Anomalous event detection using a semi-two dimensional hidden Markov model

The rapid increase in the deployment of CCTV systems has led to a greater demand for algorithms that are able to process incoming video feeds. These algorithms are designed to extract information of interest for human operators. During the past several years, there has been a large effort to detect abnormal activities through computer vision techniques. Typically, the problem is formulated as a novelty detection task where the system is trained on normal data and is required to detect events which do not fit the learned `normal' model. Many researchers have tried various sets of features to train different learning models to detect abnormal behaviour in video footage. In this work we propose using a Semi-2D Hidden Markov Model (HMM) to model the normal activities of people. The outliers of the model with insufficient likelihood are identified as abnormal activities. Our Semi-2D HMM is designed to model both the temporal and spatial causalities of the crowd behaviour by assuming the current state of the Hidden Markov Model depends not only on the previous state in the temporal direction, but also on the previous states of the adjacent spatial locations. Two different HMMs are trained to model both the vertical and horizontal spatial causal information. Location features, flow features and optical flow textures are used as the features for the model. The proposed approach is evaluated using the publicly available UCSD datasets and we demonstrate improved performance compared to other state of the art methods.

Asset health prediction using the explicit hazard model

The ability to estimate the asset reliability and the probability of failure is critical to reducing maintenance costs, operation downtime, and safety hazards. Predicting the survival time and the probability of failure in future time is an indispensable requirement in prognostics and asset health management. In traditional reliability models, the lifetime of an asset is estimated using failure event data, alone; however, statistically sufficient failure event data are often difficult to attain in real-life situations due to poor data management, effective preventive maintenance, and the small population of identical assets in use. Condition indicators and operating environment indicators are two types of covariate data that are normally obtained in addition to failure event and suspended data. These data contain significant information about the state and health of an asset. Condition indicators reflect the level of degradation of assets while operating environment indicators accelerate or decelerate the lifetime of assets. When these data are available, an alternative approach to the traditional reliability analysis is the modelling of condition indicators and operating environment indicators and their failure-generating mechanisms using a covariate-based hazard model. The literature review indicates that a number of covariate-based hazard models have been developed. All of these existing covariate-based hazard models were developed based on the principle theory of the Proportional Hazard Model (PHM). However, most of these models have not attracted much attention in the field of machinery prognostics. Moreover, due to the prominence of PHM, attempts at developing alternative models, to some extent, have been stifled, although a number of alternative models to PHM have been suggested. The existing covariate-based hazard models neglect to fully utilise three types of asset health information (including failure event data (i.e. observed and/or suspended), condition data, and operating environment data) into a model to have more effective hazard and reliability predictions. In addition, current research shows that condition indicators and operating environment indicators have different characteristics and they are non-homogeneous covariate data. Condition indicators act as response variables (or dependent variables) whereas operating environment indicators act as explanatory variables (or independent variables). However, these non-homogenous covariate data were modelled in the same way for hazard prediction in the existing covariate-based hazard models. The related and yet more imperative question is how both of these indicators should be effectively modelled and integrated into the covariate-based hazard model. This work presents a new approach for addressing the aforementioned challenges. The new covariate-based hazard model, which termed as Explicit Hazard Model (EHM), explicitly and effectively incorporates all three available asset health information into the modelling of hazard and reliability predictions and also drives the relationship between actual asset health and condition measurements as well as operating environment measurements. The theoretical development of the model and its parameter estimation method are demonstrated in this work. EHM assumes that the baseline hazard is a function of the both time and condition indicators. Condition indicators provide information about the health condition of an asset; therefore they update and reform the baseline hazard of EHM according to the health state of asset at given time t. Some examples of condition indicators are the vibration of rotating machinery, the level of metal particles in engine oil analysis, and wear in a component, to name but a few. Operating environment indicators in this model are failure accelerators and/or decelerators that are included in the covariate function of EHM and may increase or decrease the value of the hazard from the baseline hazard. These indicators caused by the environment in which an asset operates, and that have not been explicitly identified by the condition indicators (e.g. Loads, environmental stresses, and other dynamically changing environment factors). While the effects of operating environment indicators could be nought in EHM; condition indicators could emerge because these indicators are observed and measured as long as an asset is operational and survived. EHM has several advantages over the existing covariate-based hazard models. One is this model utilises three different sources of asset health data (i.e. population characteristics, condition indicators, and operating environment indicators) to effectively predict hazard and reliability. Another is that EHM explicitly investigates the relationship between condition and operating environment indicators associated with the hazard of an asset. Furthermore, the proportionality assumption, which most of the covariate-based hazard models suffer from it, does not exist in EHM. According to the sample size of failure/suspension times, EHM is extended into two forms: semi-parametric and non-parametric. The semi-parametric EHM assumes a specified lifetime distribution (i.e. Weibull distribution) in the form of the baseline hazard. However, for more industry applications, due to sparse failure event data of assets, the analysis of such data often involves complex distributional shapes about which little is known. Therefore, to avoid the restrictive assumption of the semi-parametric EHM about assuming a specified lifetime distribution for failure event histories, the non-parametric EHM, which is a distribution free model, has been developed. The development of EHM into two forms is another merit of the model. A case study was conducted using laboratory experiment data to validate the practicality of the both semi-parametric and non-parametric EHMs. The performance of the newly-developed models is appraised using the comparison amongst the estimated results of these models and the other existing covariate-based hazard models. The comparison results demonstrated that both the semi-parametric and non-parametric EHMs outperform the existing covariate-based hazard models. Future research directions regarding to the new parameter estimation method in the case of time-dependent effects of covariates and missing data, application of EHM in both repairable and non-repairable systems using field data, and a decision support model in which linked to the estimated reliability results, are also identified.

