Semi batch learning with store management using enhanced conjugate gradient

This paper explores the performance of sliding-window based training, termed as semi batch, using multilayer perceptron (MLP) neural network in the presence of correlated data. The sliding window training is a form of higher order instantaneous learning strategy without the need of covariance matrix, usually employed for modeling and tracking purposes. Sliding-window framework is implemented to combine the robustness of offline learning algorithms with the ability to track online the underlying process of a function. This paper adopted sliding window training with recent advances in conjugate gradient direction with application of data store management e.g. simple distance measure, angle evaluation and the novel prediction error test. The simulation results show the best convergence performance is gained by using store management techniques. © 2012 Springer-Verlag.

Tracking smart grid hackers

The next-generation smart grid will rely highly on telecommunications infrastructure for data transfer between various systems. Anywhere we have data transfer in a system is a potential security threat. When we consider the possibility of smart grid data being at the heart of our critical systems infrastructure it is imperative that we do all we can to ensure the confidentiality, availability and integrity of the data. A discussion on security itself is outside the scope of this paper, but if we assume the network to be as secure as possible we must consider what we can do to detect when that security fails, or when the attacks comes from the inside of the network. One way to do this is to setup a hacker-trap, or honeypot. A honeypot is a device or service on a network which appears legitimate, but is in-fact a trap setup to catch breech attempts. This paper identifies the different types of honeypot and describes where each may be used. The authors have setup a test honeypot system which has been live for some time. The test system has been setup to emulate a device on a utility network. The system has had many hits, which are described in detail by the authors. Finally, the authors discuss how larger-scale systems in utilities may benefit from honeypot placement.

Securing the smart grid

The availability of electricity is fundamental to modern society. It is at the top of the list of critical infrastructures and its interruption can have severe consequences. This highly important system is now evolving to become more reliable, efficient, and clean. This evolving infrastructure has become known as the smart grid; and these future smart grid systems will rely heavily on ICT. This infrastructure will require many servers and due to the nature of the grid, many of these systems will be geographically diverse requiring communication links. At the heart of this ICT infrastructure will be security. At each level of the smart grid from smart metering right through to remote sensing and control networks, security will be a key factor for system design consideration. With an increased number of ICT systems in place the security risk also increases. In this paper the authors discuss the changing nature of security in relation to the smart grid by looking at the move from legacy systems to more modern smart grid systems. The potential planes of attack for future smart grid systems are identified, and the general anatomy of a cyber-attack is presented. The authors then introduce the various threat levels of different types of attack and the mitigation techniques that could be put in place for each. Finally, the authors' introduce a Phasor Measurement Unit (PMU) communication system (operated by the authors) that can be used as a test-bed for some of the proposed future security research.

Small Signal Stability Performance of Power System during High Penetration of Wind Generation

The small signal stability of interconnected power systems is one of the important aspects that need to be investigated since the oscillations caused by this kind of instability have caused many incidents. With the increasing penetration of wind power in the power system, particularly doubly fed induction generator (DFIG), the impact on the power system small signal stability performance should be fully investigated. Because the DFIG wind turbine integration is through a fast action converter and associated control, it does not inherently participate in the electromechanical small signal oscillation. However, it influences the small signal stability by impacting active power flow paths in the network and replacing synchronous generators that have power system stabilizer (PSS). In this paper, the IEEE 39 bus test system has been used in the analysis. Furthermore, four study cases and several operation scenarios have been conducted and analysed. The selective eigenvalue Arnoldi/lanczos's method is used to obtain the system eigenvalue in the range of frequency from 0.2 Hz to 2 Hz which is related to electromechanical oscillations. Results show that the integration of DFIG wind turbines in a system during several study cases and operation scenarios give different influence on small signal stability performance.

Issues in fe modelling of FRP strengthened masonry arch bridges

Externally bonding of FRP composites is an effective technique for retrofitting historical masonry arch structures. A major failure mode in such strengthened structures is the debonding of FRP from the masonry. The bond behaviour between FRP and masonry thus plays a crucial role in these structures. Major challenges exist in the finite element modelling of such structures, such as modelling of mixed Mode-I and Mode-II bond behaviour between the FRP and the curved masonry substrate, modelling of existing damages in the masonry arches, consideration of loading history in the unstrengthened and strengthened structure etc. This paper presents a rigorous FE model for simulating FRP strengthened masonry arch structures. A detailed solid model was developed for simulating the masonry and a mixed-mode interface model was used for simulating the FRP-to-masonry bond behaviour. The model produces results in very close agreement with test results.

