International Journal of Intelligent Engineering and Systems

11

Authors: Widi Aribowo

The improvement of the economy country along with developments in each sector will be followed by technological progress. This has led to an increase in electricity demand in Indonesia every year. Electricity load forecasting has an important role in the energy management system. The purpose of electricity load forecasting is as an effort to balance between electricity demand and electricity supply. Power system management can be said to be good if the planning carried out will make a major contribution to the development of electric power systems. In this study, the Feed Forward Backpropagation Neural Network method (FFBNN) is optimized using the Teaching-Learning-Based Optimization Algorithm (TLBO) to predict long-term electricity load. The results were measured and validated using the Mean Absolute Percentage Error (MAPE) method. The performance of the proposed method will be compared with the actual data, feed forward backpropagation neural network (FFBNN) and cascade forward backpropagation neural network (CFBNN). It was found that the TLBO-FFBNN method had an average MAPE value of 0.00004936%. The Comparison results show the effectiveness of the proposed method. The proposed method shows that is adapted and has a good MAPE value.

33

Authors: Ahmad Alwarid, Agus Sihabuddin

Fuzzy Time Series models have been developed in various ways, one of which is determining the intervals. Several methods were applied to determine the intervals, but the performances are still not optimal. This paper proposes a new approach that uses a combination of Absolute Differences and K-means Clustering in the Fuzzy Time Series Markov Chain model. K-means Clustering made the interval more flexible and compact based on the data it clustered. In addition, Absolute Differences was used as the based method to define how many intervals to be made. This study used Taiwan Capitalization Weighted Stock Index (TAIEX) as benchmark data to evaluate the proposed method, which produced an average Mean Absolute Percentage Error (MAPE) value of 0.42, and an average Root Mean Squared Error (RMSE) value of 51.09 for the test data. The proposed method outperformed other compared researches at the end of this paper in terms of prediction accuracy.

55

Authors: Fathima Ghouse, Kavitha Paranjothi, Revathi Vaithiyanathan

Dysgraphia is a handwriting disorder and the classification of dysgraphia in children's handwritten images helps to identify the dysgraphia patient effectively and also prevents low self-esteem. Traditional methods of dysgraphia classification based on experts manually classify the dysgraphia that requires more cost and time. Few researches apply the machine learning and deep learning techniques for the classification of dysgraphia and existing models have the limitations of overfitting problems in the training process. In this research, the Non-Discrimination Regularization in Rotational Region Convolutional Neural Network (NDR-R2CNN) is proposed to improve the efficiency of dysgraphia classification. The balancing parameters are introduced in the loss function to balance the class in the training and eliminate the features to reduce the overfitting problem. The collected children's handwriting data were used to evaluate the performance of the proposed NDR-R2CNN model. The proposed NDR-R2CNN model has the advantages of effective feature analysis and non-discrimination word analysis. The NDR-R2CNN model has the accuracy of 98.2 % and the SMOTE+SVM method has 90.2 % accuracy in Dysgraphia classification. The result shows that the NDR-R2CNN model has 98.2 % accuracy and the existing CNN has the accuracy of 94.2 %.

75

Authors: Nagaratna Parameshwar Hegde, Sireesha Vikkurty, Gnyanee Kandukuri, Sriya Musunuru, Ganapatikrishna Parameshwar Hegde

The COVID-19 pandemic has essentially transformed the way of leading a life for millions of people across the world. As offices remained closed for months, employees expressed conflicting sentimental analysis on the work-from-home culture. People worldwide use social media platforms such as Twitter to talk about their daily lives made a trend in the online platform. This research study aims to gauge the public's sentiment on working from home/ remote locations during the COVID-19 pandemic by tracking their opinions on Twitter. The existing random forest model trained the data faster but failed to predict the results faster. Therefore, an ensemble model is proposed to predict an outcome using a distinct modeling algorithm. An ensemble classifier has been used for enhancing the performances using the base learning classifiers such as Naive Bayes (NB), Random Forest (RF), Support Vector Machine (SVM), Logistic Regression (LR) form an Ensemble classifier. The proposed ensemble model aggregates each base model for the prediction and results for the unseen data. These tokens are then passed to the Ensemble classifier that classifies the sentiments and assigns a polarity (positive, negative, neutral) to every tweet. The proposed Ensemble method improve the average prediction performance over any contributing member in the ensemble. The results obtained by the proposed Ensemble model reached accuracy of 97.47 % when compared to the existing models such as Deep LSTM, SVM model that obtained accuracy of 83 %, 84.46 %.

