Khaled Bashir Shaban

Classification of common partial discharge types in oil-paper insulation system using acoustic signals

This paper addresses classifying different common partial discharge (PD) types under different acoustic emission (AE) measurement conditions. Four types of PDs are considered for the multi-class classification problem, namely; PD from a sharp point to ground plane, surface discharge, PD from a void in the insulation, and PD from semi parallel planes. The collected AE signals are processed using pattern classification techniques to identify their corresponding PD types. The measurement conditions include the influences of various PD locations, oil temperatures, and having a barrier in the line-of-sight between the PD source and the AE sensor. A recognition rate of 94% is achieved when classifying the different PD types measured at the same conditions. In addition, it has been found that the different PD source locations, oil temperatures, and barrier insertion have an impact on the recognition rate. However, by including AE samples at these different conditions in the training process, a recognition rate around 90% for all cases is achieved.

A cascade of artificial neural networks to predict transformers oil parameters

In this paper artificial neural networks have been constructed to predict different transformers oil parameters. The prediction is performed through modeling the relationship between the insulation resistance measured between distribution transformers high voltage winding, low voltage winding and the ground and the breakdown strength, interfacial tension acidity and the water content of the transformers oil. The process of predicting these oil parameters statuses is carried out using various configurations of neural networks. First, a multilayer feed forward neural network with a back-propagation learning algorithm was implemented. Subsequently, a cascade of these neural networks was deemed to be more promising, and four variations of a three stage cascade were tested. The first configuration takes four inputs and outputs four parameter values, while the other configurations have four neural networks, each with two or three inputs and a single output; the output from some networks are pipelined to some others to produce the final values. Both configurations are evaluated using real-world training and testing data and the accuracy is calculated across a variety of hidden layer and hidden neuron combinations. The results indicate that even with a lack of sufficient data to train the network, accuracy levels of 84% for breakdown voltage, 95% for interfacial tension, 56% for water content, and 75% for oil acidity predictions were obtained by the cascade of neural networks.

MINTED: M ulticast VI rtual N e T work E mbedding in Cloud Data Centers With D elay Constraints

Network virtualization is regarded as the pillar of cloud computing, enabling the multi-tenancy concept where multiple Virtual Networks (VNs) can cohabit the same substrate network. With network virtualization, the problem of allocating resources to the various tenants, commonly known as the Virtual Network Embedding problem, emerges as a challenge. Its NP-Hard nature has drawn a lot of attention from the research community, many of which however overlooked the type of communication that a given VN may exhibit, assuming that they all exhibit a one-to-one (unicast) communication only. In this paper, we motivate the importance of characterizing the mode of communication in VN requests, and we focus our attention on the problem of embedding VNs with a one-to-many (multicast) communication mode. Throughout this paper, we highlight the unique properties of multicast VNs and its distinct Quality of Service (QoS) requirements, most notably the end-delay and delay-variation constraints for delay-sensitive multicast services. Further, we showcase the limitations of handling a multicast VN as unicast. To this extent, we formally define the VNE problem for Multicast VNs (MVNs) and prove its NP-Hard nature. We propose two novel approach to solve the Multicast VNE (MVNE) problem with end-delay and delay variation constraints: A 3-Step MVNE technique, and a Tabu-Search algorithm. We motivate the intuition behind our proposed embedding techniques, and provide a competitive analysis of our suggested approaches over multiple metrics and against other embedding heuristics.

Urban Air Pollution Monitoring System With Forecasting Models

A system for monitoring and forecasting urban air pollution is presented in this paper. The system uses low-cost air-quality monitoring motes that are equipped with an array of gaseous and meteorological sensors. These motes wirelessly communicate to an intelligent sensing platform that consists of several modules. The modules are responsible for receiving and storing the data, preprocessing and converting the data into useful information, forecasting the pollutants based on historical information, and finally presenting the acquired information through different channels, such as mobile application, Web portal, and short message service. The focus of this paper is on the monitoring system and its forecasting module. Three machine learning (ML) algorithms are investigated to build accurate forecasting models for one-step and multi-step ahead of concentrations of ground-level ozone (O3), nitrogen dioxide (NO2), and sulfur dioxide (SO2). These ML algorithms are support vector machines, M5P model trees, and artificial neural networks (ANN). Two types of modeling are pursued: 1) univariate and 2) multivariate. The performance evaluation measures used are prediction trend accuracy and root mean square error (RMSE). The results show that using different features in multivariate modeling with M5P algorithm yields the best forecasting performances. For example, using M5P, RMSE is at its lowest, reaching 31.4, when hydrogen sulfide (H2S) is used to predict SO2. Contrarily, the worst performance, i.e., RMSE of 62.4, for SO2 is when using ANN in univariate modeling. The outcome of this paper can be significantly useful for alarming applications in areas with high air pollution levels.

