Anomaly detection in online social networks : using data-mining techniques and fuzzy logic

This research is a step forward in improving the accuracy of detecting anomaly in a data graph representing connectivity between people in an online social network. The proposed hybrid methods are based on fuzzy machine learning techniques utilising different types of structural input features. The methods are presented within a multi-layered framework which provides the full requirements needed for finding anomalies in data graphs generated from online social networks, including data modelling and analysis, labelling, and evaluation.

Fuzzy logic to evaluate driving maneuvers: An integrated approach to improve training

Driver training is one of the interventions aimed at mitigating the number of crashes that involve novice drivers. Our failure to understand what is really important for learners, in terms of risky driving, is one of the many drawbacks restraining us to build better training programs. Currently, there is a need to develop and evaluate Advanced Driving Assistance Systems that could comprehensively assess driving competencies. The aim of this paper is to present a novel Intelligent Driver Training System (IDTS) that analyses crash risks for a given driving situation, providing avenues for improvement and personalisation of driver training programs. The analysis takes into account numerous variables acquired synchronously from the Driver, the Vehicle and the Environment (DVE). The system then segments out the manoeuvres within a drive. This paper further presents the usage of fuzzy set theory to develop the safety inference rules for each manoeuvre executed during the drive. This paper presents a framework and its associated prototype that can be used to comprehensively view and assess complex driving manoeuvres and then provide a comprehensive analysis of the drive used to give feedback to novice drivers.

Theoretical modelling of momentum transfer function of bi-disperse porous media

The aim of this paper is to obtain the momentum transfer coefficient between the two phases, denoted by f and p, occupying a bi-disperse porous medium by mapping the available experimental data to the theoretical model proposed by Nield and Kuznetsov. Data pertinent to plate-fin heat exchangers, as bi-disperse porous media, were used. The measured pressure drops for such heat exchangers are then used to give the overall permeability which is linked to the porosity and permeability of each phase as well as the interfacial momentum transfer coefficient between the two phases. Accordingly, numerical values are obtained for the momentum transfer coefficient for three different fin spacing values considered in the heat exchanger experiments.

Application of TS-Fuzzy Controller for Active Power and DC Capacitor Voltage Control in DFIG-Based Wind Energy Conversion Systems

This chapter focuses on the implementation of the TS (Tagaki-Sugino) fuzzy controller for the Doubly Fed Induction Generator (DFIG) based wind generator. The conventional PI control loops for mantaining desired active power and DC capacitor voltage is compared with the TS fuzzy controllers. DFIG system is represented by a third-order model where electromagnetic transients of the stator are neglected. The effectiveness of the TS-fuzzy controller on the rotor speed oscillations and the DC capacitor voltage variations of the DFIG damping controller on converter ratings is also investigated. The results from the time domain simulations are presented to elucidate the effectiveness of the TS-fuzzy controller over the conventional PI controller in the DFIG system. The proposed TS-fuzzy con-troller can improve the fault ride through capability of DFIG compared to the conventional PI controller.

Rough Fuzzy Control of SVC for Power System Stability Enhancement

Modern power systems have become more complex due to the growth in load demand, the installation of Flexible AC Transmission Systems (FACTS) devices and the integration of new HVDC links into existing AC grids. On the other hand, the introduction of the deregulated and unbundled power market operational mechanism, together with present changes in generation sources including connections of large renewable energy generation with intermittent feature in nature, have further increased the complexity and uncertainty for power system operation and control. System operators and engineers have to confront a series of technical challenges from the operation of currently interconnected power systems. Among the many challenges, how to evaluate the steady state and dynamic behaviors of existing interconnected power systems effectively and accurately using more powerful computational analysis models and approaches becomes one of the key issues in power engineering. The traditional computing techniques have been widely used in various fields for power system analysis with varying degrees of success. The rapid development of computational intelligence, such as neural networks, fuzzy systems and evolutionary computation, provides tools and opportunities to solve the complex technical problems in power system planning, operation and control.