95

Authors: Lukman Hakim, Huipeng Zheng, Takio Kurita

The neighboring relationship of the pixels is fundamental information in the image. Utilizing neighboring pixel information provides an advantage over using only pixel-to-pixel information. This study presents a regularization term based on the differences of neighboring pixels. We define the differences of neighboring pixels by using the Graph Theory approach. We then propose an effective graph Laplacian regularization to enforce the differences between pixels as the nodes and differences between nodes as edges. In our scheme, Graph Laplacian constructs from the output of the convolutional neural network and ground-truth image. The generated differences matrices from the output and ground-truth are combined as a Laplacian regularization term used as the deep convolutional neural network's new objective function. Experiments show that our scheme successfully captures pixel neighbor relations and improves the performance of the convolutional neural network better than the baseline without a regularization term. Qualitative and quantitative results show that our developed regularizer proved to enhance the boundary structure on image super-resolution and image segmentation tasks, which is achieved Structural Similarity Index (SSIM) of 0.9604 and 0.8263 on Cartoon set and Manga109 datasets and area under the curve (AUC) of 0.9740 and 0.9561 on DRIVE and STARE datasets, respectively.

117

Authors: Jirasak Ponchua, Suchada Sitjongsataporn

Explosive growth and usage of internet while COVID-19 pandemic demands an access technology in our lifestyle that means the Internet of Things (IoT) and Passive Optical Networks (PONs) are increasingly being used in order to support the digital lifestyle. Mobile Internet allows us firstly to exchange the content of information and multimedia while on the move. Then, IoT makes it possible to connect the smart devices with the low latency. PONs have been suggested for providing the ultra-low latency for Quality of Service (QoS) applications such as the Tactile Internet (TI) that is the next evolution to enable real-time control of IoT. A one-millisecond round-trip delay is required to support the tactile internet. As stated on the improving bandwidth allocation for PONs and Passive Optical LAN (POLAN), we propose the FiWi-Dynamic Wavelength and Bandwidth Allocation (F-DWBA) algorithm to reduce the round-trip delay at lower than one-millisecond, which the proposed F-DWBA algorithm based on the PONs and WLAN 802.11b should have a maximum performance. The lowest round-trip delay will make the Fiber-Wireless (FiWi) system can work efficiently and support TI applications. Simulation setup are tested on the latency budget of communication for the TI system starting from the Mobile Unit (MU) through the access point (AP) connected to the Optical Network Unit (ONU) including with the Optical Line Terminal (OLT) to the server. Performance analysis is focused on three periods of the latency time at the optical access network, radio interface and MU interface. Simulation results show that the proposed F-DWBA algorithm has a better round-trip delay than the Gigabit-Passive Optical Network (G-PON) standard about 40 percentage and better than the conventional Dynamic Wavelength and Bandwidth Allocation (DWBA) algorithm about 33 percentage at the full working of the tactile internet system.

141

Authors: Gemilang Santiyuda, Retantyo Wardoyo, Afiahayati

The adaptive Lévy flower pollination algorithm (ALFPA) is a recent addition to variants of flower pollination algorithm (FPA). Despite its excellent performance on single objective problem instances, it has shown inefficiency in the number of function evaluation (FE). Inspired by this, this paper proposed two algorithms extending ALFPA to solve multiple objective problem while also improving FE efficiency. The first algorithm proposed is ALFPA with non-dominated sorting denoted as MO-ALFPA alongside two variants which are proposed to improve its FE efficiency denoted as MO-ALFPAT and MO-ALFPAB. The second proposed algorithm, MOEA/D-ALFPA, uses decomposition strategy instead of non-dominated sorting on MO-ALFPA. The empirical study on two benchmark suits shows that MO-ALFPAT and MOEA/D-ALFPA performed better than other methods. Furthermore, MO-ALFPAT and MOEA/D-ALFPA produced the best results in three benchmark instances, based on inverted generational distance indicator, and two and three best results based on the hypervolume indicator, respectively.

