Reliable Fault Diagnosis for Low-Speed Bearings Using Individually Trained Support Vector Machines With Kernel Discriminative Feature Analysis

This paper proposes a highly reliable fault diagnosis approach for low-speed bearings. The proposed approach first extracts wavelet-based fault features that represent diverse symptoms of multiple low-speed bearing defects. The most useful fault features for diagnosis are then selected by utilizing a genetic algorithm (GA)-based kernel discriminative feature analysis cooperating with one-against-all multicategory support vector machines (OAA MCSVMs). Finally, each support vector machine is individually trained with its own feature vector that includes the most discriminative fault features, offering the highest classification performance. In this study, the effectiveness of the proposed GA-based kernel discriminative feature analysis and the classification ability of individually trained OAA MCSVMs are addressed in terms of average classification accuracy. In addition, the proposedGA- based kernel discriminative feature analysis is compared with four other state-of-the-art feature analysis approaches. Experimental results indicate that the proposed approach is superior to other feature analysis methodologies, yielding an average classification accuracy of 98.06% and 94.49% under rotational speeds of 50 revolutions-per-minute (RPM) and 80 RPM, respectively. Furthermore, the individually trained MCSVMs with their own optimal fault features based on the proposed GA-based kernel discriminative feature analysis outperform the standard OAA MCSVMs, showing an average accuracy of 98.66% and 95.01% for bearings under rotational speeds of 50 RPM and 80 RPM, respectively.

Stormwater management: An introduction to Green Gully

In recent years there has been increasing interest in the use of water resources generated within the urban boundary for potable supply substitution as a means of augmenting the current supply capacity. These urban water resources include roof and stormwater runoff. Expanding the use of stormwater runoff to add to the water supply and reduce water pollution are important objectives all over Australia. This book presents the background, significance and objectives of the research, as well as the reasons why stormwater plays a significant role as an alternative source of water.

Scale adaptive tracking using mean shift and efficient feature matching

The mean shift tracker has achieved great success in visual object tracking due to its efficiency being nonparametric. However, it is still difficult for the tracker to handle scale changes of the object. In this paper, we associate a scale adaptive approach with the mean shift tracker. Firstly, the target in the current frame is located by the mean shift tracker. Then, a feature point matching procedure is employed to get the matched pairs of the feature point between target regions in the current frame and the previous frame. We employ FAST-9 corner detector and HOG descriptor for the feature matching. Finally, with the acquired matched pairs of the feature point, the affine transformation between target regions in the two frames is solved to obtain the current scale of the target. Experimental results show that the proposed tracker gives satisfying results when the scale of the target changes, with a good performance of efficiency.

Relocating a flood-affected community : good planning or good politics?

Problem, research strategy and findings: On January 10, 2011, the town of Grantham, Queensland (Australia), was inundated with a flash flood in which 12 of the town's 370 residents drowned. The overall damage bill in Queensland was AUD∃2.38 billion (USD∃2.4 billion) with 35 deaths, and more than three-quarters of the state was declared a flood disaster zone. In this study, we focus on the unusual and even rare decision to relocate Grantham in March 2011. The Lockyer Valley Regional Council (LVRC) acquired a 377-hectare (932-acre) site to enable a voluntary swap of equivalent-sized lots. In addition, planning regulations were set aside to streamline the relocation of a portion of the town. We review the natural hazard literature as it relates to community relocation, state and local government documents related to Grantham, and reports and newspaper articles related to the flood. We also analyze data from interviews with key stakeholders. We document the process of community relocation, assess the relocation process in Grantham against best practice, examine whether the process of community relocation can be upscaled and if the Grantham relocation is an example of good planning or good politics. Takeaway for practice: Our study reveals two key messages for practice. Community relocation (albeit a small one) is possible, and the process can be done quickly; some Grantham residents moved into their new, relocated homes in December 2012, just 11 months after the flood. Moreover, the role of existing planning regulations can be a hindrance to quick action; political leadership, particularly at the local level, is key to implementing the relocation.

Application of anomaly technique in wind turbine bearing fault detection

Bearing faults are the most common cause of wind turbine failures. Unavailability and maintenance cost of wind turbines are becoming critically important, with their fast growing in electric networks. Early fault detection can reduce outage time and costs. This paper proposes Anomaly Detection (AD) machine learning algorithms for fault diagnosis of wind turbine bearings. The application of this method on a real data set was conducted and is presented in this paper. For validation and comparison purposes, a set of baseline results are produced using the popular one-class SVM methods to examine the ability of the proposed technique in detecting incipient faults.

Overvoltage and overloading prevention using coordinated PV inverters in distribution network

Overvoltage and overloading due to high utilization of PVs are the main power quality concerns for future distribution power systems. This paper proposes a distributed control coordination strategy to manage multiple PVs within a network to overcome these issues. PVs reactive power is used to deal with over-voltages and PVs active power curtailment are regulated to avoid overloading. The proposed control structure is used to share the required contribution fairly among PVs, in proportion to their ratings. This approach is examined on a practical distribution network with multiple PVs.

Considerations for a social and geographical framework for agent-based epidemics

Understanding the dynamics of disease spread is of crucial importance, in contexts such as estimating load on medical services to risk assessment and intervention policies against large-scale epidemic outbreaks. However, most of the information is available after the spread itself, and preemptive assessment is far from trivial. Here, we investigate the use of agent-based simulations to model such outbreaks in a stylised urban environment. For most diseases, infection of a new individual may occur from casual contact in crowds as well as from repeated interactions with social partners such as work colleagues or family members. Our model therefore accounts for these two phenomena.Presented in this paper is the initial framework for such a model, detailing implementation of geographical features and generation of social structures. Preliminary results are a promising step towards large-scale simulations and evaluation of potential intervention policies.

