Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Automated tool for conducting emotion analysis studies in perception surveys
Full text linkConsidering the growing need for companies to automate the analysis of customer opinions from different digital media, this paper outlines the development of an automated tool for emotion analysis in survey responses utilizing Ekman’s six-emotion model (joy, excitement, anger, sadness, fear, and boredom). The tool processes spreadsheets containing qualitative responses and generates the percentage distribution of emotions at both individual and aggregated levels. A case study conducted with 46 systems engineering students at the University of Cartagena during the COVID-19 pandemic showed that 'anger' was the most prevalent emotion (29.3%), followed by 'excitement' (19.4%), while 'boredom' was the least frequent (2.6%). The tool demonstrated an accuracy rate of 92% in classifying emotions, compared to 90% achieved through manual coding. These results highlight the tool’s effectiveness in automating emotion analysis, providing statistical and graphical reports that aid decision-making in academic and organizational contexts
Effect of gamma radiation on semi-crystalline polyvinyl chloride polymer for low-voltage cable insulator
Full text linkThis study explores the properties of semi-crystalline polyvinyl chloride (PVC) polymer as insulation material for low-voltage (LV) cables under high gamma radiation exposure. Test samples underwent gamma radiation (60Co) at doses of 25, 50, 100, 200, 400, and 800 kGy. The evaluation encompassed surface morphology, electrical conductivity, thermal characteristics, and mechanical properties via tensile tests. Electron microscopy observation indicated surface smoothing and flattening occurred at an irradiation dose of 800 kGy. Gamma radiation with increasing doses results in similar thermogram profiles with slight differences in melting temperature and residue mass. The sample irradiated at a gamma dose of 25 kGy generates an increase in the percentage of crystallinity, indicating the occurrence of crosslinking, while other doses exhibit a decrease of crystallinity with increasing radiation dose. Tensile stress significantly dropped up to 400 kGy but increased at 800 kGy. Elongation at break (EAB) decreased with higher gamma radiation doses. Overall, materials up to 800 kGy remained non-brittle, serving as effective insulators and demonstrating thermal stability within high gamma radiation exposure conditions
Three-level common emitter-current source inverter equipped with MPPT system for photovoltaic power converter
Full text linkCommon emitter-current source inverter (CE-CSI) has unique features with its common emitter structure of its power switches. Hence, instead of its simpler power supply for gate drive circuits, it also allows higher switching operation because of zero gradient voltage of its power switches. One of interesting applications of current source power inverter is for photovoltaic (PV) power converter. This paper discussed the three-level CE-CSI equipped with current based incremental resistance maximum power point tracking (MPPT) system as a new alternative for PV system power converter. Test results revealed some characteristics of PV power conversion using this inverter. Moreover, in order to investigate the system performance approaching real condition, testing during partial shading condition of PV modules were also conducted. Test results verified the efficacy of the incremental resistance based MPPT algorithm implemented in the CE-CSI circuits for increasing the performance of PV power generation
Optimization of dynamic transmission network expansion planning using binary particle swarm optimization algorithm
Full text linkIncreasing power demand is usually met by the expansion of generation capacity. The transmission network should be expanded in tandem to ensure power is evacuated from generation points to the load centres. Inadequate power capacity causes congestion. Congestion results due to under-voltages and violation of transmission lines’ loading limits. Constructing additional transmission lines is required to alleviate the congestion after measures of increasing the transmission line’s transfer capability are exploited. Transmission network expansion planning (TNEP) determines the transmission lines to be added to a power system at minimal construction cost, without violating network constraints. In this research, voltage limit violations are penalized in a constrained dynamic TNEP problem for a 10-year planning horizon. The optimal location and number of new transmission lines required at minimal construction cost, and transmission losses associated with the transmission network operations are determined. Improved binary particle swarm optimization (IBPSO) algorithm is applied to optimize the dynamic transmission network expansion planning (DTNEP) results. The developed model is tested on Garver’s 6-bus system using MATLAB. The construction cost for new transmission lines is minimized, and transmission losses reduced when compared to other published works without violating voltage limits (±5%) and transmission lines’ thermal capacities. The transmission network system adequacy is improved
Solving problems of the flexible scheduling machines
Full text linkFlexible job scheduling problem (JSP) as an optimization problem, tends to find solution for allowing different operations to be processed faster. This problem could be solved by genetic algorithm, as we have proven in another experiment. Now, we have tried to outperform state of the art, by using parallel genetic algorithm. Parallel genetic algorithm has two types and we have chosen the most popular one coarsed grained genetic algorithm, for our specific case. The results have improved time wise and are promising in some of the datasets, while a need exists for improving on other ones. In the future, we will compare both versions of parallel genetic algorithms but also compare the results to another algorithm
Optimizing turbine location in upgraded wind farm using grasshopper optimization algorithm
