsmodellen sowie durch Experimente an Versuchsständen werden die Merkmalsverläufe im normalen Referenzzustand und bei auftretenden Fehlern aufgenommen und die Ergebnisse in einer Wissensbasis abgelegt. Diese dient als Grundlage zur Festlegung von Grenzwerten und Regeln für die Überwachung des Systems und zur Erstellung wissensbasierter Diagnosemodelle. Bei der Überwachung werden die Merkmalsausprägungen auf das Überschreiten von Grenzwerten überprüft, Informationen über erkannte Fehler und Betriebszustände gebildet sowie gegebenenfalls Alarmmeldungen ausgegeben. Sich langsam anbahnende Fehler können durch die Berechnung der Merkmalstrends mit Hilfe der Regressionsanalyse erkannt werden. Über die bisher bei aktiven Magnetlagern übliche Überwachung von Grenzwerten hinaus erfolgt bei der Fehlerdiagnose eine Verknüpfung der extrahierten Merkmale zur Identifizierung und Lokalisierung auftretender Fehler. Die Diagnose geschieht mittels regelbasierter Fuzzy-Logik, dies gestattet die Einbeziehung von linguistischen Aussagen in Form von Expertenwissen sowie die Berücksichtigung von Unbestimmtheiten und ermöglicht damit eine Diagnose komplexer Systeme. Für Aktor-, Sensor- und Reglerfehler im Magnetlagerregelkreis sowie Fehler durch externe Kräfte und Unwuchten werden Diagnosemodelle erstellt und verifiziert. Es erfolgt der Nachweis, dass das entwickelte Diagnosekonzept mit beherrschbarem Rechenaufwand korrekte Diagnoseaussagen liefert. Durch Kaskadierung von Fuzzy-Logik-Modulen wird die Transparenz des Regelwerks gewahrt und die Abarbeitung der Regeln optimiert. Endresultat ist ein neuartiges hybrides Diagnosekonzept, welches signal- und modellgestützte Verfahren der Merkmalsgewinnung mit wissensbasierten Methoden der Fehlerdiagnose kombiniert. Das entwickelte Diagnosekonzept ist für die Anpassung an unterschiedliche Anforderungen und Anwendungen bei rotierenden Maschinen konzipiert.

Improving sequence-based fold recognition by using 3D model quality assessment

Motivation: The ability of a simple method (MODCHECK) to determine the sequence–structure compatibility of a set of structural models generated by fold recognition is tested in a thorough benchmark analysis. Four Model Quality Assessment Programs (MQAPs) were tested on 188 targets from the latest LiveBench-9 automated structure evaluation experiment. We systematically test and evaluate whether the MQAP methods can successfully detect native-likemodels. Results: We show that compared with the other three methods tested MODCHECK is the most reliable method for consistently performing the best top model selection and for ranking the models. In addition, we show that the choice of model similarity score used to assess a model's similarity to the experimental structure can influence the overall performance of these tools. Although these MQAP methods fail to improve the model selection performance for methods that already incorporate protein three dimension (3D) structural information, an improvement is observed for methods that are purely sequence-based, including the best profile–profile methods. This suggests that even the best sequence-based fold recognition methods can still be improved by taking into account the 3D structural information.

A modified rat model of neonatal anoxia: Development and evaluation by pulseoximetry, arterial gasometry and Fos immunoreactivity

Neonatal anoxia is a worldwide clinical problem that has serious and lasting consequences. The diversity of models does not allow complete reproducibility, so a standardized model is needed. In this study, we developed a rat model of neonatal anoxia that utilizes a semi-hermetic system suitable for oxygen deprivation. The validity of this model was confirmed using pulse oximetry, arterial gasometry, observation of skin color and behavior and analysis of Fos immunoreactivity in brain regions that function in respiratory control. For these experiments, 87 male albino neonate rats (Rattus norvegicus, lineage Wistar) aged approximate 30 postnatal hours were divided into anoxia and control groups. The pups were kept in an euthanasia polycarbonate chamber at 36 +/- 1 degrees C, with continuous 100% nitrogen gas flow at 3 L/min and 101.7 kPa for 25 min. The peripheral arterial oxygen saturation of the anoxia group decreased 75% from its initial value. Decreased pH and partial pressure of oxygen and increased partial pressure of carbon dioxide were observed in this group, indicating metabolic acidosis, hypoxia and hypercapnia. respectively. Analysis of neuronal activation showed Fos immunoreactivity in the solitary tract nucleus, the lateral reticular nucleus and the area postrema, confirming that those conditions activated areas related to respiratory control in the nervous system. Therefore, the proposed model of neonatal anoxia allows standardization and precise control of the anoxic condition, which should be of great value in indentifying both the mechanisms underlying neonatal anoxia and novel therapeutic strategies to combat or prevent this widespread public health problem. (C) 2011 Elsevier B.V. All rights reserved.