167

Authors: Alek Nur Fatman, Tohari Ahmad

Information has been crucial in supporting daily activities. Many tasks must be accomplished by transferring not only public but also private information. This situation has brought the necessity for protecting it during transmission. Some mechanisms introduced, like data hiding, are still challenging, considering the quality of the data and the secret size can be protected. This research presents a reversible data hiding method for digital video based on prediction error and histogram shifting to approach those problems. There are six kinds of prediction errors used, three are created from two adjacent frames, and three are from a single frame. The selection of the shift location is determined by the histogram location, which has the highest peak point. The histogram is obtained by calculating the frequency of the subtraction of each value from its predicted value. The PSNR calculated from the stego and the original videos has an average of 73.45 dB. This value demonstrates that the proposed method outperforms the previous research.

188

Authors: Ravi Kumar Sanapala, Sreenivasa Rao Duggirala

Wireless Sensor Networks (WSNs) contain thousands of small and inexpensive sensor nodes distributed in the specific area for collecting the desired information from the environment. However, the sensor nodes have restricted battery energy, hence the random deployment of sensors creates issues while managing the resources for effective data delivery over the network. Therefore, this research aims to develop an energy-efficient cluster-based WSN to minimize the energy consumption of the nodes. In this paper, the Low-Energy Adaptive Clustering Hierarchy (LEACH) algorithm is used for selecting the optimal Cluster Heads (CHs) from the network. Subsequently, the Improved Spider Monkey Optimization (ISMO) is used for creating the optimal routing through the CHs to the Base Station (BS) where the multiple fitness functions such as residual energy, distance, and routing traffic are considered to optimize the ISMO. Hence, the data communication between the intermediate CHs in hierarchical cluster-based architecture helps to minimize the energy consumption of the nodes. The performance of the proposed LEACH-ISMO method is analyzed using network lifetime, total energy consumption, normalized overhead, throughput, End-to-End Delay (EED) and Packet Delivery Ratio (PDR). The existing methods namely Fuzzy C-Means (FCM) with Artificial Bee Colony (ABC), Meta-Heuristic Ant Colony Optimization based Unequal Clustering (MHACO-UC) and Optimized Quality of Service-based Clustering and Multipath Routing Protocol (OQoS-CMRP) are used to evaluate the LEACH-ISMO method. The PDR of the LEACH-ISMO method is 96.7 % for 200s, which is high when compared to the FCM-ABC, MHACO-UC and OQoS-CMRP.

214

Authors: Gouramma Halidoddi, Rubini Pandu

Wireless Sensor Networks (WSNs) have gained huge attention in different areas due to their self-configurability, easy maintenance and scalability features. WSN is configured with more nodes to transfer the data inside the network. The sensor networks are usually classified by low bandwidth, limited energy, limited power supplies and small size memory which leads to a very demanding environment to provide security. In recent years, accomplishing the network security objective with low energy utilization is a major issue in WSN. In this paper, two-level security is proposed to secure data transmission over the WSN. First, the secure clustering and routing operation are performed by using the Multi-Objective Trust-based Bat Optimization Algorithm (MOTBOA). Second, the data security of the netw¬ork is improved by the Enhanced Homomorphic Cryptosystem (EHC). This paper aims to offer secure communication in WSN. It is achieved by using the EHC method for the operation of encryption and decryption over the data during communication among the nodes. The performance of the proposed MOTBOA-EHC method is evaluated in terms of detection rate, delay, throughput, routing load and Packet Delivery Ratio (PDR). The existing methods namely Trust based Dynamic Source Routing (TDSR), Monarch-Cat Swarm Optimization-based routing protocol and Secured Quality of Service aware Energy Efficient Routing (SQEER) are used to evaluate the performances of the MOTBOA-EHC method. The PDR of the MOTBOA-EHC method is 92% at 100 rounds that are high when compared to the TDSR, MCSO and SQEER.

