International Journal of Intelligent Engineering and Systems

11

Authors: Boonlert Suechoey, Chongrak Boonseng, Chow Chompooinwai, Chai Chompoo-inwai

Under renewable generations mixed in today’s modern grid system, many of renewable generation resources tend to increase the fault level to an existing power system. Rules and regulation for power equipment and devices need to be revised and updated, of course, no exception for the distribution transformers. This article, therefore, presents the evaluation and analysis of the winding temperature on both loading and short circuit conditions of an oil-immersed distribution transformer and propose the novel transformer design method for improving short circuit withstand capability. In the study, the methodologies of measuring and estimating the winding temperature of the transformer during short circuited are presented and implemented. In the study, various winding parameters are analysed including: the winding temperature, the hottest temperature of the winding after a short circuit, the short circuit current, short circuit force and short circuit duration. The tested and analysed result are benefit for the newly proposed distribution transformer design of a 400 kVA 3 phases 50 Hz 22 kV-400/230 V, Dyn11. The new design approaches will enable designers to find a weak spot and proper selection of raw materials, such as winding size, insulation thickness and properties of the silicon steel for a better quality of distribution transformer. Moreover, the new design can offer lower winding temperature rise of transformer while loading or experiencing with short circuit conditions meaning that it can prolong transformer insulation and extend transformer lifetime.

33

Authors: Jayanthi Chandrashekar, Arun Manoharan

Providing a reliable and secure communication in Mobile Adhoc Network (MANET) is a highly challenging and demanding task in recent days. For this purpose, different techniques are developed in the traditional works, but it lacks with the drawbacks of difficult key revocation, reduced security, increased delay, and bandwidth usage. To solve these problems, this paper motives to develop a routing technique, namely, Identity Based Key Management (IBKM) for providing a secure communication in MANET. Here, the group polynomial equation is generated at first for generating a unique ID to each node in the network. Then, the neighboring nodes of the source and destination are identified and registered in the network table. After that, the routing path between these nodes are validated in order to enable a reliable communication. The node pairing is performed and the public key is distributed to the nodes that are successfully paired. Sequentially, the signature of the packet is generated and verified, if both generated and received signature are same, the packet is received; otherwise, the misbehavior node is identified and blocked. To evaluate the results of the proposed mechanism, various performance measures such as malicious node detection probability, average message overhead, Packet Delivery Ratio (PDR), end-to-end delay, throughput, and routing overhead are used. Here, the results are evaluated with respect to various malicious node ratio, speed (m/s), and simulation time (s). Also, the existing Dynamic Source Routing (DSR) and Cooperative Bait Detection Scheme (CBDS) are compared with the proposed mechanism for proving the efficiency. From the results, it is analyzed that the PDR and throughput of the network is increased with the reduced message overhead and delay. Also, the proposed mechanism uses the identity based key management technique, which efficiently detects the malicious nodes in the network.

52

Authors: Sunil Swamilingappa Harakannanavar, Prashanth Chikkanayakanahalli Renukamurthy, Kori Basava Raja

Personal identification based on face recognition is receiving extensive attention over the last few years in both research and real-time applications due to increasing emphasis on security. In this paper, Face Recognition using the Fusion of bit-plane and binary image compression techniques is presented. Face images are resized to 256×256 to obtain the uniformity in the size of face images. The Bit-Plane compression algorithm is applied on the resized image to extract the features of face. Similarly, the Binary image compression technique is applied on 256×256 resized images to extract the features. The features produced from both the compression techniques are added to obtain the final facial features. Finally, the test features and database features are compared using Euclidean Distance classifier. Extensive experiments are conducted on combined, Indian female, Indian male and JAFFE face database. The true success rate of the proposed model provides better recognition rate than other existing state-of-the-art methods.

75

Authors: Raja Murali Prasad, Gandluru Arthur Edwin Satish Kumar

As cluster heads chosen by nodes, gathers the data from their individual groups and exchanges the collected data to the mobile sink as proposed in the heuristic tour planning technique. In this research, the mobile sink starts the collection of data irregularly from the initial site, which is stored in cluster heads in an individual hop extend and at last returns to the initial site. In existing approaches, scheduling time for data delivery is less. To increase the data delivery, there is a need in extending the concept to higher level routing process. To solve this issue, this paper proposed a collection oriented distribution scheme with k-medoids (CODS-KM). The method can be exhausted for distribute the collection points in network in a particular manner. In this method, k-medoids algorithm for clustering and cluster head process is used. To collect the data purpose, it can be used CH (Cluster Head), CP (Collection Point) and RP (Rendezvous Point). The proposed system helps in mitigation of travelling time when compared to existing approaches. To find the effectiveness of proposed algorithm, NS2 (Network simulator-2) software is used for simulation verification.