From bio-inspired computing to e-Biology

Several algorithms and techniques widely used in Computer Science have been adapted from, or inspired by, known biological phenomena. This is a consequence of the multidisciplinary background of most early computer scientists. The field has now matured, and permits development of tools and collaborative frameworks which play a vital role in advancing current biomedical research. In this paper, we briefly present examples of the former, and elaborate upon two of the latter, applied to immunological modelling and as a new paradigm in gene expression.

Impact of mobility constraints on epidemic broadcast mechanisms in delay-tolerant networks

In this paper, we investigate the effect of mobility constraints on epidemic broadcast mechanisms in DTNs (Delay-Tolerant Networks). Major factors affecting epidemic broadcast performances are its forwarding algorithm and node mobility. The impact of forwarding algorithm and node mobility on epidemic broadcast mechanisms has been actively studied in the literature, but those studies generally use unconstrained mobility models. The objective of this paper is therefore to quantitatively investigate the effect of mobility constraints on epidemic broadcast mechanisms. We evaluate the performances of three classes of epidemic broadcast mechanisms - P-BCAST (PUSH-based BroadCast), SA-BCAST (Self-Adaptive BroadCast), and HP-BCAST (History-based P-BCAST) - with a random waypoint mobility model with mobility constraints. Our finding includes that the existence of mobility constraints significantly improves the reach ability and dissemination speed of epidemic broadcast mechanisms while degrading their efficiency.

Impact of mobility and topology on information diffusion in MANETs

In some delay-tolerant communication systems such as vehicular ad-hoc networks, information flow can be represented as an infectious process, where each entity having already received the information will try to share it with its neighbours. The random walk and random waypoint models are popular analysis tools for these epidemic broadcasts, and represent two types of random mobility. In this paper, we introduce a simulation framework investigating the impact of a gradual increase of bias in path selection (i.e. reduction of randomness), when moving from the former to the latter. Randomness in path selection can significantly alter the system performances, in both regular and irregular network structures. The implications of these results for real systems are discussed in details.

High-performance computing for data analytics

One of the main challenges in data analytics is that discovering structures and patterns in complex datasets is a computer-intensive task. Recent advances in high-performance computing provide part of the solution. Multicore systems are now more affordable and more accessible. In this paper, we investigate how this can be used to develop more advanced methods for data analytics. We focus on two specific areas: model-driven analysis and data mining using optimisation techniques.

Performance evaluation of cloud-based parallel computing

As computational models in fields such as medicine and engineering get more refined, resource requirements are increased. In a first instance, these needs have been satisfied using parallel computing and HPC clusters. However, such systems are often costly and lack flexibility. HPC users are therefore tempted to move to elastic HPC using cloud services. One difficulty in making this transition is that HPC and cloud systems are different, and performance may vary. The purpose of this study is to evaluate cloud services as a means to minimise both cost and computation time for large-scale simulations, and to identify which system properties have the most significant impact on performance. Our simulation results show that, while the performance of Virtual CPU (VCPU) is satisfactory, network throughput may lead to difficulties.

Efficiency of parallelisation of genetic algorithms in the data analysis context

Most real-life data analysis problems are difficult to solve using exact methods, due to the size of the datasets and the nature of the underlying mechanisms of the system under investigation. As datasets grow even larger, finding the balance between the quality of the approximation and the computing time of the heuristic becomes non-trivial. One solution is to consider parallel methods, and to use the increased computational power to perform a deeper exploration of the solution space in a similar time. It is, however, difficult to estimate a priori whether parallelisation will provide the expected improvement. In this paper we consider a well-known method, genetic algorithms, and evaluate on two distinct problem types the behaviour of the classic and parallel implementations.

Optimisation algorithms for microarray biclustering

In providing simultaneous information on expression profiles for thousands of genes, microarray technologies have, in recent years, been largely used to investigate mechanisms of gene expression. Clustering and classification of such data can, indeed, highlight patterns and provide insight on biological processes. A common approach is to consider genes and samples of microarray datasets as nodes in a bipartite graphs, where edges are weighted e.g. based on the expression levels. In this paper, using a previously-evaluated weighting scheme, we focus on search algorithms and evaluate, in the context of biclustering, several variations of Genetic Algorithms. We also introduce a new heuristic “Propagate”, which consists in recursively evaluating neighbour solutions with one more or one less active conditions. The results obtained on three well-known datasets show that, for a given weighting scheme,optimal or near-optimal solutions can be identified.

Mini-Orb : A personal indoor climate preference feedback interface

The control of environmental factors in open-office environments, such as lighting and temperature is becoming increasingly automated. This development means that office inhabitants are losing the ability to manually adjust environmental conditions according to their needs. In this paper we describe the design, use and evaluation of MiniOrb, a system that employs ambient and tangible interaction mechanisms to allow inhabitants of office environments to maintain awareness of environmental factors, report on their own subjectively perceived office comfort levels and see how these compare to group average preferences. The system is complemented by a mobile application, which enables users to see and set the same sensor values and preferences, but using a screen-based interface. We give an account of the system’s design and outline the results of an in-situ trial and user study. Our results show that devices that combine ambient and tangible interaction approaches are well suited to the task of recording indoor climate preferences and afford a rich set of possible interactions that can complement those enabled by more conventional screen-based interfaces.