A new class of survival regression models with heavy-tailed errors: robustness and diagnostics

Birnbaum-Saunders models have largely been applied in material fatigue studies and reliability analyses to relate the total time until failure with some type of cumulative damage. In many problems related to the medical field, such as chronic cardiac diseases and different types of cancer, a cumulative damage caused by several risk factors might cause some degradation that leads to a fatigue process. In these cases, BS models can be suitable for describing the propagation lifetime. However, since the cumulative damage is assumed to be normally distributed in the BS distribution, the parameter estimates from this model can be sensitive to outlying observations. In order to attenuate this influence, we present in this paper BS models, in which a Student-t distribution is assumed to explain the cumulative damage. In particular, we show that the maximum likelihood estimates of the Student-t log-BS models attribute smaller weights to outlying observations, which produce robust parameter estimates. Also, some inferential results are presented. In addition, based on local influence and deviance component and martingale-type residuals, a diagnostics analysis is derived. Finally, a motivating example from the medical field is analyzed using log-BS regression models. Since the parameter estimates appear to be very sensitive to outlying and influential observations, the Student-t log-BS regression model should attenuate such influences. The model checking methodologies developed in this paper are used to compare the fitted models.

Neuronal Assembly Detection and Cell Membership Specification by Principal Component Analysis

LOPES-DOS-SANTOS, V. , CONDE-OCAZIONEZ, S. ; NICOLELIS, M. A. L. , RIBEIRO, S. T. , TORT, A. B. L. . Neuronal assembly detection and cell membership specification by principal component analysis. Plos One, v. 6, p. e20996, 2011.

Estendendo CRefine para o suporte de táticas de refinamento

The use of increasingly complex software applications is demanding greater investment in the development of such systems to ensure applications with better quality. Therefore, new techniques are being used in Software Engineering, thus making the development process more effective. Among these new approaches, we highlight Formal Methods, which use formal languages that are strongly based on mathematics and have a well-defined semantics and syntax. One of these languages is Circus, which can be used to model concurrent systems. It was developed from the union of concepts from two other specification languages: Z, which specifies systems with complex data, and CSP, which is normally used to model concurrent systems. Circus has an associated refinement calculus, which can be used to develop software in a precise and stepwise fashion. Each step is justified by the application of a refinement law (possibly with the discharge of proof obligations). Sometimes, the same laws can be applied in the same manner in different developments or even in different parts of a single development. A strategy to optimize this calculus is to formalise these application as a refinement tactic, which can then be used as a single transformation rule. CRefine was developed to support the Circus refinement calculus. However, before the work presented here, it did not provide support for refinement tactics. The aim of this work is to provide tool support for refinement tactics. For that, we develop a new module in CRefine, which automates the process of defining and applying refinement tactics that are formalised in the tactic language ArcAngelC. Finally, we validate the extension by applying the new module in a case study, which used the refinement tactics in a refinement strategy for verification of SPARK Ada implementations of control systems. In this work, we apply our module in the first two phases of this strategy

Particle Competition and Cooperation in Networks for Semi-Supervised Learning

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Parameter recovery for a skew-normal IRT model under a Bayesian approach: hierarchical framework, prior and kernel sensitivity and sample size