95

Authors: Ngoc Thuy Pham, Diep Phu Nguyen, Khuong Huu Nguyen, Nho Van Nguyen

In this paper, a novel Stator Current Model Reference Adaptive System (SC_MRAS) based scheme using Neural Network (NN) for sensorless controlled of Six Phase Induction Motor (SPIM) drive is presented. For this scheme, the measured stator current components are used as the reference model to avoid the pure integrator problems and reduce influence of motor parameter variation. The two layer linear NN stator current observer is used as an adaptive model, which has been trained online by means of a Back Propagation Network (BPN) algorithm, to estimate rotor speed and update stator resistance value. The Voltage Model (VM) based rotor flux identifier with value of stator resistance is updated online is used to provide the rotor flux for the adaptive model, this helps to overcome the instability problem in the regenerating mode of operation that occur when using current model for rotor flux identifier and enhance the low speed operation performance of the proposed scheme. Simulation results have demonstrated that the proposed MRAS observer is significantly improved especially at low and near zero speed range, Sensorless results also confirm the instability problem not exist in the regenerating mode in low speed region.

118

Authors: Tulasi Sunitha Manepalli, Chamakuzhi Subramanian

Rainfall is considered as the most important phenomena of the climate system. Due to the lack of adequate irrigation facilities, agriculture becomes vulnerable, which is the backbone of a country's economy. The rainfall can be able to predict by using selective appropriate predictors. Though several models have been developed for forecasting and predicting in Time Series (TS), there is no ideal model to predict the rainfall. In recent years, Automata is useful for forecasting and prediction of hydrological TS because automata help to predict the rainfall from the uncertainty data. The motivation of this work is to design a reliable tool for predicting daily rainfall in advance using Regression Automata (RA) models. The proposed method uses three different RA models for predicting rainfall from the collected data for four stations in Queensland State. The results clearly show that the all the three RA models can predict the rainfall very efficiently in various terms such as error rate, coefficients and mean square error.

137

Authors: Kishore Verma Samraj, Rajesh Appusamy, Ramya Ravi Shankar

Increase in the use of digital platforms generate a very huge amount of user data which on processing reveals more valuable information about users while mining or it may also reveal some future events. Privacy preserving data mining (PPDM) is the current developing area of research that precisely ensures a certain level of privacy by increasing the utility of information. Data Anonymization is a PPDM technique that protects sensitive information in the recordset with high confidence. Anonymizing recordsets that consists of null values habitually referred as deficient recordsets, suffer from serious information loss owing to null value contamination. This would be arisen because of the null values present in the original recordset. In this paper, we proposed an enhanced ARX framework to anonymize the deficient recordsets through a higher level of privacy and utility. We explore the characteristics of null value contamination on generalization, driven by these characteristics, we formulate an Enhanced ARX Anonymization- (K,R_n^N )Anonymity Model to preserve data utility on deficient recordsets. The results obtained by executing our framework over the real - world dataset proved to be optimized in providing better trade-off between utility and privacy for deficient recordsets in Cell oriented Anonymization (CoA), Attribute oriented Anonymization (AoA) and Record oriented Anonymization (RoA) than the existing procedures.

156

Authors: Vinay Balasubramanyam, Kalappan Balachander

Electromyography (EMG) signal is one of the important tools for the detection of skeletal related muscular information. In this scenario, we analyse the normality and abnormality of knee movements for pre-predicting the knee dystonia diseases. It is characterized by involuntary sustained muscle contractions affecting one or more sides of the body, frequently causing twisting and repetitive movements or abnormal postures. In this paper, EMG based knee dystonia pre-prediction is assessed by employing EMG-Lower Limb dataset. Then, transformation is done by using discrete wavelet Transformation (DWT), which delivers a superior time and frequency localization ability. After transformation, hybrid feature extraction is performed by employing yule-walker, burg’s, voltage level (minimum and maximum), Renyi entropy, and Peak-Magnitude to Root Mean Square Ratio (PMRS) for achieving optimal feature subsets and also to reject the redundant and irrelevant features. In most of the existing studies, an individual feature or two different features are combined for extracting the feature values from the acquired signal. In this research, five effective entropy and autoregressive features are combined to obtain more active features. After obtaining the active features, a superior binary classifier: Kernel Nearest Neighbour (KNN) is implemented for classifying the normality and abnormality of knee movements. The experimental outcome proves that the proposed methodology effectively distinguishes the normal and abnormal knee movement in terms of sensitivity, specificity, recall, precision, accuracy and E-rate. The proposed methodology improves the classification accuracy in knee movement detection upto 0.785-1.65% compared to the existing methods.