Characterising metal build-up on urban road surfaces

Reliable approaches for predicting pollutant build-up are essential for accurate urban stormwater quality modelling. Based on the in-depth investigation of metal build-up on residential road surfaces, this paper presents empirical models for predicting metal loads on these surfaces. The study investigated metals commonly present in the urban environment. Analysis undertaken found that the build-up process for metals primarily originating from anthropogenic (copper and zinc) and geogenic (aluminium, calcium, iron and manganese) sources were different. Chromium and nickel were below detection limits. Lead was primarily associated with geogenic sources, but also exhibited a significant relationship with anthropogenic sources. The empirical prediction models developed were validated using an independent data set and found to have relative prediction errors of 12-50%, which is generally acceptable for complex systems such as urban road surfaces. Also, the predicted values were very close to the observed values and well within 95% prediction interval.

Intercalation of kaolins by alkaline earth metal salts

A suite of new materials, based on chemical modification of kaolins, has been successfully prepared via manipulation of the kaolin structure and subsequent intercalation by CaCl2 and MgCl2. A standard kaolinite(KGa-1)and a commercially available halloysite (New Zealand china clay) were used for this study. The kaolins are given several cycles of intercalation and deintercalation using a common intercalant such as potassium acetate. The number of cycles given depends on the type of kaolin. After this treatment, both kaolinite and halloysite hydrate show considerable broadening of the (00l) reflections which indicate extensive exfoliation of the layers. In the case of kaolinite, exfoliated layers roll to form tubes similar to proper halloysite. Kaolins modified by the above treatment readily intercalate MgCl2 and CaCl2 from saturated solutions of these salts. On intercalation with CaCl2 and MgCl2, kaolinite layers expand to 10A and 9.8A, and those of halloysite to 12.8A and 15.5A, respectively. To our knowledge, this is the first report of successful intercalation of alkaline-earth halides by kaolins.

First-principles prediction of metal-free magnetism and intrinsic half-metallicity in graphitic carbon nitride

Transition metal-free magnetism and half-metallicity recently has been the subject of intense research activity due to its potential in spintronics application. Here we, for the first time, demonstrate via density functional theory that the most recently experimentally realized graphitic carbon nitride (g-C4N3) displays a ferromagnetic ground state. Furthermore, this novel material is predicted to possess an intrinsic half-metallicity never reported to date. Our results highlight a new promising material toward realistic metal-free spintronics application.

