Unbalance identification in large steam turbo-generator unit using a model-based method

Fault identification in industrial machine is a topic of major importance under engineering point of view. In fact, the possibility to identify not only the type, but also the severity and the position of a fault occurred along a shaft-line allows quick maintenance and shorten the downtime. This is really important in the power generation industry where the units are often of several tenths of meters long and where the rotors are enclosed by heavy and pressure-sealed casings. In this paper, an industrial experimental case is presented related to the identification of the unbalance on a large size steam turbine of about 1.3 GW, belonging to a nuclear power plant. The case history is analyzed by considering the vibrations measured by the condition monitoring system of the unit. A model-based method in the frequency domain, developed by the authors, is introduced in detail and it is then used to identify the position of the fault and its severity along the shaft-line. The complete model of the unit (rotor – modeled by means of finite elements, bearings – modeled by linearized damping and stiffness coefficients and foundation – modeled by means of pedestals) is analyzed and discussed before being used for the fault identification. The assessment of the actual fault was done by inspection during a scheduled maintenance and excellent correspondence was found with the identified one by means of authors’ proposed method. Finally a complete discussion is presented about the effectiveness of the method, even in presence of a not fine tuned machine model and considering only few measuring planes for the machine vibration.

An application of the Driver Behaviour Questionnaire in a large Australian organisational fleeting setting : can it predict crashes and demerit point loss?

This study reports on the utilisation of the Manchester Driver Behaviour Questionnaire (DBQ) to examine the self-reported driving behaviours of a large sample of Australian fleet drivers (N = 3414). Surveys were completed by employees before they commenced a one day safety workshop intervention. Factor analysis techniques identified a three factor solution similar to previous research, which was comprised of: (a) errors, (b) highway-code violations and (c) aggressive driving violations. Two items traditionally related with highway-code violations were found to be associated with aggressive driving behaviours among the current sample. Multivariate analyses revealed that exposure to the road, errors and self-reported offences predicted crashes at work in the last 12 months, while gender, highway violations and crashes predicted offences incurred while at work. Importantly, those who received more fines at work were at an increased risk of crashing the work vehicle. However, overall, the DBQ demonstrated limited efficacy at predicting these two outcomes. This paper outlines the major findings of the study in regards to identifying and predicting aberrant driving behaviours and also highlights implications regarding the future utilisation of the DBQ within fleet settings.

The BRAKE Driver Awareness Program & Senior Health Education

The BRAKE Driver Awareness Program provides evidence-based behaviour, risk, attitude and knowledge education for young drivers. BRAKE was founded during 2006 by Queensland Police Sergeant Rob Duncan and has been delivered to more than 35,000 senior secondary students since 2007. BRAKE is a participant directed program supported by resources provided at no cost. It includes eight parts able to be delivered in different configurations. BRAKE is endorsed by the Queensland Police and Queensland Ambulance Services. It is recognised by the Queensland Studies Authority as a Queensland Certificate of Education registered life skills course. This session is a must attend for secondary teachers, coordinators, staff in senior leadership positions and other stakeholders seeking a unique approach to adolescent road safety education. It will conclude with an opportunity to consider how BRAKE can be integrated into the senior secondary Health Education curriculum or pastoral care, social action and personal development programs.

A new method of interfacing battery/supercapacitor energy storage systems for distributed energy sources

Battery-supercapacitor hybrid energy storage systems can achieve better power and energy performances compared to their individual use. These hybrid systems require separate dc-dc converters, or at least one dc-dc converter for the supercapacitor bank, to connect them to the dc-link of the grid connecting inverter. However, the use of such dc-dc converters introduces additional cost and power losses. Therefore, the possibility of direct connection of energy storage systems, to the dc-link of a diode clamped 3-level inverter is investigated in this paper. Even though the proposed topology does not use dc-dc converters, a vector selection method is proposed to produce a similar control flexibility that is found in the separate dc-dc converter topology. The major issue with the proposed system is the imminent imbalance of the neutral point potential. A PWM technique with modified carriers is used to solve this problem. Simulations are carried out using MATLAB/SIMULINK to verify the efficacy of the proposed system.