A technique for reducing strain concentrations in FRP wrapped concrete columns

Existing studies have shown conclusively that the measured fibre reinforced polymer (FRP) rupture strain in FRP wrapped concrete columns is usually significantly smaller than the rupture strain obtained from flat coupon tests. One of the main causes for this phenomenon is the existence of geometrical discontinuities at both ends of the FRP sheets. This study proposes a new strengthening method in which continuous FRP spiral wrapping is used to eliminate strain concentrations due to the geometrical discontinuities and thus increase the FRP rupture strain at column failure. The effect of the spiral angle of FRP on the FRP rupture strain in FRP wrapped specimens was experimentally investigated. The test results indicate that the spiral wrapping with a small angle with respect to the column circumference can significantly increase the strain efficiency of FRP and thus enhance the axial compression capacity of the strengthened cylinders.

Bond behavior of NSM FRP strips between two adjacent cracks in concrete beams: A numerical study

Strengthening RC structures with near-surface mounted (NSM) fibre reinforced polymer (FRP) composites has a number of advantages compared with that with externally bonded (EB) FRP sheets/plates. As with EB FRP, the performance of the bond between NSM FRP and concrete is one of the key factors affecting the behaviour of the strengthened structure. This paper presents a numerical investigation into the behaviour of NSM FRP loaded at its both ends to simulate the NSM FRP-toconcrete bond between two adjacent cracks in RC members. The main objective of this study is to quantitatively clarify the effect of the bondline damage during slip reversal on the ultimate load (bond strength). The results show that the bondline damage has a significant effect on the load-carrying capacity of the NSM FRP-to-concrete bonded interface and should be considered in FE modeling of the interface.

Mitigation of concrete cover separation in frpstrengthened RC beams using FRP U-jackets

Concrete cover separation is a common failure mode of reinforced concrete (RC) beams strengthened with a fibre-reinforced polymer (FRP) plate bonded to the tension face (FRP-plated RC beams). Plate-end FRP U-jackets have previously been explored as a mitigation measure to delay or suppress concrete cover separation, although its effectiveness needs further clarification. The paper presents the first systemic experimental study on the use of FRP U-jackets of different forms for mitigating the concrete cover separation failure. A total of ten full-scale FRP-plated RC beams were tested. The test results show that both the ultimate load and the ductility of the beams were enhanced by the U-jackets. Among the forms of U-jackets explored, those inclined at 45o are the most effective.

Effect of Cyclic Carbonation on Chloride Ingression in GGBS Concrete

Carbonation and chloride ingress are the two main causes of corrosion in reinforced concrete structures. An investigation to monitor the ingress of chlorides and the effect of carbonation on chloride ingression during an accelerated 12 month cyclic wetting and drying exposure regime that simulates conditions in which multiple mode transport mechanisms are active was conducted on ground granulated blast furnace slag (GGBS) concrete. The penetration of chloride and carbon dioxide was evaluated using water and acid soluble chloride profiles and phenolphthalein indicator, respectively. The results indicated that when chloride and carbon dioxide ingress concomitantly the effects can be adverse. Carbonation has a detrimental effect on the binding capacity of the concrete, increasing the concentration of free (water soluble) chlorides. This contributed to greater concentration and greater penetration of chlorides and thus an increased corrosion risk.

A decision support system for managing irrigation in agriculture

In this paper, an automatic Smart Irrigation Decision Support System, SIDSS, is proposed to manage irrigation in agriculture. Our system estimates the weekly irrigations needs of a plantation, on the basis of both soil measurements and climatic variables gathered by several autonomous nodes deployed in field. This enables a closed loop control scheme to adapt the decision support system to local perturbations and estimation errors. Two machine learning techniques, PLSR and ANFIS, are proposed as reasoning engine of our SIDSS. Our approach is validated on three commercial plantations of citrus trees located in the South-East of Spain. Performance is tested against decisions taken by a human expert.