Item response theory (IRT) comprises a set of statistical models which are useful in many fields, especially when there is an interest in studying latent variables (or latent traits). Usually such latent traits are assumed to be random variables and a convenient distribution is assigned to them. A very common choice for such a distribution has been the standard normal. Recently, Azevedo et al. [Bayesian inference for a skew-normal IRT model under the centred parameterization, Comput. Stat. Data Anal. 55 (2011), pp. 353-365] proposed a skew-normal distribution under the centred parameterization (SNCP) as had been studied in [R. B. Arellano-Valle and A. Azzalini, The centred parametrization for the multivariate skew-normal distribution, J. Multivariate Anal. 99(7) (2008), pp. 1362-1382], to model the latent trait distribution. This approach allows one to represent any asymmetric behaviour concerning the latent trait distribution. Also, they developed a Metropolis-Hastings within the Gibbs sampling (MHWGS) algorithm based on the density of the SNCP. They showed that the algorithm recovers all parameters properly. Their results indicated that, in the presence of asymmetry, the proposed model and the estimation algorithm perform better than the usual model and estimation methods. Our main goal in this paper is to propose another type of MHWGS algorithm based on a stochastic representation (hierarchical structure) of the SNCP studied in [N. Henze, A probabilistic representation of the skew-normal distribution, Scand. J. Statist. 13 (1986), pp. 271-275]. Our algorithm has only one Metropolis-Hastings step, in opposition to the algorithm developed by Azevedo et al., which has two such steps. This not only makes the implementation easier but also reduces the number of proposal densities to be used, which can be a problem in the implementation of MHWGS algorithms, as can be seen in [R.J. Patz and B.W. Junker, A straightforward approach to Markov Chain Monte Carlo methods for item response models, J. Educ. Behav. Stat. 24(2) (1999), pp. 146-178; R. J. Patz and B. W. Junker, The applications and extensions of MCMC in IRT: Multiple item types, missing data, and rated responses, J. Educ. Behav. Stat. 24(4) (1999), pp. 342-366; A. Gelman, G.O. Roberts, and W.R. Gilks, Efficient Metropolis jumping rules, Bayesian Stat. 5 (1996), pp. 599-607]. Moreover, we consider a modified beta prior (which generalizes the one considered in [3]) and a Jeffreys prior for the asymmetry parameter. Furthermore, we study the sensitivity of such priors as well as the use of different kernel densities for this parameter. Finally, we assess the impact of the number of examinees, number of items and the asymmetry level on the parameter recovery. Results of the simulation study indicated that our approach performed equally as well as that in [3], in terms of parameter recovery, mainly using the Jeffreys prior. Also, they indicated that the asymmetry level has the highest impact on parameter recovery, even though it is relatively small. A real data analysis is considered jointly with the development of model fitting assessment tools. The results are compared with the ones obtained by Azevedo et al. The results indicate that using the hierarchical approach allows us to implement MCMC algorithms more easily, it facilitates diagnosis of the convergence and also it can be very useful to fit more complex skew IRT models.

Ingegneria di sistemi auto‐organizzanti con il paradigma multiagente

Self-organisation is increasingly being regarded as an effective approach to tackle modern systems complexity. The self-organisation approach allows the development of systems exhibiting complex dynamics and adapting to environmental perturbations without requiring a complete knowledge of the future surrounding conditions. However, the development of self-organising systems (SOS) is driven by different principles with respect to traditional software engineering. For instance, engineers typically design systems combining smaller elements where the composition rules depend on the reference paradigm, but typically produce predictable results. Conversely, SOS display non-linear dynamics, which can hardly be captured by deterministic models, and, although robust with respect to external perturbations, are quite sensitive to changes on inner working parameters. In this thesis, we describe methodological aspects concerning the early-design stage of SOS built relying on the Multiagent paradigm: in particular, we refer to the A&A metamodel, where MAS are composed by agents and artefacts, i.e. environmental resources. Then, we describe an architectural pattern that has been extracted from a recurrent solution in designing self-organising systems: this pattern is based on a MAS environment formed by artefacts, modelling non-proactive resources, and environmental agents acting on artefacts so as to enable self-organising mechanisms. In this context, we propose a scientific approach for the early design stage of the engineering of self-organising systems: the process is an iterative one and each cycle is articulated in four stages, modelling, simulation, formal verification, and tuning. During the modelling phase we mainly rely on the existence of a self-organising strategy observed in Nature and, hopefully encoded as a design pattern. Simulations of an abstract system model are used to drive design choices until the required quality properties are obtained, thus providing guarantees that the subsequent design steps would lead to a correct implementation. However, system analysis exclusively based on simulation results does not provide sound guarantees for the engineering of complex systems: to this purpose, we envision the application of formal verification techniques, specifically model checking, in order to exactly characterise the system behaviours. During the tuning stage parameters are tweaked in order to meet the target global dynamics and feasibility constraints. In order to evaluate the methodology, we analysed several systems: in this thesis, we only describe three of them, i.e. the most representative ones for each of the three years of PhD course. We analyse each case study using the presented method, and describe the exploited formal tools and techniques.