Further investigations of framing effects on cooperative choices in a provision point mechanism

We investigate whether framing effects of voluntary contributions are significant in a provision point mechanism. Our results show that framing significantly affects individuals of the same type: cooperative individuals appear to be more cooperative in the public bads game than in the public goods game, whereas individualistic subjects appear to be less cooperative in the public bads game than in the public goods game. At the aggregate level of pooling all individuals, the data suggests that framing effects are negligible, which is in contrast with the established result.

Dual inverter based battery energy storage system for grid connected photovoltaic systems

This paper explores a new breed of energy storage system interfacing for grid connected photovoltaic (PV) systems. The proposed system uses the popular dual inverter topology in which one inverter is supplied by a PV cell array and the other by a Battery Energy Storage System (BESS). The resulting conversion structure is controlled in a way that both demand matching and maximum power point tracking of the PV cell array are performed simultaneously. This dual inverter topology can produces 2, 3, 4 and 5 level inverter voltage waveforms at the dc-link voltage ratios of 0:1, 1:1, 2:1 and 3:2 respectively. Since the output voltage of the PV cell array and the battery are uncorrelated and dynamically change, the resulting dc-link voltage ratio can take non-integer values as well. These noninteger dc-link voltage ratios produce unevenly distributed space vectors. Therefore, the main issue with the proposed system is the generation of undistorted current even in the presence of unevenly distributed and dynamically changing space vectors. A modified space vector modulation method is proposed in this paper to address this issue and its efficacy is proved by simulation results. The ability of the proposed system to act as an active power source is also verified.

Bayesian hierarchical models for analysing spatial point-based data at a grid level: A comparison of approaches

Spatial data are now prevalent in a wide range of fields including environmental and health science. This has led to the development of a range of approaches for analysing patterns in these data. In this paper, we compare several Bayesian hierarchical models for analysing point-based data based on the discretization of the study region, resulting in grid-based spatial data. The approaches considered include two parametric models and a semiparametric model. We highlight the methodology and computation for each approach. Two simulation studies are undertaken to compare the performance of these models for various structures of simulated point-based data which resemble environmental data. A case study of a real dataset is also conducted to demonstrate a practical application of the modelling approaches. Goodness-of-fit statistics are computed to compare estimates of the intensity functions. The deviance information criterion is also considered as an alternative model evaluation criterion. The results suggest that the adaptive Gaussian Markov random field model performs well for highly sparse point-based data where there are large variations or clustering across the space; whereas the discretized log Gaussian Cox process produces good fit in dense and clustered point-based data. One should generally consider the nature and structure of the point-based data in order to choose the appropriate method in modelling a discretized spatial point-based data.

Sing fox to me : an investigation into the "use" of Down Syndrome in both the Down Syndrome and Gothic novel

This project investigates the current borders around and within, what I have in this exegesis termed, "the Down Syndrome novel", using a close reading analysis of literary texts containing characters with Down syndrome and contextualised by theoretical works drawn from both disability and literary theory. This practice-led thesis introduces and discusses select fictional characters with Down syndrome from numerous genres, revealing them as highly contained, or "boundaried", spoken for, and generally used for narrative conflict rather than included as individuals with agency and a legitimate, autonomous voice and narrative point of view. In reframing the Australian landscape as "disabled" this exegesis illustrates that the Australian Gothic novel can shift, and sometimes even remove, the boundary around characters with intellectual disabilities, allowing a space where the stories of characters with Down syndrome can emerge.

Scene signatures : localised and point-less features for localisation

This paper is about localising across extreme lighting and weather conditions. We depart from the traditional point-feature-based approach as matching under dramatic appearance changes is a brittle and hard thing. Point feature detectors are fixed and rigid procedures which pass over an image examining small, low-level structure such as corners or blobs. They apply the same criteria applied all images of all places. This paper takes a contrary view and asks what is possible if instead we learn a bespoke detector for every place. Our localisation task then turns into curating a large bank of spatially indexed detectors and we show that this yields vastly superior performance in terms of robustness in exchange for a reduced but tolerable metric precision. We present an unsupervised system that produces broad-region detectors for distinctive visual elements, called scene signatures, which can be associated across almost all appearance changes. We show, using 21km of data collected over a period of 3 months, that our system is capable of producing metric localisation estimates from night-to-day or summer-to-winter conditions.