235

Authors: Jeny Varghese, Jagannatha Sreenivasaiah

Multiple cloud computing services are collectively managed as federated cloud. The growth of users into this federated cloud for accessing a variety of services has introduced challenges on resource utilization and load imbalance that consumes larger waiting time of users. To address these two issues, this paper proposes data center (DC) clustering, virtual machine (VM) clustering, resource mapping and task scheduling. Initially the DCs are clustered using region based fuzzy possibilistic C-means clustering (R-FPCM) algorithm. DC clustering is performed by gathering the information of data dependency, million instructions per second (MIPS), latency, storage, bandwidth and counts of VM. Depending on DC clustering, the VMs are clustered by multi-objective density-based spatial clustering based on the estimated capacity and bandwidth. In order to balance load, Markov chain is applied to predict future load and balance accordingly. An intermediate broker is employed to monitor the VM resources and map based on the user’s task requirement. Service level agreement (SLA) is satisfied for every user by the broker and then fast 1 to N resource mapping algorithm is involved for mapping resources. This is followed by entropy-based monotonic scheduling algorithm that arranges user tasks in an order that is determined from task type, task size, task arrival time and deadline. The dynamic computation of entropy enables to improve scheduling with the current status of the system. The extended simulations of this proposed system are simulated in Cloudsim and the results are evaluated in terms of execution time, latency, resource utilization and response time and compared with the performance of Capacity based VM clustering algorithm. These simulation results show that the performance of the proposed system is much better than the existing approach.

262

Authors: Pulung Nurtantio Andono, Guruh Fajar Shidik, Dwi Puji Prabowo, Dewi Pergiwati, Ricardus Anggi Pramunendar

Indonesia is rich in flora and fauna diversity, but it is experiencing an increasing number of endangered populations. The public can prevent the expanding population threatened with extinction by knowing the types of fauna, especially birds threatened with extinction. This study proposes an automatic grouping method to recognize bird sound patterns in an open environment. This experiment used bird data from a public dataset and obtained bird sound patterns by recording from a distance far enough to make sounds without feeling disturbed freely. However, the sound of birds may contain noise and need data processing using YAMME to be carried out the noise, and birds’ voices can be separated. After that, the data was extracted using the combination of Mel Frequency Cepstral Coefficients (MFCC) and Gammatone Cepstral Coefficients (GTCC) methods also reduced the dimensions of the feature before completing the identification. The bird identification obtained provides an accuracy performance that reaches 78.11%, and these results are higher than other feature extraction methods that also apply dimensional reduction.

282

Authors: Yunifa Miftachul Arif, Hani Nurhayati, Supeno Mardi Susiki Nugroho, Mochamad Hariadi

Halal tourism is one of the tourism products that have the prospect of contributing to economic growth in Indonesia. Therefore, the government needs to increase promotions to increase tourist interest in halal tourism destinations in Indonesia. Game is one of the alternative promotional media that can also function as an educational medium for choosing halal tourism that is fun for potential tourists. This study proposes a recommendation system to support knowledge sources in the Indonesian halal tourism game. We use destinations ratings-based multi-criteria recommender system (MCRS) to generate recommendation rankings as a reference for visualizing halal travel for players as potential tourists. This method improves the ability of the conventional tourism recommendation system, which is generally based on a single criterion. In this study, we use eight destinations rating criteria as a reference for calculating the recommender system in the halal tourism game. Each of these criteria is a reference for tourists' assessment of halal tourist destinations in Indonesia. Next, we integrate the cosine-based similarity technique in MCRS to measure the level of similarity between players and previous tourist data sets. This research's testing phase uses the theme of halal tourism destinations in the Batu City area. The test results show that the number and composition of the tourism destinations item rating as input of the recommender system affect the accuracy, precision, recall, and F1 scores. Based on 40 experiments with different tourism destination item rating input configurations, it shows that the average value of accuracy = 0.60, precision = 0.67, recall = 0.64 and F1 score = 0.65.