Nanoporous Graphitic-C3N4@Carbon Metal-Free Electrocatalysts for Highly Efficient Oxygen Reduction

Based on theoretical prediction, a g-C3N4@carbon metal-free oxygen reduction reaction (ORR) electrocatalyst was designed and synthesized by uniform incorporation of g-C3N4 into a mesoporous carbon to enhance the electron transfer efficiency of g-C3N4. The resulting g-C3N4@carbon composite exhibited competitive catalytic activity (11.3 mA cm–2 kinetic-limiting current density at −0.6 V) and superior methanol tolerance compared to a commercial Pt/C catalyst. Furthermore, it demonstrated significantly higher catalytic efficiency (nearly 100% of four-electron ORR process selectivity) than a Pt/C catalyst. The proposed synthesis route is facile and low-cost, providing a feasible method for the development of highly efficient electrocatalysts.

CASNS - A heuristic algorithm for the nesting of irregular-shaped sheet-metal blanks

This article reports on the design and implementation of a computer-aided sheet nesting system (CASNS) for the nesting of two-dimensional irregular-shaped sheet-metal blanks on a given sheet stock or coil stock. The system is designed by considering several constraints of sheet-metal stamping operations, such as bridge width and grain orientation, and design requirements such as maximizing the strength of the part hen subsequent bending is involved, minimization of scrap, and economic justification for'a single or multiple station operation. Through many practical case studies, the system proves its efficiency, effectiveness and usefulness.

CADDS: an automated die design system for sheet-metal blanking

This article reports on the design and implementation of a Computer-Aided Die Design System (CADDS) for sheet-metal blanks. The system is designed by considering several factors, such as the complexity of blank geometry, reduction in scrap material, production requirements, availability of press equipment and standard parts, punch profile complexity, and tool elements manufacturing method. The interaction among these parameters and how they affect designers' decision patterns is described. The system is implemented by interfacing AutoCAD with the higher level languages FORTRAN 77 and AutoLISP. A database of standard die elements is created by parametric programming, which is an enhanced feature of AutoCAD. The greatest advantage achieved by the system is the rapid generation of the most efficient strip and die layouts, including information about the tool configuration.

Metal ion levels post unilateral primary Exeter total hip replacement

There has been much discussion and controversy in the media recently regarding metal toxicity following large head metal on metal (MoM) total hip replacement (THR). Patients have been reported as having hugely elevated levels of metal ions with, at times, devastating systemic, neurolgical and/or orthopaedic sequelae. However, no direct correlation between metal ion level and severity of metallosis has yet been defined. Normative levels of metal ions in well functioning, non Cobalt-Chrome hips have also not been defined to date. The Exeter total hip replacement contains no Cobalt-Chrome (Co-Cr) as it is made entirely from stainless steel. However, small levels of these metals may be present in the modular head of the prosthesis, and their effect on metal ion levels in the well functioning patient has not been investigated. We proposed to define the “normal” levels of metal ions detected by blood test in 20 well functioning patients at a minimum 1 year post primary Exeter total hip replacement, where the patient had had only one joint replaced. Presently, accepted normal levels of blood Chromium are 10–100 nmol/L and plasma Cobalt are 0–20 nmol/L. The UK Modern Humanities Research Association (MHRA) has suggested that levels of either Cobalt or Chromium above 7 ppb (equivalent to 135 nmol/L for Chromium and 120 nmol/L for Cobalt) may be significant. Below this level it is indicated that significant soft tissue reaction and tissue damage is less likely and the risk of implant failure is reduced. Hips were a mixture of cemented and hybrid procedures performed by two experienced orthopaedic consultants. Seventy percent were female, with a mixture of head sizes used. In our cohort, there were no cases where the blood Chromium levels were above the normal range, and in more than 70% of cases, levels were below recordable levels. There were also no cases of elevated plasma Cobalt levels, and in 35% of cases, levels were negligible. We conclude that the implantation with an Exeter total hip replacement does not lead to elevation of blood metal ion levels.