In pursuit of an architecture realism

The noble idea of studying seminal works to ‘see what we can learn’ has turned in the 1990s into ‘let’s see what we can take’ and in the last decade a more toxic derivative ‘what else can’t we take’. That is my observation as a student of architecture in the 1990s, and as a practitioner in the 2000s. In 2010, the sense that something is ending is clear. The next generation is rising and their gaze has shifted. The idea of classification (as a means of separation) was previously rejected by a generation of Postmodernists; the usefulness of difference declined. It’s there in the presence of plurality in the resulting architecture, a decision to mine history and seize in a willful manner. This is a process of looking back but never forward. It has been a mono-culture of absorption. The mono-culture rejected the pursuit of the realistic. It is a blanket suffocating all practice of architecture in this country from the mercantile to the intellectual. Independent reviews of Australia’s recent contributions to the Venice Architecture Biennales confirm the malaise. The next generation is beginning to reconsider classification as a means of unification. By acknowledging the characteristics of competing forces it is possible to bring them into a state of tension. Seeking a beautiful contrast is a means to a new end. In the political setting, this is described by Noel Pearson as the radical centre[1]. The concept transcends the political and in its most essential form is a cultural phenomenon. It resists the compromised position and suggests that we can look back while looking forward. The radical centre is the only demonstrated opportunity where it is possible to pursue a realistic architecture. A realistic architecture in Australia may be partially resolved by addressing our anxiety of permanence. Farrelly’s built desires[2] and Markham’s ritual demonstrations[3] are two ways into understanding the broader spectrum of permanence. But I think they are downstream of our core problem. Our problem, as architects, is that we are yet to come to terms with this place. Some call it landscape others call it country. Australian cities were laid out on what was mistaken for a blank canvas. On some occasions there was the consideration of the landscape when it presented insurmountable physical obstacles. The architecture since has continued to work on its piece of a constantly blank canvas. Even more ironic is the commercial awards programs that represent a claim within this framework but at best can only establish a dialogue within itself. This is a closed system unable to look forward. It is said that Melbourne is the most European city in the southern hemisphere but what is really being described there is the limitation of a senseless grid. After all, if Dutch landscape informs Dutch architecture why can’t the Australian landscape inform Australian architecture? To do that, we would have to acknowledge our moribund grasp of the meaning of the Australian landscape. Or more precisely what Indigenes call Country[4]. This is a complex notion and there are different ways into it. Country is experienced and understood through the senses and seared into memory. If one begins design at that starting point it is not unreasonable to think we can arrive at an end point that is a counter trajectory to where we have taken ourselves. A recent studio with Masters students confirmed this. Start by finding Country and it would be impossible to end up with a building looking like an Aboriginal man’s face. To date architecture in Australia has overwhelmingly ignored Country on the back of terra nullius. It can’t seem to get past the picturesque. Why is it so hard? The art world came to terms with this challenge, so too did the legal establishment, even the political scene headed into new waters. It would be easy to blame the budgets of commerce or the constraints of program or even the pressure of success. But that is too easy. Those factors are in fact the kind of limitations that opportunities grow out of. The past decade of economic plenty has, for the most part, smothered the idea that our capitals might enable civic settings or an architecture that is able to looks past lot line boundaries in a dignified manner. The denied opportunities of these settings to be prompted by the Country they occupy is criminal. The public realm is arrested in its development because we refuse to accept Country as a spatial condition. What we seem to be able to embrace is literal and symbolic gestures usually taking the form of a trumped up art installations. All talk – no action. To continue to leave the public realm to the stewardship of mercantile interests is like embracing derivative lending after the global financial crisis.Herein rests an argument for why we need a resourced Government Architect’s office operating not as an isolated lobbyist for business but as a steward of the public realm for both the past and the future. New South Wales is the leading model with Queensland close behind. That is not to say both do not have flaws but current calls for their cessation on the grounds of design parity poorly mask commercial self interest. In Queensland, lobbyists are heavily regulated now with an aim to ensure integrity and accountability. In essence, what I am speaking of will not be found in Reconciliation Action Plans that double as business plans, or the mining of Aboriginal culture for the next marketing gimmick, or even discussions around how to make buildings more ‘Aboriginal’. It will come from the next generation who reject the noxious mono-culture of absorption and embrace a counter trajectory to pursue an architecture of realism.

