Indonesian Journal of Electrical Engineering and Informatics (IJEEI)
Not a member yet
776 research outputs found
Sort by
Feature Optimization for Machine Learning Based Bearing Fault Classification
Full text linkThe most critical and essential parts of rotating machinery are bearings. The main problem of the bearing fault classification is to select the fault features effectively because all extracted features are not useful, and the high-dimensional features give poor performances and slow down the training process. Due to the effective feature selection problem, the bearing fault diagnosis method does not achieve a satisfactory result. The main goal of this paper is to extract the effective fault features with an optimization technique to classify the bearing faults using machine learning algorithms. Since wavelet entropy can determine complexity and degree of order of a vibration signal, this research uses it in features optimization. The proposed wavelet entropy-based optimization technique reduces the dimensionality of input, elapsed time and raises the learning process. Four Machine learning algorithms (naïve Bayes, support vector machine, artificial neural network and KNN) are applied to classify the bearing faults using the optimized features. To evaluate the proposed method, Case Western Reserve University’s (CWRU’s) bearing dataset is used which consists of three types of bearing faults. The accuracy and robustness of the bearing fault classification are tested by adding noise to the vibration raw signals at various levels of Signal-to-Noise Ratio (SNR). Experimental results show that the proposed method is very highly reliable in detecting bearing faults compared to the conventional methods
Solving Dynamic Combined Economic Environemental Dispatch Problem with Renewable Energies and Constraints Using Gorilla Troops Optimizer
Full text linkThe primary goal is to optimize the hourly allocation of power generation outputs by minimizing operational costs, pollutant emissions, and transmission losses, and ensuring compliance with a range of equality and inequality constraints. To tackle this challenge, a novel metaheuristic algorithm inspired by gorilla’s behavior is proposed. Gorilla Troops Optimizer (GTO) was applied to 5- and 10-generator unit systems, integrating variable wind and solar energies over a day with varying load demands. To demonstrate the effectiveness of the GTO algorithm in handling the hybrid dynamic combined economic and environmental dispatch problem, including equality constraints, transmission losses, valve-point effects, prohibited operating zones, ramp rates, and power limits, its performance was compared with other optimization techniques. The findings indicate that GTO provides the optimal scheduling of power generators, leading to significant reductions in daily operational costs and emissions with high percentages. Moreover, the integration of renewable energy significantly reduces pollutant gas emissions, fuel costs, and transmission losses, while meeting all imposed constraints. This research positively contributes to enhancing the reliability of power supply systems, while simultaneously reducing environmental pollution, transmission losses, and fuel costs
Artificial Intelligence-Based DS-PSO Algorithm for Enhanced Frequency Response in Digital IIR Filters
Full text linkDigital elliptic filters, as a type of infinite impulse response (IIR) digital filter, play a crucial role in signal processing applications. Despite their widespread use, there remains a significant research gap in optimizing their frequency response to better approximate desired magnitude responses. This study addresses this gap by introducing an innovative optimization technique that leverages the DS-PSO (Dynamic & Static-Particle Swarm Optimization) algorithm. Based on artificial intelligence, the DS-PSO method uniquely integrates topologies (dynamic and static) into particle swarm optimization (PSO), enabling more precise analysis of pole positions derived from a filter's transfer function coefficients. The primary research problem lies in approximating the frequency response of digital IIR elliptic filters to match a desired magnitude response. Traditional methods often fail to achieve this due to limitations in their optimization techniques. The proposed DS-PSO algorithm addresses this by setting a slightly more significant maximum pole radius (Rmax) than 1.0, surpassing the pre-established pole radius (R). This approach allows for a more accurate approximation of the frequency response. This feature distinguishes it from previous studies that employed genetic algorithms (GA) and semi-definite programming (SDP) techniques, which reported lower Rmax values. The results of this study demonstrate the effectiveness of the DS-PSO algorithm in improving the frequency response of digital IIR elliptic filters. The proposed method successfully approximates the desired magnitude response by designing 4th and 12th-order lowpass digital IIR elliptic filters while maintaining stability at a high average. This makes the technique particularly suitable for determining frequency response boundaries in electronics or communications systems. The contribution of this research extends beyond the immediate results. By introducing and validating the DS-PSO algorithm, this study provides a robust framework for future research in optimizing digital IIR filters. The findings not only enhance the design of digital elliptic filters but also open new avenues for improving other types of IIR filters and signal processing applications. This paper establishes a foundation for further research in signal processing and other fields, with significant theoretical and practical implications