Molecular dynamics simulation of structural characteristics in metal cluster deposition on surfaces

The deposition of small metal clusters (Cu, Au and Al) on f.c.c. metals (Cu, Au and Ni) has been studied by molecular dynamics simulation using Finnis–Sinclair (FS) potential. The impact energy varied from 0.01 to 10 eV/atom. First, the deposition of single cluster was simulated. We observed that, even at much lower energy, a small cluster with (Ih) icosahedral symmetry was reconstructed to match the substrate structure (f.c.c.) after deposition. Next, clusters were modeled to drop, one after the other, on the surface. The nanostructure was found by soft landing of Au clusters on Cu with increasing coverage, where interfacial energy dominates. While at relatively higher deposition energy (a few eV), the ordered f.c.c.-like structure was observed in the first adlayer of the film formed by Al clusters depositing on Ni substrate. This characteristic is mainly attributive to the ballistic collision. Our results indicate that the surface morphology synthesized by cluster deposition could be controlled by experimental parameters, which will be helpful for controlled design of nanostructure.

Inflow cannula design for biventricular assist devices

Cardiovascular diseases are a leading cause of death throughout the developed world. With the demand for donor hearts far exceeding the supply, a bridge-to-transplant or permanent solution is required. This is currently achieved with ventricular assist devices (VADs), which can be used to assist the left ventricle (LVAD), right ventricle (RVAD), or both ventricles simultaneously (BiVAD). Earlier generation VADs were large, volume-displacement devices designed for temporary support until a donor heart was found. The latest generation of VADs use rotary blood pump technology which improves device lifetime and the quality of life for end stage heart failure patients. VADs are connected to the heart and greater vessels of the patient through specially designed tubes called cannulae. The inflow cannulae, which supply blood to the VAD, are usually attached to the left atrium or ventricle for LVAD support, and the right atrium or ventricle for RVAD support. Few studies have characterized the haemodynamic difference between the two cannulation sites, particularly with respect to rotary RVAD support. Inflow cannulae are usually made of metal or a semi-rigid polymer to prevent collapse with negative pressures. However suction, and subsequent collapse, of the cannulated heart chamber can be a frequent occurrence, particularly with the relatively preload insensitive rotary blood pumps. Suction events may be associated with endocardial damage, pump flow stoppages and ventricular arrhythmias. While several VAD control strategies are under development, these usually rely on potentially inaccurate sensors or somewhat unreliable inferred data to estimate preload. Fixation of the inflow cannula is usually achieved through suturing the cannula, often via a felt sewing ring, to the cannulated chamber. This technique extends the time on cardiopulmonary bypass which is associated with several postoperative complications. The overall objective of this thesis was to improve the placement and design of rotary LVAD and RVAD inflow cannulae to achieve enhanced haemodynamic performance, reduced incidence of suction events, reduced levels of postoperative bleeding and a faster implantation procedure. Specific objectives were: * in-vitro evaluation of LVAD and RVAD inflow cannula placement, * design and in-vitro evaluation of a passive mechanism to reduce the potential for heart chamber suction, * design and in-vitro evaluation of a novel suture-less cannula fixation device. In order to complete in-vitro evaluation of VAD inflow cannulae, a mock circulation loop (MCL) was developed to accurately replicate the haemodynamics in the human systemic and pulmonary circulations. Validation of the MCL’s haemodynamic performance, including the form and magnitude of pressure, flow and volume traces was completed through comparisons of patient data and the literature. The MCL was capable of reproducing almost any healthy or pathological condition, and provided a useful tool to evaluate VAD cannulation and other cardiovascular devices. The MCL was used to evaluate inflow cannula placement for rotary VAD support. Left and right atrial and ventricular cannulation sites were evaluated under conditions of mild and severe heart failure. With a view to long term LVAD support in the severe left heart failure condition, left ventricular inflow cannulation was preferred due to improved LVAD efficiency and reduced potential for thrombus formation. In the mild left heart failure condition, left atrial cannulation was preferred to provide an improved platform for myocardial recovery. Similar trends were observed with RVAD support, however to a lesser degree due to a smaller difference in right atrial and ventricular pressures. A compliant inflow cannula to prevent suction events was then developed and evaluated in the MCL. As rotary LVAD or RVAD preload was reduced, suction events occurred in all instances with a rigid inflow cannula. Addition of the compliant segment eliminated suction events in all instances. This was due to passive restriction of the compliant segment as preload dropped, thus increasing the VAD circuit resistance and decreasing the VAD flow rate. Therefore, the compliant inflow cannula acted as a passive flow control / anti-suction system in LVAD and RVAD support. A novel suture-less inflow cannula fixation device was then developed to reduce implantation time and postoperative bleeding. The fixation device was evaluated for LVAD and RVAD support in cadaveric animal and human hearts attached to a MCL. LVAD inflow cannulation was achieved in under two minutes with the suture-less fixation device. No leakage through the suture-less fixation device – myocardial interface was noted. Continued development and in-vivo evaluation of this device may result in an improved inflow cannulation technique with the potential for off-bypass insertion. Continued development of this research, in particular the compliant inflow cannula and suture-less inflow cannulation device, will result in improved postoperative outcomes, life span and quality of life for end-stage heart failure patients.