177

Authors: Deepa Nataraj, Rajaji Loganathan, Moorthy Veerasamy, Viswanatharao Jawalkar

A novel metaheuristic optimization procedure for optimizing electrical distribution system is proposed that may help the system planners to know the minimum and maximum objective values. In which the optimal distribution network reconfiguration (DNR) framework and implementation of modified grey wolf optimization (mGWO) algorithm to minimize power loss reduction and node voltage deviation indices have delineated. The DNR involves nonlinear and multimodal function that has been optimized under practical constraints. For the purpose, mGWO algorithm is employed for ascertaining optimum switching position while reconfiguring the distribution system at a minimum fitness value. In fact grey wolf’s update its position linearly from a higher value to zero in the search vicinity that provides perfect balance among intensification and diversification to ascertain the best fitness function and exhibits quick and steady convergence. Moreover, the proposed method seems to be a promising optimization tool for the electrical utilities, thereby modifying their operating strategy of distribution system under steady state conditions. This study is conducted on standard IEEE 33-bus and 69-bus distribution systems, the simulated results have analyzed and compared with several recent methods. It shows that the numerical results have provided, by mGWO is superior among the contestant algorithms.

199

Authors: Medi Sandhya Rani, Redamalla Rekha, Kota Venkata Naga Sunitha

Mobile Ad-Hoc Network (MANET) is a self-configuring network of mobile nodes associated by wireless channel. The nodes are free to move arbitrarily anywhere in the network. Hence, lack of fixed infrastructure and host flexibility, providing secure communications is a big challenge. Because a temporary device recurrently joins or leaves a network, the authentication and security techniques should be equipped for the unorganised users. In cryptography it is ineffective if its key management is fragile. Key management is one of the vital aspects for security. Thus, the network's wireless topology may be disorganised and may adjust rapidly, the efficient-route is recognised using Multi-path AODV Routing-Protocol. The ID based SKMS (Secret Key Management Scheme) method is implemented in secured environment by using Rivest-Shamir Adleman (RSA) method and the Node ID based Secure Key Management Scheme (SKMS) mechanism for securing data packets from source to destination. The ID based is used to share secret key of an individual users. A multi-cast (AODV) group receiver or group-source-key pair is conserved and recognised by an ID. Each clustering group has its own exclusive identity number. “ID based Multicast SKMS using RSA” method increase in packet delivery ratio (6%) along with the decrease in drop (5%), delay (8%), and energy consumption (6%) in secured ad-hoc environment.

220

Authors: Saïd Nouh, Moulay Seddiq El Kasmi Alaoui, Mostafa Belkasmi, Abdelaziz Marzak

Quadratic Residue codes are among the best codes. They have high capacity of error correction but they are very difficult to enumerate and therefore to analyse. Despite all developed methods in this domain, the weights enumerators of Quadratic Residue codes are known only for lengths less than or equal to 167. For the lengths 191 and 199 only estimations are available. In this paper, we present a new method based on the Multiple Impulse Method (MIM) and hash techniques to find the weights enumerators of Quadratic Residue codes having lengths in the form 8m-1, for an integer m. The proposed method Hash_MIM_Weights_Enumerators is validated on all Quadratic Residue codes of known weights enumerators; its reduced spatial and temporal complexities yields to new important results. So, the weights enumerators for the lengths 191, 199 and 223 are determined. These three codes are the best binary linear block codes in terms of minimum distance known until today and their analytical performances are remained unknowns in more than 60 years ago and they are available now.