Empirical Evaluation of Energy-assisted Protocols for Wireless Sensor Networks
Full text linkWireless sensor networks (WSNs) have emerged as a transformative technology with widespread applications in various fields, such as environmental monitoring, healthcare, and industrial automation. This investigation provides a comprehensive evaluation and comparison of an existing protocol, the Energy-Efficient Backbone-assisted protocol for Load Balancing (EBLBP), against two established protocols: Ad-hoc On-Demand Distance Vector (AODV) and Destination Sequenced Distance Vector (DSDV). Through extensive simulations, we analyzed the performance of these protocols across four critical metrics: scalability, efficiency, network lifetime, and energy consumption. Our findings reveal the inherent strengths and weaknesses of EBLBP, AODV, and DSDV, offering insights into their suitability for various WSN deployment scales and conditions. In the simulation environment, EBLBP achieved an impressive 66.67% reduction in overall energy consumption of 100 to 600 node positions, which underlined its positive impact on energy efficiency. The NS2 simulator was used for this investigation. The measured results validate the advantages of EBLBP in terms of energy optimization
Improved Lung Sound Classification Model Using Combined Residual Attention Network and Vision Transformer for Limited Dataset
Full text linkAccording to WHO data, the prevalence of respiratory disorders is increasing, exacerbated by a shortage of skilled medical professionals. Consequently, there is an urgent need for an automated lung sound classification system. Current methods rely on deep learning, but limited lung sound data resulted in low model accuracy. The widely used ICBHI 2017 dataset has an imbalanced class distribution, with a normal class at 52.8%, wheezing at 27.0%, crackles at 12.8%, and combined wheeze and crackles at 7.3%. The imbalance of the dataset may affect the model's efficiency and performance in classifying lung sounds. Given these data limitations, we propose a hybrid model, combining residual attention network (RAN) and vision transformer (ViT), to construct an effective respiratory sound classification model with a small dataset. We employ feature fusion techniques between convolutional neural network (CNN) feature maps and image patches to enrich lung sound features. Additionally, our preprocessing involves bandpass filtering, resampling sounds to 16 kHz, and normalizing volume to 15 dB. Our model achieves impressive ICBHI scores with 97.28% specificity, 92.83% sensitivity, and an average score of 95.05%, marking a 10% improvement over state-of-the-art models in previous research
An Exhaustive Survey on Authentication Classes in the IoT Environments
Full text linkIn today's world, devices are interconnected across various fields, ranging from intelligent buildings and smart cities to Industry 4.0 and smart healthcare. IoT security is still the biggest obstacle to deployment despite the exponential growth of IoT usage in our world. The principal objective of IoT security is to warrant the accessibility of services offered by an IoT environment, protect privacy, and confidentiality, and ensure the safety of IoT users, infrastructures, data, and devices. Authentication has become a top priority for everyone because it is the first line of defense against security threats and can allow or prevent users from accessing resources according to their legitimacy. Consequently, studying and researching authentication issues within IoT is extremely important. Our paper provides a comparative study of current IoT security research; it analyzes recent authentication protocols from 2018 to 2024. This survey’s goal is to provide an IoT security research summary, the biggest susceptibilities, and attacks, the appropriate technologies, and the most used simulators
Partition-Based Technique to Enhance Missing Data Prediction
Full text linkManaging missing data is a critical aspect of preprocessing in data mining endeavors, significantly influencing output accuracy during both model development and utilization phases. This study introduces a novel approach to predicting missing values by partitioning data into disjoint subsets based on partitioning measures. The rationale behind this approach is the elimination of unrelated data through partitioning, thereby improving the accuracy of missing value prediction within each subset. Through a combination of expert panel insights and statistical tests (including the Chi-square test and Cramer's V coefficient), the database partitioning measure was determined using operational data from the Mashhad Fire and Safe Services Organization. Models were constructed for each partition, and missing data were segmented accordingly, with the corresponding models employed for prediction. The results revealed that in 44% of cases, models built on partitioned data outperformed those constructed on the entire dataset. The evaluation of this method underscores its capability to predict missing values with heightened accuracy. Notably, this approach is independent of the method employed for missing value prediction, enabling seamless integration into existing methods as an additional step to bolster prediction accuracy.