A semi-supervised graph-based algorithm for detecting outliers in online-social-networks

In this paper, we propose a semi-supervised approach of anomaly detection in Online Social Networks. The social network is modeled as a graph and its features are extracted to detect anomaly. A clustering algorithm is then used to group users based on these features and fuzzy logic is applied to assign degree of anomalous behavior to the users of these clusters. Empirical analysis shows effectiveness of this method.

Carbon dioxide reforming of methane to produce synthesis gas over metal-supported catalysts: State of the art

Carbon dioxide reforming of methane produces synthesis gas with a low hydrogen to carbon monoxide ratio, which is desirable for many industrial synthesis processes. This reaction also has very important environmental implications since both methane and carbon dioxide contribute to the greenhouse effect. Converting these gases into a valuable feedstock may significantly reduce the atmospheric emissions of CO2 and CH4. In this paper, we present a comprehensive review on the thermodynamics, catalyst selection and activity, reaction mechanism, and kinetics of this important reaction. Recently, research has centered on the development of catalysts and the feasible applications of this reaction in industry. Group VIII metals supported on oxides are found to be effective for this reason. However, carbon deposition causing catalyst deactivation is the major problem inhibiting the industrial application of the CO2/CH4 reaction. Ni-based catalysts impregnated on certain supports show carbon-free operation and thus attract much attention. To develop an effective catalyst for CO2 reforming of CH4 and accelerate the commercial application of the reaction, the following are identified to be the most important areas for future work: (1) selection of metal and support and studying the effect of their interaction on catalyst activity; (2) the effect of different promoter on catalyst activity; (3) the reaction mechanism and kinetics; and (4) pilot reactor performance and scale-up operation.

Graphyne and graphdiyne : versatile catalysts for dehydrogenation of light metal complex hydrides

The interaction between new two-dimensional carbon allotropes, i.e. graphyne (GP) and graphdiyne (GD), and light metal complex hydrides LiAlH4, LiBH4, and NaAlH4 was studied using density functional theory (DFT) incorporating long range van der Waals dispersion correction. The light metal complex hydrides show much stronger interaction with GP and GP than that with fullerene due to the well defined pore structure. Such strong interactions greatly affect the degree of charge donation from the alkali metal atom to AlH4 or BH4, consequently destabilizing the Al-H or B-H bonds. Compared to the isolated light metal complex hydride, the presence of GP or GD can lead to a significant reduction of the hydrogen removal energy. Most interestingly, the hydrogen removal energies for LiBHx on GP and with GD are found to be lowered at all the stages (x from 4 to 1) whereas the H-removal energy in the third stage is increased for LiBH4 on fullerene. In addition, the presence of uniformly distributed pores on GP and GD is expected to facilitate the dehydrogenation of light metal complex hydrides. The present results highlight new interesting materials to catalyze light metal complex hydrides for potential application as media for hydrogen storage. Since GD has been successfully synthesized in a recent experiment, we hope the present work will stimulate further experimental investigations in this direction.