305

Authors: Priyanka Rajanikanth, Kalli Satyanarayan Reddy

The Last decade the demand for wireless communication has increased massively. Wireless Sensor Networks (WSN) have turned out as a potential solution for various real-time applications namely environmental monitoring, health and military applications. These networks follow the random placement strategy hence, identifying the location of sensor nodes is a tedious task. Moreover, these networks suffer from various issues such as limited power resources. This work, focusses on the sensor node localization and network lifetime enhancement of Wireless Sensor Networks (WSN). In localization, security is an important issue, so a trust-based model to secure localization is investigated in this work. Further, energy aware clustering and Cluster Head (CH) selection has been undertaken to prolong the lifetime of the network. Finally, a data aggregation model which decreases the network energy depletion by discarding the redundant data is presented. The outcome of the model which is proposed has been assessed and expressed in the form of localization error, energy depletion and alive node. The comparative analysis shows that the localization error for varied anchor nodes is obtained as 1.51, 2.28, 2.52 and 1.25 using existing techniques DMA (Distance Mapping Algorithm), MDS-Map (Multidimensional Scaling), DV-Hop (Distance Vector-Hop) and Proposed Approach, respectively. Similarly, the average energy consumption is obtained as 1.02, 1.67, 1.90 and 0.82 using DMA (Distance Mapping Algorithm), MDS-Map (Multidimensional Scaling), DV-Hop (Distance Vector-Hop) and Proposed Hybrid Approach. The comparative analysis shows that proposed hybrid approach of localization, CH (Cluster Head) formation and aggregation attains better performance when compared with existing techniques in WSN(Wireless Sensor Networks).

329

Authors: Cucun Very Angkoso, Hapsari Peni Agustin Tjahyaningtijas, I Ketut Eddy Purnama, Mauridhi Hery Purnomo

One of the objectives of medical imaging research is to develop an effective and reliable clinical support tool for the early detection of various neurological conditions in patients such as Alzheimer's Disease (AD). Classification based on three-dimensional (3D) MRI (Magnetic-Resonance-Imaging) images have shown a good performance. However, solving 3D object classification as a 3D object using classical machine learning or deep learning is computationally high. 3D images from MRI can be reconstructed into three depth directions: axial, coronal, and sagittal. However, the multi-view-based method cannot explore the content of each image slice for all planes in 3D-MRI. Our study proposes a multiplane method by using three two-dimensional (2D) CNNs to capture discriminatory information on all available planes. Our method is a new fusion strategy to deal with the disadvantages of shape-based multi-view techniques. Our framework selects three slices: the three largest 3-planes to represent whole 3D-MRI objects as multi-2DCNN inputs. Extensive experiments have shown that the proposed method can outperform the 2DCNN method, which uses only one plane. More importantly, by taking full advantage of 2DCNN, we offer a new method for identifying 3D objects that is both easy and efficient. We called this new architecture Multiplane Convolutional Neural Network (Mp-CNN) since it used multiple inputs in its design. We evaluated the proposed method using T1-weighted structural MRI data consisting of 500 AD, 500 Mild Cognitive Impairment (MCI), and 500 Normal Cognition (NC) subjects collected from the MRI database of Alzheimer's Disease Neuroimaging Initiative (ADNI). From the performed experiments, the proposed method achieves 93% accuracy for multiclass AD-MCI-NC and good precision AD 93%, MCI 91%, and NC 95%, respectively. Our study also demonstrates that the proposed multiplane approach outperforms the single-plane approach.

350

Authors: Muhammad Abdillah, Ronald Cornelius Batubara, Nita Indriani Pertiwi, Herlambang Setiadi

Maximum Power Point Tracking (MPPT) is a technique developed to obtain maximum power transfer on solar panels. Various MPPT methods developed can overcome various conditions that affect the power transfer of solar panels, such as irradiation fluctuations and an increase in the solar panels surface temperature. To handle the problem fuzzy logic controller (FLC) is employed as the MPPT method. The input of FLC utilizes the error (E) and delta error (∆E) which is obtained from solar panel or photovoltaic (PV) output including voltage Vpv and active power Ppv. Then, the output of FLC is to determine the appropriate duty cycle (D) of the single-ended primary inductor converter (SEPIC) converter. This proposed scheme is to get better performance in terms of stable voltage output produced by SEPIC converter. The novelty of this paper is the combination of MPPT based on FLC or MPPT FLC that have less ripple and SEPIC that can produce stable voltage output. Hence, the ripple out the load can be reduced significantly by using this proposed method. From the simulation results, it is found that when irradiation conditions and temperatures vary, MPPT FLC can change the operating point of the solar panels close to MPP to obtain maximum power transfer. In terms of power ripple, MPPT FLC can reduce ripple (the total ripple is around 4.9 W) compared to conventional MPPT such as perturb and observer (P&O) (the total ripple is around 9.2 W). Ripple reduction occurs in both irradiation and temperature variations.