Full text linkThis research explores the use of the grasshopper optimization algorithm (GOA) for optimizing the placement of additional turbines in an established wind farm. The primary objective is to increase the annual energy production (AEP) of the wind farm while minimizing the wake effects caused by both existing and new turbines. The research evaluates three different turbine types (1.5 MW, 2.0 MW, and 2.5 MW) to identify the most appropriate choice for increasing the wind farm's capacity. The GOA’s performance is compared with the commercial software windPRO and validated using WAsP software for energy calculations. Numerical results indicate that the GOA effectively improves wind farm layout, with the 1.5 MW turbines identified as the optimal choice for maximizing AEP and minimizing wake interactions. This study provides practical insights for wind farm operators and contributes to the development of advanced optimization techniques in wind energy
Lung diseases identification using hybrid transfer learning and bidirectional long short-term memory
Full text linkLung diseases rank as the third most prevalent cause of mortality globally. Accurate identification of lung disease is essential to provide appropriate medical intervention for patients. This research devised a categorization system for lung diseases using chest X-Rays (CXR). The system can identify bacterial pneumonia, viral pneumonia, COVID-19, tuberculosis, and normal CXR. The approach for detecting lung diseases utilize a combination of hybrid transfer learning and bidirectional long short-term memory. The research included convolutional neural network (CNN) models including Resnet50-BiLSTM, VGG19-BiLSTM, InceptionV3-BiLSTM, Resnet50, VGG19, and InceptionV3. The Resnet50-BiLSTM model outperforms other models in terms of accuracy and overall performance. The Resnet50-BiLSTM model achieved an accuracy of 99.87%. The models that achieve the second greatest accuracy are Resnet50, VGG19-BiLSTM, VGG19, InceptionV3-BiLSTM, and InceptionV3. The research utilizes precision, recall, and F1-Measure to demonstrate that Resnet50-BiLSTM outperforms other methods by achieving the greatest value. This research improves the performance outcomes when compared to earlier studies
Eagle strategy-based crow search algorithm for UCP: integration of pumped storage units in smart grid environment
Full text linkThis paper proposes a hybrid eagle strategy with crow search algorithm (ES-CSA) as local optimizer to solve the unit commitment problem (UCP) in power systems. The algorithm aims to minimize total operational costs while considering pumped storage units as spinning reserves. The proposed methodology combines the exploration capability of ES with CSA's local search efficiency to determine optimal generator scheduling and power dispatch. The approach is validated using standard test cases from the literature, demonstrating improved convergence and cost reduction compared to existing methods. Results confirm the effectiveness of integrating pumped storage units in reducing overall system costs while maintaining reliable operation
Improving nutrient prediction models with polynomial and ratio features and mRMR selection
Full text linkDue to limited space and regulations, food labels often lack information on micronutrients, i.e., vitamins and minerals. Accurately predicting missing these micronutrient data is essential yet challenging. This study explores the feasibility of using machine learning to predict these missing nutrients based on a limited reported nutrient (protein and carbs). Using the Tabel Komposisi Pangan Indonesia (TKPI) dataset, we evaluated the performance of 12 diverse classifiers to predict binary classes ("low" or "high") for 13 target micronutrients. Random forest emerged as the best performing classifier with an average accuracy of 0.7421 across all target nutrients. Additionally, we introduced feature engineering techniques by incorporating polynomial and ratio features to enhance model performance. Minimum redundancy maximum relevance (mRMR) feature selection was then applied to identify the most informative features. This approach boosted the average accuracy of the random forest classifier to 0.7591. These findings highlight the efficacy of feature engineering and selection in enhancing nutrient prediction models, demonstrating the potential to improve consumer knowledge about unknown nutrients in food
Theoretical and experimental implementation of PID and sliding mode control on an inverted pendulum system
Full text linkThis paper explores the theoretical and experimental implementation of proportional-integral-derivative (PID) and sliding mode control (SMC) on an inverted pendulum system, a well-known problem in control engineering that is inherently unstable and highly nonlinear. The primary objective of this study is to evaluate and compare the effectiveness of these two control strategies in achieving system stabilization and robustness against disturbances. The PID controller, widely utilized due to its straightforward design and implementation, is developed based on the linearized model of the inverted pendulum. On the other hand, the SMC technique, known for its robustness to system uncertainties and external disturbances, is employed to tackle the nonlinear nature of the system. The controllers are tested in both simulation and real-time experimental environments to ensure the reliability of the findings. The results from the experiments indicate that while the PID controller performs adequately under nominal conditions, it struggles to maintain stability when faced with parameter variations and external perturbations. In contrast, the SMC exhibits superior performance by consistently stabilizing the pendulum even under adverse conditions, demonstrating its robustness and effectiveness in managing nonlinear systems. This comparative analysis provides valuable insights into the practical applications of PID and SMC, highlighting the trade-offs between simplicity and robustness in control system design