244

Authors: Alugoju Sreelatha, Chagunta Venkata Guru Rao, Porika Sammulal

Cross Media Retrieval (CMR) is one of the emerging research areas in the field of internet services and multimedia technology. The primary objective of this paper is to develop an effective dictionary learning methodology to reduce the semantic gap between the low-level and high-level features. In this experimental research, the image and text data in the Wikipedia dataset are separated, after separating the image and text data, feature extraction is performed individually. A suitable combination of feature extraction methods (normalized histogram colour feature and bag-of-words) are under-taken for image and text feature extraction. The normalized histogram colour feature has very quick generation, compared to other feature vectors. In addition, Bag-of-words measures the repetition of the words in the bag and retrieves the data similar to the query data. Also, it removes the conjunction words and the remaining words are named as keywords. These keywords are used as the query text and it compared to the proposed dictionary learning. Then, retrieves the text and image related to the keywords based on Minkowski distance measure. Finally, the experimental outcome shows that the proposed approach delivers better performance in terms of Mean Average Precision (MAP) value, retrieval efficiency, precision and recall. The proposed methodology improved the MAP value up to 0.24-0.20 compared to the existing methods.

261

Authors: Adna Sento, Yuttana Kitjaidure

This paper presents a new model of the Neural Network PID-Like controller using an Actor-Critic reinforcement algorithm, called the Neural Network PID-Like controller using an Actor-Critic reinforcement algorithm (NNPID-AC). The proposed NNPID-AC controller is designed to develop the performances and the speed of calculation under the iterative learning algorithm. In the learning algorithm, the critic algorithm receives the reward value and control input to criticize the current state using the action-state value function approximation. Furthermore, instead of applying every available action to predict the local successor state, the algorithm only uses one-step estimation using the fifth degree spherical-radial cubature rule algorithm. To evaluate the proposed NNPID-AC controller, the robot arm MATLAB simulations have been implemented and provide the control system with the load and noise to prove the robustness and fault tolerance, respectively. From the results, the robot arm control system simulation under the control of the proposed NNPID-AC controller can potentially track the error and gives the best responses compared with the other conventional controller either with or without the load and the noise disturbance.

282

Authors: Sindia Thuraipandian Vimala, Julia Punitha Malar Dhas

The usage of cloud computing is skyrocketing now-a-days and so is the network traffic. The adversaries intend to attack the cloud servers due to some intentional reasons. The most frequent attack is the data theft, which is followed by the DDoS attack. Though numerous solutions exist to handle DDoS attack, SDN based solutions for cloud computing is scarce. Understanding the need of the DDoS attack detection system, this work proposes a SDN based solution for detecting DDoS attacks in cloud computing environment, which relies on ensemble classifier. This work collects the real time traffic data with the help of Wireshark network analyser tool. The ensemble classification relies on the classifiers k-Nearest Neighbour (k-NN), Support Vector Machine (SVM) and Extreme Learning Machine (ELM). The performance of the proposed approach is analysed in terms of accuracy, sensitivity, specificity and the results are compared with the existing approaches. Additionally, in order to prove the potentiality of the ensemble classification, this work employs the classifier individually and the results are compared. Finally, the average attack detection time is measured and compared. From the experimental results, it is observed that the proposed approach proves better results in terms of accuracy, sensitivity and specificity.

300

Authors: Abdelali Zakrani, Mustapha Hain, Abdelwahed Namir

There is evidence that Software Development Effort Estimation (SDEE) plays a crucial role in managing the software project and controlling its whole lifecycle; an accurate effort estimate allows an effective monitoring and efficient scheduling of tasks and resources. Although extensive research has been carried out on SDEE techniques, no single technique has been shown to be superior to other in all situation. Recently, there has been an increasing amount of literature on predicting software effort using Machine Learning (ML) methods. Among these ML techniques, regression tree-based models have gained a considerable attention due to their generalization ability and understandability. So far, very few studies have investigated the potential of Random Forests (RF) in software effort estimation. In this paper, a RF model is designed and adjusted empirically by varying the values of its key parameters. Prior to the parameters adjustment, we analysed their impact on RF model accuracy which allows an efficient tuning of the model during the training stage. The performance of the RF is then evaluated and compared with that of classical Regression Trees (RT). The evaluation was performed through the 30% hold-out validation method using five datasets: ISBSG R8, Tukutuku, COCOMO, Desharnais and Albrecht. To identify the most accurate technique, we employed three widely known accuracy measures: Pred(0.25), MMRE and MdMRE. The results obtained show that the adjusted random forest outperforms the regression trees model on all evaluation criteria. Moreover, the proposed model performs better than some recent techniques reported in the literature for software effort estimation.