Trust-based Enhanced ACO Algorithm for Secure Routing in IoT
Full text linkThe Internet of Things (IoT) is an expanding paradigm of object connectivity using a range of resource types and architectures to deliver ubiquitous and requested services. There are security issues associated with the proliferation of IoT-connected devices, allowing IoT applications to evolve. In order to provide an energy-efficient and secure routing method for sensors deployed within a dynamic IoT network, this paper presents a trust-aware enhanced ant colony optimization (ACO)-based routing algorithm, incorporating a lightweight trust evaluation model. As it is challenging to implement security in resource-constrained IoT networks, the presented model adopted bioinspired approaches, offering an improved version of ACO towards secure data transmission cost-effectively while taking into consideration residual energy and the trust score of the sensor to be optimized. The trust evaluation system has been enhanced in the development of the proposed routing algorithm and the node trust value is evaluated, sensor node misbehavior is identified, and energy conservation is maximized. The performance evaluation is demonstrated utilizing MATLAB. In comparison to the standard bioinspired algorithms and existing secure routing protocols, the proposed system reduces average energy consumption by nearly 50% regardless of the increase in the number of nodes and end-to-end delay of 40%, while finding the secure and optimal path in unison is designed to ensure trust in the IoT environment
Semantic Similarity Measure Using a Combination of Word2Vec and WordNet Models
Full text linkThe cognitive effort required for humans to perceive similarities and relationships between words is considerable. Measuring similarity and relatedness between text components such as words, texts, or documents is challenging, and it continues to be an active area of research across various domains. The complexity of language and the diverse factors that influence similarity and relatedness make this task an ongoing research focus. Researchers are exploring diverse approaches, to improve the accuracy and effectiveness of measuring similarity and relatedness in text. The utilization of knowledge sources, such as WordNet, has been a popular approach for modeling semantic relationships between words. However, Recently, distributional semantic models, such as Word2Vec, have demonstrated their ability to effectively capture semantic information and outperform lexiconbased methods in terms of unidirectional contextual similarity outcomes. In contrast to lexicon-based approaches, which rely on structure, distributional models leverage context to capture semantics. This study proposes a novel approach that linearly combines the lexical databases WordNet and Word2Vec to measure semantic similarity, focusing on improving upon previous techniques. The proposed approach is thoroughly detailed and evaluated using popular datasets to determine its effectiveness. The experimental results indicate that the proposed approach achieves highly satisfactory results and surpasses the performance of individual methods
ML-ACID: a Modified Machine Learning Algorithm Coupled With a Novel Ant Colony Approach for Intrusion Detection in IOT
Full text linkSoftware Defined Networks is becoming increasingly important in IoT because it allows devices to communicate more easily it provides the flexibility and centralized management, however in recent years these networks have witnessed a widespread spread of cyber-attacks that has a significant and negative impact on the availability of services. In this paper, we propose a novel approach for intrusion detection in Software Defined Networks for IoT. our work inspired by the self-defense mechanism of ant colonies. The approach uses a self-adaptable colony fingerprint and based on multiple parameters, it makes the detection of intrusions easy and filters out every other legitimate communication within the network. A machine learning model is used to provide basic predictions about the communication that later drives the evolution of the colony in terms of self-defence. The whole approach is implemented in a simple switch using Ryu-controller and analyses OpenFlow datagrams. The meta-heuristic implication of using ant colony optimization improved approach provides the system with reliability and high performance of detecting and blocking threats. in the end interesting results based on several scenarios shows the usability of our approach