Using the Driver Behaviour Questionnaire to predict crashes and demerit point loss : does it get better with larger sample sizes?

The Driver Behaviour Questionnaire (DBQ) continues to be the most widely utilised self-report scale globally to assess crash risk and aberrant driving behaviours among motorists. However, the scale also attracts criticism regarding its perceived limited ability to accurately identify those most at risk of crash involvement. This study reports on the utilisation of the DBQ to examine the self-reported driving behaviours (and crash outcomes) of drivers in three separate Australian fleet samples (N = 443, N = 3414, & N = 4792), and whether combining the samples increases the tool’s predictive ability. Either on-line or paper versions of the questionnaire were completed by fleet employees in three organisations. Factor analytic techniques identified either three or four factor solutions (in each of the separate studies) and the combined sample produced expected factors of: (a) errors, (b) highway-code violations and (c) aggressive driving violations. Highway code violations (and mean scores) were comparable across the studies. However, across the three samples, multivariate analyses revealed that exposure to the road was the best predictor of crash involvement at work, rather than DBQ constructs. Furthermore, combining the scores to produce a sample of 8649 drivers did not improve the predictive ability of the tool for identifying crashes (e.g., 0.4% correctly identified) or for demerit point loss (0.3%). The paper outlines the major findings of this comparative sample study in regards to utilising self-report measurement tools to identify “at risk” drivers as well as the application of such data to future research endeavours.

Improving national hospice/palliative care service symptom outcomes systematically through point-of-care data collection, structured feedback and benchmarking

PURPOSE Every health care sector including hospice/palliative care needs to systematically improve services using patient-defined outcomes. Data from the national Australian Palliative Care Outcomes Collaboration aims to define whether hospice/palliative care patients' outcomes and the consistency of these outcomes have improved in the last 3 years. METHODS Data were analysed by clinical phase (stable, unstable, deteriorating, terminal). Patient-level data included the Symptom Assessment Scale and the Palliative Care Problem Severity Score. Nationally collected point-of-care data were anchored for the period July-December 2008 and subsequently compared to this baseline in six 6-month reporting cycles for all services that submitted data in every time period (n = 30) using individual longitudinal multi-level random coefficient models. RESULTS Data were analysed for 19,747 patients (46 % female; 85 % cancer; 27,928 episodes of care; 65,463 phases). There were significant improvements across all domains (symptom control, family care, psychological and spiritual care) except pain. Simultaneously, the interquartile ranges decreased, jointly indicating that better and more consistent patient outcomes were being achieved. CONCLUSION These are the first national hospice/palliative care symptom control performance data to demonstrate improvements in clinical outcomes at a service level as a result of routine data collection and systematic feedback.

Factors influencing robustness and effectiveness of conditional random fields in active learning frameworks

Active learning approaches reduce the annotation cost required by traditional supervised approaches to reach the same effectiveness by actively selecting informative instances during the learning phase. However, effectiveness and robustness of the learnt models are influenced by a number of factors. In this paper we investigate the factors that affect the effectiveness, more specifically in terms of stability and robustness, of active learning models built using conditional random fields (CRFs) for information extraction applications. Stability, defined as a small variation of performance when small variation of the training data or a small variation of the parameters occur, is a major issue for machine learning models, but even more so in the active learning framework which aims to minimise the amount of training data required. The factors we investigate are a) the choice of incremental vs. standard active learning, b) the feature set used as a representation of the text (i.e., morphological features, syntactic features, or semantic features) and c) Gaussian prior variance as one of the important CRFs parameters. Our empirical findings show that incremental learning and the Gaussian prior variance lead to more stable and robust models across iterations. Our study also demonstrates that orthographical, morphological and contextual features as a group of basic features play an important role in learning effective models across all iterations.