Bi-modal biometric authentication on mobile phones in challenging conditions

This paper examines the issue of face, speaker and bi-modal authentication in mobile environments when there is significant condition mismatch. We introduce this mismatch by enrolling client models on high quality biometric samples obtained on a laptop computer and authenticating them on lower quality biometric samples acquired with a mobile phone. To perform these experiments we develop three novel authentication protocols for the large publicly available MOBIO database. We evaluate state-of-the-art face, speaker and bi-modal authentication techniques and show that inter-session variability modelling using Gaussian mixture models provides a consistently robust system for face, speaker and bi-modal authentication. It is also shown that multi-algorithm fusion provides a consistent performance improvement for face, speaker and bi-modal authentication. Using this bi-modal multi-algorithm system we derive a state-of-the-art authentication system that obtains a half total error rate of 6.3% and 1.9% for Female and Male trials, respectively.

Differing phagocytic capacities of accessory and main olfactory ensheathing cells and the implication for olfactory glia transplantation therapies

The rodent olfactory systems comprise the main olfactory system for the detection of odours and the accessory olfactory system which detects pheromones. In both systems, olfactory axon fascicles are ensheathed by olfactory glia, termed olfactory ensheathing cells (OECs), which are crucial for the growth and maintenance of the olfactory nerve. The growth-promoting and phagocytic characteristics of OECs make them potential candidates for neural repair therapies such as transplantation to repair the injured spinal cord. However, transplanting mixed populations of glia with unknown properties may lead to variations in outcomes for neural repair. As the phagocytic capacity of the accessory OECs has not yet been determined, we compared the phagocytic capacity of accessory and main OECs in vivo and in vitro. In normal healthy animals, the accessory OECs accumulated considerably less axon debris than main OECs in vivo. Analysis of freshly dissected OECs showed that accessory OECs contained 20% less fluorescent axon debris than main OECs. However, when assayed in vitro with exogenous axon debris added to the culture, the accessory OECs phagocytosed almost 20% more debris than main OECs. After surgical removal of one olfactory bulb which induced the degradation of main and accessory olfactory sensory axons, the accessory OECs responded by phagocytosing the axon debris. We conclude that while accessory OECs have the capacity to phagocytose axon debris, there are distinct differences in their phagocytic capacity compared to main OECs. These distinct differences may be of importance when preparing OECs for neural transplant repair therapies.

Fuzzy-DL perception for multi-robot systems

For future planetary robot missions, multi-robot-systems can be considered as a suitable platform to perform space mission faster and more reliable. In heterogeneous robot teams, each robot can have different abilities and sensor equipment. In this paper we describe a lunar demonstration scenario where a team of mobile robots explores an unknown area and identifies a set of objects belonging to a lunar infrastructure. Our robot team consists of two exploring scout robots and a mobile manipulator. The mission goal is to locate the objects within a certain area, to identify the objects, and to transport the objects to a base station. The robots have a different sensor setup and different capabilities. In order to classify parts of the lunar infrastructure, the robots have to share the knowledge about the objects. Based on the different sensing capabilities, several information modalities have to be shared and combined by the robots. In this work we propose an approach using spatial features and a fuzzy logic based reasoning for distributed object classification.

An application of fuzzy DL-Based semantic perception to soil container classification

Semantic perception and object labeling are key requirements for robots interacting with objects on a higher level. Symbolic annotation of objects allows the usage of planning algorithms for object interaction, for instance in a typical fetchand-carry scenario. In current research, perception is usually based on 3D scene reconstruction and geometric model matching, where trained features are matched with a 3D sample point cloud. In this work we propose a semantic perception method which is based on spatio-semantic features. These features are defined in a natural, symbolic way, such as geometry and spatial relation. In contrast to point-based model matching methods, a spatial ontology is used where objects are rather described how they "look like", similar to how a human would described unknown objects to another person. A fuzzy based reasoning approach matches perceivable features with a spatial ontology of the objects. The approach provides a method which is able to deal with senor noise and occlusions. Another advantage is that no training phase is needed in order to learn object features. The use-case of the proposed method is the detection of soil sample containers in an outdoor environment which have to be collected by a mobile robot. The approach is verified using real world experiments.