Wavelet Transform With Histogram-Based Threshold Estimation for Online Partial Discharge Signal Denoising

Online condition assessment of the power system devices and apparatus is considered vital for robust operation, where partial discharge (PD) detection is employed as a diagnosis tool. PD measurements, however, are corrupted with different types of noises such as white noise, random noise, and discrete spectral interferences. Hence, the denoising of such corrupted PD signals remains a challenging problem in PD signal detection and classification. The challenge lies in removing these noises from the online PD signal measurements effectively, while retaining its discriminant features and characteristics. In this paper, wavelet-based denoising with a new histogram-based threshold function and selection rule is proposed. The proposed threshold estimation technique obtains two different threshold values for each wavelet sub-band and uses a prodigious thresholding function that conserves the original signal energy. Moreover, two signal-to-noise ratio (SNR) estimation techniques are derived to fit with actual PD signals corrupted with real noise. The proposed technique is applied on different acoustic and current measured PD signals to examine its performance under different noisy environments. The simulation results confirm the merits of the proposed denoising technique compared with other existing wavelet-based techniques by measuring four evaluation metrics: 1) SNR; 2) cross-correlation coefficient; 3) mean square error; and 4) reduction in noise level.

Optimizing Cloud-Service Performance: Efficient Resource Provisioning via Optimal Workload Allocation

Cloud computing is being widely accepted and utilized in the business world. From the perspective of businesses utilizing the cloud, it is critical to meet their customers’ requirements by achieving service-level-objectives. Hence, the ability to accurately characterize and optimize cloud-service performance is of great importance. In this paper a stochastic multi-tenant framework is proposed to model the service of customer requests in a cloud infrastructure composed of heterogeneous virtual machines. Two cloud-service performance metrics are mathematically characterized, namely the percentile and the mean of the stochastic response time of a customer request, in closed form. Based upon the proposed multi-tenant framework, a workload allocation algorithm, termed max-min-cloud algorithm, is then devised to optimize the performance of the cloud service. A rigorous optimality proof of the max-min-cloud algorithm is also given. Furthermore, the resource-provisioning problem in the cloud is also studied in light of the max-min-cloud algorithm. In particular, an efficient resource-provisioning strategy is proposed for serving dynamically arriving customer requests. These findings can be used by businesses to build a better understanding of how much virtual resource in the cloud they may need to meet customers’ expectations subject to cost constraints.

Enhancing medical named entity recognition with an extended segment representation technique

OBJECTIVE The objective of this paper is to formulate an extended segment representation (SR) technique to enhance named entity recognition (NER) in medical applications. METHODS An extension to the IOBES (Inside/Outside/Begin/End/Single) SR technique is formulated. In the proposed extension, a new class is assigned to words that do not belong to a named entity (NE) in one context but appear as an NE in other contexts. Ambiguity in such cases can negatively affect the results of classification-based NER techniques. Assigning a separate class to words that can potentially cause ambiguity in NER allows a classifier to detect NEs more accurately; therefore increasing classification accuracy. RESULTS The proposed SR technique is evaluated using the i2b2 2010 medical challenge data set with eight different classifiers. Each classifier is trained separately to extract three different medical NEs, namely treatment, problem, and test. From the three experimental results, the extended SR technique is able to improve the average F1-measure results pertaining to seven out of eight classifiers. The kNN classifier shows an average reduction of 0.18% across three experiments, while the C4.5 classifier records an average improvement of 9.33%.

Predicting transformers oil parameters

In this paper different configurations of artificial neural networks are applied to predict various transformers oil parameters. The prediction is performed through modeling the relationship between the transformer insulation resistance extracted from the Megger test and the breakdown strength, interfacial tension, acidity and the water content of the transformers oil. The process of predicting these oil parameters statuses is carried out using two different configurations of neural networks. First, a multilayer feed forward neural network with a back-propagation learning algorithm is implemented. Subsequently, a cascade of these neural networks is deemed to be more promising. Both configurations are evaluated using real-world training and testing data and the accuracy is calculated across a variety of hidden layer and hidden node combinations. The results indicate that even with a lack of sufficient data to train the network, accuracy levels of 83.9% for breakdown voltage, 94.6% for interfacial tension, 56.4% for water content, and 75.4% for oil acidity predictions were obtained by the cascade of neural networks.

Fault Detection, Isolation, and Service Restoration in Distribution Systems: State-of-the-Art and Future Trends

This paper surveys the conceptual aspects, as well as recent developments in fault detection, isolation, and service restoration (FDIR) following an outage in an electric distribution system. This paper starts with a discussion of the rationale for FDIR, and then investigates different areas of the FDIR problem. Recently reported approaches are compared and related to discussions on current practices. This paper then addresses some of the often-cited associated technical, environmental, and economic challenges of implementing self-healing for the distribution grid. The review concludes by pointing toward the need and directions for future research.

Progressive recovery for network virtualization after large-scale disasters

Network virtualization allows users to build customized interconnected storage/computing configurations for their business needs. Today this capability is being widely used to improve the scalability and reliability of cloud-based services, including virtual infrastructures services. However, as more and more business-critical applications migrate to the cloud, disaster recovery is now a major concern. Although some studies have looked at network virtualization design under such scenarios, most have only studied pre-fault protection provisioning. Indeed, there is a pressing need to address post-fault recovery, since physical infrastructure repairs will likely occur in a staged progressive manner due to constraints on available resources. Hence this paper studies progressive recovery design for network virtualization and evaluates several heuristic strategies.