Self-organising coordination systems

Many research fields are pushing the engineering of large-scale, mobile, and open systems towards the adoption of techniques inspired by self-organisation: pervasive computing, but also distributed artificial intelligence, multi-agent systems, social networks, peer-topeer and grid architectures exploit adaptive techniques to make global system properties emerge in spite of the unpredictability of interactions and behaviour. Such a trend is visible also in coordination models and languages, whenever a coordination infrastructure needs to cope with managing interactions in highly dynamic and unpredictable environments. As a consequence, self-organisation can be regarded as a feasible metaphor to define a radically new conceptual coordination framework. The resulting framework defines a novel coordination paradigm, called self-organising coordination, based on the idea of spreading coordination media over the network, and charge them with services to manage interactions based on local criteria, resulting in the emergence of desired and fruitful global coordination properties of the system. Features like topology, locality, time-reactiveness, and stochastic behaviour play a key role in both the definition of such a conceptual framework and the consequent development of self-organising coordination services. According to this framework, the thesis presents several self-organising coordination techniques developed during the PhD course, mainly concerning data distribution in tuplespace-based coordination systems. Some of these techniques have been also implemented in ReSpecT, a coordination language for tuple spaces, based on logic tuples and reactions to events occurring in a tuple space. In addition, the key role played by simulation and formal verification has been investigated, leading to analysing how automatic verification techniques like probabilistic model checking can be exploited in order to formally prove the emergence of desired behaviours when dealing with coordination approaches based on self-organisation. To this end, a concrete case study is presented and discussed.

Development of advanced tools and methods for the assessment and management of risk due to atypical major accident scenarios

Proper hazard identification has become progressively more difficult to achieve, as witnessed by several major accidents that took place in Europe, such as the Ammonium Nitrate explosion at Toulouse (2001) and the vapour cloud explosion at Buncefield (2005), whose accident scenarios were not considered by their site safety case. Furthermore, the rapid renewal in the industrial technology has brought about the need to upgrade hazard identification methodologies. Accident scenarios of emerging technologies, which are not still properly identified, may remain unidentified until they take place for the first time. The consideration of atypical scenarios deviating from normal expectations of unwanted events or worst case reference scenarios is thus extremely challenging. A specific method named Dynamic Procedure for Atypical Scenarios Identification (DyPASI) was developed as a complementary tool to bow-tie identification techniques. The main aim of the methodology is to provide an easier but comprehensive hazard identification of the industrial process analysed, by systematizing information from early signals of risk related to past events, near misses and inherent studies. DyPASI was validated on the two examples of new and emerging technologies: Liquefied Natural Gas regasification and Carbon Capture and Storage. The study broadened the knowledge on the related emerging risks and, at the same time, demonstrated that DyPASI is a valuable tool to obtain a complete and updated overview of potential hazards. Moreover, in order to tackle underlying accident causes of atypical events, three methods for the development of early warning indicators were assessed: the Resilience-based Early Warning Indicator (REWI) method, the Dual Assurance method and the Emerging Risk Key Performance Indicator method. REWI was found to be the most complementary and effective of the three, demonstrating that its synergy with DyPASI would be an adequate strategy to improve hazard identification methodologies towards the capture of atypical accident scenarios.

Development of Original Analytical Methods for the Determination of Drugs of Abuse in Biological Samples

Drug abuse is a major global problem which has a strong impact not only on the single individual but also on the entire society. Among the different strategies that can be used to address this issue an important role is played by identification of abusers and proper medical treatment. This kind of therapy should be carefully monitored in order to discourage improper use of the medication and to tailor the dose according to the specific needs of the patient. Hence, reliable analytical methods are needed to reveal drug intake and to support physicians in the pharmacological management of drug dependence. In the present Ph.D. thesis original analytical methods for the determination of drugs with a potential for abuse and of substances used in the pharmacological treatment of drug addiction are presented. In particular, the work has been focused on the analysis of ketamine, naloxone and long-acting opioids (buprenorphine and methadone), oxycodone, disulfiram and bupropion in human plasma and in dried blood spots. The developed methods are based on the use of high performance liquid chromatography (HPLC) coupled to various kinds of detectors (mass spectrometer, coulometric detector, diode array detector). For biological sample pre-treatment different techniques have been exploited, namely solid phase extraction and microextraction by packed sorbent. All the presented methods have been validated according to official guidelines with good results and some of these have been successfully applied to the therapeutic drug monitoring of patients under treatment for drug abuse.