Sectionalizing strategies for minimizing outage durations of critical loads in parallel power system restoration with bi-level programming

Fast restoration of critical loads and non-black-start generators can significantly reduce the economic losses caused by power system blackouts. In a parallel power system restoration scenario, the sectionalization of restoration subsystems plays a very important role in determining the pickup of critical loads before synchronization. Most existing research mainly focuses on the startup of non-black-start generators. The restoration of critical loads, especially the loads with cold load characteristics, has not yet been addressed in optimizing the subsystem divisions. As a result, sectionalized restoration subsystems cannot achieve the best coordination between the pickup of loads and the ramping of generators. In order to generate sectionalizing strategies considering the pickup of critical loads in parallel power system restoration scenarios, an optimization model considering power system constraints, the characteristics of the cold load pickup and the features of generator startup is proposed in this paper. A bi-level programming approach is employed to solve the proposed sectionalizing model. In the upper level the optimal sectionalizing problem for the restoration subsystems is addressed, while in the lower level the objective is to minimize the outage durations of critical loads. The proposed sectionalizing model has been validated by the New-England 39-bus system and the IEEE 118-bus system. Further comparisons with some existing methods are carried out as well.

Fuzzy-logic based frequency controller for wind farms augmented with energy storage systems

Displacement of conventional synchronous generators by non-inertial units such as wind or solar generators will result in reduced-system inertia affecting under-frequency response. Frequency control is important to avoid equipment damage, load shedding, and possible blackouts. Wind generators along with energy storage systems can be used to improve the frequency response of low-inertia power system. This paper proposes a fuzzy-logic based frequency controller (FFC) for wind farms augmented with energy storage systems (wind-storage system) to improve the primary frequency response in future low-inertia hybrid power system. The proposed controller provides bidirectional real power injection using system frequency deviations and rate of change of frequency (RoCoF). Moreover, FFC ensures optimal use of energy from wind farms and storage units by eliminating the inflexible de-loading of wind energy and minimizing the required storage capacity. The efficacy of the proposed FFC is verified on the low-inertia hybrid power system.

Conversion of n-type CuTCNQ into p-type nitrogen-doped CuO and the implication for room temperature gas sensing

Sensors to detect toxic and harmful gases are usually based on metal oxides that are operated at elevated temperature. However, enabling gas detection at room temperature (RT) is a significant ongoing challenge. Here, we address this issue by demonstrating that microrods of semiconducting CuTCNQ (TCNQ=7,7,8,8-tetracyanoquinodimethane) with nanostructured features can be employed as conductometric gas sensors operating at 50°C for detection of oxidizing and reducing gases such as NO2 and NH3. The sensor is evaluated at RT and up to 200°C. It was found that CuTCNQ is transformed into a N-doped CuO material with p-type conductivity when annealed at the maximum temperature. This is the first time that such a transformation, from a semiconducting charge transfer material into a N-doped metal oxide is detected. It is shown here that both the surface chemistry and the type of majority charge carrier within the sensing layer is critically important for the type of response towards oxidizing and reducing gases. A detailed physical description of NO2 and NH3 sensing mechanism at CuTCNQ and N-doped CuO is provided to explain the difference in the response. For the N-doped CuO sensor, a detection limit of 1 ppm for NO2 and 10 ppm for NH3 are achieved.

Multipurpose nanoporous alumina-carbon nanowall bi-dimensional nano-hybrid platform via catalyzed and catalyst-free plasma CVD

Simple, rapid, plasma-assisted synthesis of large-area arrays of vertically-aligned carbon nanowalls on highly-porous, transparent bare and gold-coated alumina membranes with the two pore sizes is reported. It is demonstrated that the complex patterns of vertically aligned nanowalls can nucleate and form different morphologies in the low-temperature plasmas. The process is stable, and the twofold change in the gas flow (10 and 20 sccm) does not noticeably influence the morphology of the nanowall pattern. Application of a thin (5 nm) gold layer to nanoporous membrane prior to the nanowall growth allows controlling the network morphology.