There are many methods or algorithms applicable for detecting electricity theft. However, comparative studies on supervised learning methods for electricity theft detection are still insufficient. In this paper, comparisons based on predictive accuracy, recall, precision, AUC, and F1-score of several supervised learning methods such as decision tree (DT), artificial neural network (ANN), deep artificial neural network (DANN), and AdaBoost are presented and their performances are analyzed. A public dataset from the State Grid Corporation of China (SGCC) was used for this study. The dataset consisted of power consumption in kWh unit. Based on the analysis results, the DANN outperforms compared to other supervised learning classifiers such as ANN, AdaBoost, and DT in recall, F1-Score, and AUC. A future research direction is the experiments can be performed on other supervised learning algorithms with different types of datasets and suitable preprocessing methods can be applied to produce better performance.
Brain magnetic resonance imaging (MRI) classification into normal and abnormal is a critical and challenging task. Owing to that, several medical imaging classification techniques have been devised in which Learning Vector Quantization (LVQ) is amongst the potential. The main goal of this paper is to enhance the performance of LVQ technique in order to gain higher accuracy detection for brain tumor in MRIs. The classical way of selecting the winner code vector in LVQ is to measure the distance between the input vector and the codebook vectors using Euclidean distance function. In order to improve the winner selection technique, round off function is employed along with the Euclidean distance function. Moreover, in competitive learning classifiers, the fitting model is highly dependent on the class distribution. Therefore this paper proposed a multiresampling technique for which better class distribution can be achieved. This multiresampling is executed by using random selection via preclassification. The test data sample used are the brain tumor magnetic resonance images collected from Universiti Kebangsaan Malaysia Medical Center and UCI benchmark data sets. Comparative studies showed that the proposed methods with promising results are LVQ1, Multipass LVQ, Hierarchical LVQ, Multilayer Perceptron, and Radial Basis Function.
The microgrid communication network with proper connectivity among microgrid resources is play important role to maintain a stability and reliability of the microgrid. Application of suitable communication network and protocol and highlighted the best security measurement is one of the elements to achieve those broad objectives. The communication network and protocol that has been implemented in existing microgrid has different types and objective which is depend on specific application. To secure the communication network and protocol, many security approaches is proposed. In this paper, a review of microgrid communication and its security is shown and future direction of communication network and protocol with its security also provided.
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Serial crime recognition is a critical task. Usually, police officer investigates the serial crime behavior based on their heuristics, evidence or prior information from public. Sometimes, the police officer makes inadequate decision when handling the serial crime problems due to lack of preliminary study on relationship between serial crime and amenities. Therefore, this study explores k-means to identify pattern of surroundings area at serial comersial crime scene. In Malaysia, precisely Selangor, Wilayah Persekutuan Kuala Lumpur and Wilayah Persekutuan Putrjaya, a set data of serial crime including index and non-index, and its surroundings area at crime scene are being investigated. Experimental result shows that ‘hot spot’ amenities such as bank, commercial center, restorant, place of worship, resident and school are highly involved with three types of crime namely house breaking at night, day and robbery without firearm. Furthermore, radius distance with 0.2 km and 0.3 km between the crime scene location and its amenities at surroundings area captured from Safe City Monitoring System are also being evaluated and analyzed. Consequently, our finding helps the police to easily observe and prevent criminal behavior by assigning necessary human resource based on their ‘hot spot’ amenities.
Harmony Search (HS) is the behaviour imitation of a musician looking for a balanced harmony. HS has difficulty finding the best tuning parameter, especially for Pitch Adjustment Rate (PAR). PAR plays a crucial role in selecting historical solution and adjusting it using Bandwidth (BW) value. PAR in HS requires a constant value to be initialized. Furthermore, there is a delay in convergence speed due to the disproportion of global and local search capabilities. Although some HS variants have claimed to overcome this shortcoming by introducing the self-modification of PAR, these justifications have been found to be imprecise and require more extensive experiments. Local Opposition-Based Learning Self-Adaptation Global Harmony Search (LHS) implements a heuristic factor, η for self-modification of PAR. It (η) manages the probability for selecting the adaptive step, either global adaptive step or worst adaptive step. If the value of η is large, the prospects of selecting the global adaptive step is higher, thereby allowing the algorithm to exploit a better harmony value. Conversely, if η is small, the worst adaptive step is prone to selection, therefore the algorithm is closed to the best global solution. In this paper, in addressing the existing HS obstacle, we introduce a Cosine Harmony Search (CHS) which incorporates an additional strategy rule. This additional strategy employs the η inspired by LHS and contains the cosine parameter. This allows for self-modification of pitch tuning to enlarge the exploitation capabilities. We test our proposed CHS on twelve standard static benchmark functions and compare it with basic HS and five state-of-the-art HS variants. Our proposed method and these state-of-the-art algorithms are executed using 30 and 50 dimensions. The numerical results demonstrated that the CHS has outperformed other state-of-the-art algorithms in terms of accuracy and convergence speed evaluations.
The infrastructure of and processes involved in a microgrid electrical system require advanced technology to facilitate connection among its various components in order to provide the intelligence and automation that can benefit users. As a consequence, the microgrid has vulnerabilities that can expose it to a wide range of attacks. If they are not adequately addressed, these vulnerabilities may have a destructive impact on a country’s critical infrastructure and economy. While the impact of exploiting vulnerabilities in them is understood, research on the cybersecurity of microgrids is inadequate. This paper provides a comprehensive review of microgrid cybersecurity. In particular, it (1) reviews the state-of-the-art microgrid electrical systems, communication protocols, standards, and vulnerabilities while highlighting prevalent solutions to cybersecurity-related issues in them; (2) provides recommendations to enhance the security of these systems by segregating layers of the microgrid, and (3) identifies the gap in research in the area, and suggests directions for future work to enhance the cybersecurity of microgrids.
No abstract
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.