370

Authors: Shaymaa Mahmood Mahdi, Safanah Mudheher Raafat

In this paper, the design and application of robust interactive PID control for a medical robot system is presented. The aim is to solve the problem of precise position tracking required for human walking, using a simple type of control while taking into account the interactions of the systems’ input and outputs. The medical robot system can be interpreted as pendulum links to obtain a complete motion of all muscles. Two control actions are required to be applied for the two joints. The interactions of this multivariable system have been examined using the relative gain array (RGA). In this work, PID and Fuzzy PID controllers are implemented with interactions effects. In addition, two other non-conventional types of PID control structures have been considered for comparison, where the RGA cannot be applied. These are the two-degree PID (2DoFPID) and nonlinear PID (NPID) controllers. Simulation results show that in the case of not considering the interaction effects, the Fuzzy PID controller can provide an improvement of 44% as compared with conventional PID controller. While this improvement has been increased to 70% with the consideration of interaction effects. Meanwhile, NPID controller shows higher rise and settling times (almost more than 2 seconds for the settling time). While 2DoFPID controller shows difficulty in implementation due to its highly chattered control signals with higher overshoots in the transient response as compared with the results of the interactive Fuzzy PID controller. Finally, robustness against disturbances and system uncertainties has been tested with all the developed controllers. It has been concluded that the interactive Fuzzy PID controller achieves better performance in terms of less ISE, settling time, rise time, and overshoot.

396

Authors: Ahmed Aldhahab, Ibrahim Murdas

Conventional logic gates failed to process data at a high bit rate. Therefore, researchers began to study the design of these gates with optical technology, which is characterized by high bit rate data processing. The non-linear behavior is a drawback in semiconductor optical amplifier (SOA). This nonlinearity property of the SOA is used in this paper to implement the OR optical gates. Mach–Zander interferometer (MZI)-based structures are among the numerous of structures that are available to transform a modulation of phase into a modulation of intensity. These are widely used by electro-optic (EO), or thermo-optic (TO) impacts to create various applications like optical modulators, splitters, switches, etc. In this paper, the MZI is used as switches and the SOA is employed with cross-gain modulation (XGM) to implement the optical logic gate OR at a high bit rate of 250 Gbps. Several experimental results were performed on the proposed design of the OR gate until the optimum value of the injection current, which is 0.08 A, was obtained. This optimum value led to achieve the lowest BER, highest Quality factor, and lowest power consumption, which are 0, 60.7989, and 31.13×10-3, respectively. The experimental results of the proposed design outperformed those results obtained by other state-of-the-art designs.

414

Authors: Andi Kurniawan Nugroho, Terawan Agus Putranto, I Ketut Eddy Purnama, Mauridhi Hery Purnomo

Commonly, clinicans have problems for recognising brain stroke injury images. However, with the advantages of Information technology it is expected that will be a new method that can support the clinicans’ opinion for recognising the brain stroke injury for type of stroke (hemorrhagic, ischemic, and normal). Therefore, this study aim is to discovery a new model to classify hemorrhagic, ischemic and normal based on Diffusion Weighted (DW)- Magnetic Resonance (MR) images. This study argues by using Qual Convolutional Layers (QCL-CNN) which applied in CNN can classified type of stroke. For this study experiment, this research conducted two experiment to asses the performance of QCL-CNN. The first experiments partitioned the MR image dataset into 20 percent testing and 80 percent training sets. Then, the second testing performed ten-fold cross-validation on the image dataset. The result from the first experiment of the classification accuracies obtained 93.90 percent (1st dataset) and 94.96 percent (2nd dataset). As for the second experiment, the results shows that the classification accuracies obtained 95.91 percent ( 1st data set) and 97.31 percent ( 2nd data set). The data source for this study gained from Indonesian hospital and the web sources dataset public from Ischemic Stroke Lesion Segmentation (ISLES). This study also compared, the QCL-CNN model with other architecture model such as AlexNet, ResNet50, and VGG16. The result of the comparison experiment shows that QCL-CNN architectures model has excellent performance than the others model.