Model reduction techniques in flexible multibody dynamics with application to engine cranktrain simulation

The development of a multibody model of a motorbike engine cranktrain is presented in this work, with an emphasis on flexible component model reduction. A modelling methodology based upon the adoption of non-ideal joints at interface locations, and the inclusion of component flexibility, is developed: both are necessary tasks if one wants to capture dynamic effects which arise in lightweight, high-speed applications. With regard to the first topic, both a ball bearing model and a journal bearing model are implemented, in order to properly capture the dynamic effects of the main connections in the system: angular contact ball bearings are modelled according to a five-DOF nonlinear scheme in order to grasp the crankshaft main bearings behaviour, while an impedance-based hydrodynamic bearing model is implemented providing an enhanced operation prediction at the conrod big end locations. Concerning the second matter, flexible models of the crankshaft and the connecting rod are produced. The well-established Craig-Bampton reduction technique is adopted as a general framework to obtain reduced model representations which are suitable for the subsequent multibody analyses. A particular component mode selection procedure is implemented, based on the concept of Effective Interface Mass, allowing an assessment of the accuracy of the reduced models prior to the nonlinear simulation phase. In addition, a procedure to alleviate the effects of modal truncation, based on the Modal Truncation Augmentation approach, is developed. In order to assess the performances of the proposed modal reduction schemes, numerical tests are performed onto the crankshaft and the conrod models in both frequency and modal domains. A multibody model of the cranktrain is eventually assembled and simulated using a commercial software. Numerical results are presented, demonstrating the effectiveness of the implemented flexible model reduction techniques. The advantages over the conventional frequency-based truncation approach are discussed.

Discrete Event Systems based Design Patterns for Diagnosability Analysis of Automated Manufacturing Systems

The main goal of this thesis is to facilitate the process of industrial automated systems development applying formal methods to ensure the reliability of systems. A new formulation of distributed diagnosability problem in terms of Discrete Event Systems theory and automata framework is presented, which is then used to enforce the desired property of the system, rather then just verifying it. This approach tackles the state explosion problem with modeling patterns and new algorithms, aimed for verification of diagnosability property in the context of the distributed diagnosability problem. The concepts are validated with a newly developed software tool.

Development of a Bayesian Joint Logistic Model to Better Study the Association between Haplotypes and Disease

In 2011, there will be an estimated 1,596,670 new cancer cases and 571,950 cancer-related deaths in the US. With the ever-increasing applications of cancer genetics in epidemiology, there is great potential to identify genetic risk factors that would help identify individuals with increased genetic susceptibility to cancer, which could be used to develop interventions or targeted therapies that could hopefully reduce cancer risk and mortality. In this dissertation, I propose to develop a new statistical method to evaluate the role of haplotypes in cancer susceptibility and development. This model will be flexible enough to handle not only haplotypes of any size, but also a variety of covariates. I will then apply this method to three cancer-related data sets (Hodgkin Disease, Glioma, and Lung Cancer). I hypothesize that there is substantial improvement in the estimation of association between haplotypes and disease, with the use of a Bayesian mathematical method to infer haplotypes that uses prior information from known genetics sources. Analysis based on haplotypes using information from publically available genetic sources generally show increased odds ratios and smaller p-values in both the Hodgkin, Glioma, and Lung data sets. For instance, the Bayesian Joint Logistic Model (BJLM) inferred haplotype TC had a substantially higher estimated effect size (OR=12.16, 95% CI = 2.47-90.1 vs. 9.24, 95% CI = 1.81-47.2) and more significant p-value (0.00044 vs. 0.008) for Hodgkin Disease compared to a traditional logistic regression approach. Also, the effect sizes of haplotypes modeled with recessive genetic effects were higher (and had more significant p-values) when analyzed with the BJLM. Full genetic models with haplotype information developed with the BJLM resulted in significantly higher discriminatory power and a significantly higher Net Reclassification Index compared to those developed with haplo.stats for lung cancer. Future analysis for this work could be to incorporate the 1000 Genomes project, which offers a larger selection of SNPs can be incorporated into the information from known genetic sources as well. Other future analysis include testing non-binary outcomes, like the levels of biomarkers that are present in lung cancer (NNK), and extending this analysis to full GWAS studies.