441

Authors: Annisa Kesy Garside, Ikhlasul Amallynda

This paper considers the permutation flow shop scheduling problem to minimize total earliness and tardiness. We initiated a metaheuristic algorithm, namely a crossbreed discrete artificial bee colony. We modifications to the essential artificial bee colony and propose two versions of the crossbreed discrete artificial bee colony. Taguchi experimental design is used to test the performance of this. Several computational experiments using Vallada’s benchmark instances have been carried out to prove the performance of the proposed algorithms. The statistical test results show that the proposed algorithms have the largest negative mean of BRE’s value, each of -0.1959 for CDABC_ver1 and -0.1954 for CDABC_ver2. It means that the proposed algorithms perform significantly better than other algorithms. The results of the Kruskal Wallis H test show that the proposed algorithm has better performance than some dispatching rules and heuristic algorithms. Furthermore, the proposed algorithms can deliver better results in less time than mathematical model solution.

464

Authors: Lubna A. Alnabi, Abbas K. Dhaher, Mohammed B. Essa

The power loss in electrical networks is considered unavoidable, because of its inherent resistance, for effective and economical operation; network loss should be reduced to maximum extent. There are two goals for this study, the first one is extracting the optimal size and location of distribution generators (DG) and the optimal reconfiguration with the aim of decreasing power loss and enhance voltage profile, while the second objective is to prove the success of the methods of Newton Raphson (NR) and guess sidle(GS) in the study and analysis of radial distribution networks. In order to reduce power loss, the genetic algorithm (GA) is used to address the network distribution problem. Software package MATPOWER-7 and MATLAB are utilized to simulate a 33-bus testing system for achieving four cases of various Distribution generator (DG) numbers. A comparison with another previous studies done and the conclusion indicated a positive impact on the efficiency of the system, this work support loss reduction and voltage improvement with the lowest Distribution generator (DG) size and prove that the increase of Distribution generator (DG) number not always can give better result though system cost increase, maintenance, along with the units’ distance for gas supplying increases. There are 4 cases with distinctive Distribution generator (DG) number applied in this paper and a compassion is done of power losses and voltage profile with references where case4 show better result with voltage profile 0.96 and power losses 59.09. As load flow method the Gauss-Seidel (GS) method used because it is a simple iterative method for solving n number of load flow equations by using iterative method. Where partial derivatives not require. And a Newton-Raphson (NR) method used where it is based on Taylor’s series and partial derivatives. The Newton-Raphson (NR) method advantages are less number of iterations needs to reach convergence, takes less computer time when computation cost is less and the convergence is definite. Also it is more accurate, and is insensitive to factors like slack bus selection, regulating transformers. and the number of iterations required in this method is almost independent of the system size. Genetic algorithm (GA) has a good ability to making a global search and explore the search space using their different kind of crossover. Also you can combine it with local search method to increase the exploitation search of the Genetic algorithm. In this paper the introduction part introduced where it is the general part of the paper then the research method proposed in section 3 with details then a Genetic algorithm proposed finally the result with conclusions are proposed.

490

Authors: Nam-Soo Kim

Recently, the non-orthogonal multiple access (NOMA) technique, which transmits multiple users in the same time-frequency block, has been focused on for future cellular networks. Conventional NOMA-based aerial cellular networks consider an unmanned aerial vehicle (UAV) as an aerial user, in which secure control information for UAV is important. In recent days, the deployment of urban air mobility (UAM) has been actively discussed for a new aerial transport service in an urban area. Since UAM conveys passengers, both the traffic and the control information for passengers and a vehicle are important. Therefore, this paper considers a NOMA-based aerial cellular network with a UAM as an aerial user and a ground user as a terrestrial user. And we derive closed-form outage probabilities of the forward and reverse links of UAM, and the Monet-Carlo simulation verifies the results. Further, the condition to maintain the balance between the forward and the reverse links for the identical quality of service (QoS) is derived. Numerical results show that the outage probability of the forward link decreases with the increase of the received signal-to-noise ratio (SNR). However, we noticed that the outage probability of the reverse link shows the error floor, which does not decrease with the received SNR and remains constant, caused by the interference from the ground user (GU). Also, it is shown that there is an average transmit power of UAM to conform the link balance regardless of strong or weak users.

509

Authors: Riyadh Rahef Nuiaa, Selvakumar Manickam, Ali Hakem Alsaeedi, Dhiah Eadan Jabor Al-Shammary

Distinguishing between network traffic activity, intrusion, and normal bahavour is very difficult and very time-consuming. An analyst has to review all the large and wide data to find the order of intrusion in the network connection. Therefore, a method that can detect network intrusion and reflect the current network traffic is required. In this paper, a new EDFC model (Evolving Dynamic Fuzzy clustering) algorithm is generated to improve and enhance the detection mechanism. The proposed model contains two parts: the cluster part and evolving part. This paper's main objective is to design and implement a novel and data density-based clustering scheme that provides high system performance and persistent grouping of data with high similarity and performance on big data for efficient machine learning. Compared to previous techniques, the suggested model's performance with several standard datasets such as the UNSW-NB15 dataset, KDD99 dataset, and NSLKDD dataset indicates a higher silhouette coefficient. In the EDFC model two metrics have been used to verify the quality of clusters, and these are the silhouette coefficient and the number of clusters. The EDFC model has achieved a high silhouette coefficient, and a low number of clusters compare to other models. Our focus is to enhance the detection mechanism for the DRDoS_DNS attacks. Therefore, The EDFC model has been implemented on the standard CICDDoS2019 dataset which contains DRDoS_DNS attacks and achieved a silhouette coefficient of 0.76 and a number of cluster 13.

530

Authors: Purba Daru Kusuma, Ratna Astuti Nugrahaeni

This work proposes a new flow-shop scheduling model that consists of several flow-shops. Every flow-shop acts as an independent entity, but there is a collaboration among them. Although the relation between flow-shops and their customers is exclusive, collaboration through production sharing is possible. This circumstance is different from most studies in flow-shop scheduling problems (FSP), for example, parallel or distributed, where all jobs come from a single point and are then distributed to the production resources. In the multiple independent flow-shops, each flow-shop has its own processing time and production cost. Through collaboration, efficiency can be achieved in the make-span and total cost aspects, which becomes the objective of this work. This model is developed by combining the first price sealed bid auction and cloud theory-based simulated annealing. The first price sealed bid auction is conducted to minimize the total production cost. Meanwhile, the cloud theory-based simulated annealing is conducted to minimize the make-span. This model is then compared with the existing non-dominated sorting genetic algorithm (NSGA II) based flow-shop scheduling models. The first existing model is a parallel flow-shop, while the second one is a collaborative flow-shop. The simulation result shows that the proposed model outperforms the existing models in the total cost aspect. The proposed model creates a 13 to 29 percent lower total cost than the NSGA II-based parallel flow-shop. Meanwhile, the proposed model creates a 16 to 28 percent higher make-span than the NSGA II-based parallel flow-shop.

549

Authors: Abdelghafour Herizi, Riyadh Rouabhi

The efficiency of a wind turbine depends primarily on the power of the wind, the power curve of the turbine and the ability of the generator to respond to fluctuations in the wind. This article then proposes a robust control strategy of the doubly-fed induction generator applied in the wind energy conversion system, the active and reactive powers, which are generated by the DFIG will be decoupled by the orientation of the stator flux. The proposed control is combined between sliding mode control and type-2 fuzzy logic, this technique is robust to the modeling uncertainties of the generator and the wind generator, and which makes it possible to optimize energy production (efficiency). Finally, the performance of the system was tested and compared by simulation in terms of follow-up of instructions and the robustness with respect to the parameter variations of the DFIG. The results obtained show the interest of such a control in this system.

575

Authors: Mala Sinnoor, Shanthi Kaliyil Janardhan

The Electrocardiogram signal (ECG) is highly susceptible to the electrical environment from where the motion artifacts were being recorded. The accurate representation of the ECG signal will necessitate removing the noise from various sources thereby resulting in a noise-free model. This present research work uses Multi scale Local Polynomial Transform (MLPT) technique that provides wavelet transform as an alternative when the MLPT model was non-equispaced. The proposed research combines two approaches such as MLPT and Ensemble Empirical Mode Decomposition (EEMD) forming a Hybrid Transform Model which is used for denoising in the research. The present research work considers white Gaussian noise and experimental results are based on the MIT-BIH Physionet Database. The results of the proposed hybrid MLPT-EEMD obtained better SNR values of 25.93 dB when compared to the existing Empirical Mode Decomposition technique (EMD) of 5.43dB, EMD with adaptive switching mean filter obtained 9.135 dB and S Transform based Time–Frequency Filtering Approach obtained 13.82 dB.