Taiwan Association of Engineering and Technology Innovation: E-Journals
Not a member yet
887 research outputs found
Sort by
Tool Wear Prediction Combining Global Feature Attention and Long Short-Term Memory Network
Get PDFThis study aims to accurately predict tool flank wear in milling and simplify the complexity of feature selection. A hybrid approach is proposed to eclectically integrate the advantages between the long short-term memory (LSTM) network and the global feature attention (GFA) module. First, the feature matrix is calculated using the multi-domain feature extraction method. Subsequently, a parallel network is employed to achieve feature fusion. The stacked LSTM network extracts the temporal dependencies between features and the GFA module is used to adaptively complement key features representing global information of samples. Finally, the output features are concatenated, and tool wear prediction is achieved through a fully connected layer. Experiments demonstrate the optimal performance in predicting tool flank wear. Specifically, using the proposed GFA-LSTM framework reduces the mean absolute error (MAE) by 36.9%, 17.7%, and 25.2% in three experiments compared to the simple LSTM, validating the effectiveness of the proposed method
A Score-Based Evaluation Model for Rehabilitation of Existing Pumped Storage Hydropower Plant Construction
Get PDFAs the proportion of new and renewable energy increases, power control demands are becoming more frequent due to variability in power generation. As a complementary means against this, the pumped storage hydropower plants (PSHP) are attracting attention as energy storage systems (ESS), but it has high construction costs. Therefore, this study aims to improve the economic feasibility by developing the evaluation model of the existing infrastructure into an upper/lower dam suitable for PSHP. The concept of upper dam capacity is newly defined, and the evaluation index is constructed using normalization. A new evaluation system is presented for five factors: environment, stability, energy, capacity, and economy. Finally, it is tested in the pilot area in Korea. Several candidates, including the PSHP in operation, are found to have been distributed with higher scores. These results will help to satisfy the selection of candidates during the preliminary feasibility study phase, and programming them will enable more accurate and rapid assessment
Estimating Macronutrient Content of Paddy Soil Based on Near-Infrared Spectroscopy Technology Using Multiple Linear Regression
Get PDFThis study investigates the feasibility of employing near-infrared (NIR) spectroscopy with multiple linear regression (MLR) to estimate macronutrients in paddy soil compared with partial least squares (PLS) and principal component regression (PCR). Seventy-nine soil samples from West Java Province, Indonesia, are subject to conventional nutrient analysis and NIR spectroscopy (1000-2500 nm). The reflectance data undergoes various pretreatment techniques, and MLR models are calibrated using the forward method to achieve correlations exceeding 0.90. The best model calibrations are selected based on high correlation coefficients, determination coefficients, RPD, and low RMSE values. Meanwhile, the comparison of performance MLR is made with the PLS and PCR models. Results indicate that simple MLR models perform less than PLS for all nutrients, better than PCR for nitrogen, and below PCR for phosphorus and potassium. However, MLR reliably estimates soil nitrogen, phosphorus, and potassium content with ratio of performance to deviation (RPD) exceeding 2.0. This study demonstrates the potential of MLR for precise macronutrient estimation in paddy soil
Analysis of Strength Parameters of Polymer–Glass Composites Modified with Rubber Recyclate Through Bending Tests
Get PDFThe study aims to analyze the strength properties obtained from three-point bending tests of epoxy-glass composite samples modified by adding rubber recyclate. A pure epoxy-glass composite is used as a comparative variant. The tested materials, which varies in the percentage of rubber recyclate and distribution, are cut through waterjet cutting to minimize the influence of temperature. The results undergo statistical analysis, and the microstructures are examined using scanning electron microscopy. The decreasing bending strength of the composites is observed, as the content of rubber recyclate in the material increased. However, adding rubber recyclate directly into the resin subtly decreases in bending strength compared to adding in the layers between the glass mat layers. Composites with rubber recyclate exhibits lower deflection under load compared to pure composites. The most favorable bending test parameters are obtained for the material containing 5% rubber recyclate distributed in three layers
Optimizing Dynamic Stability in Power Systems: A Robust Approach with FOPID Controller Tuning Using HHO Algorithm
Get PDFThis study investigates the stability improvement in power systems by using fractional order proportional-integral-derivative (FOPID) controllers that have been improved with the Harris hawks optimization (HHO) algorithm. It showcases a novel integration of fractional order control and nature-inspired optimization approaches in single-machine infinite bus (SMIB) systems. Introducing FOPID controllers allows for precise control, which is essential for maintaining stability under varying conditions. This research utilizes HHO, a nature-inspired optimization technique, to optimize FOPID parameters. The research involves initializing the SMIB model, defining an objective function to minimize control errors, and applying HHO to fine-tune the FOPID controller iteratively. This proposed HHO-FOPID-SMIB method surpasses existing strategies, achieving a notable reduction in settling time to 6.29 seconds, thus demonstrating efficiency in stabilizing the SMIB system’s response faster than competing methodologies. Simulation results demonstrate improved stability, reduced overshoot, faster settling time, and transient response
Improving Healthcare Communication: AI-Driven Emotion Classification in Imbalanced Patient Text Data with Explainable Models
Get PDFSentiment analysis is crucial in healthcare to understand patients’ emotions, automatically identifying the feelings of patients suffering from serious illnesses (cancer, AIDS, or Ebola) with an artificial intelligence model that constitutes a major challenge to help health professionals. This study presents a comparative study on different machine learning (logistic regression, naive Bayes, and LightGBM) and deep learning models: long short-term memory (LSTM) and bidirectional encoder representations from transformers (BERT) for classify health feelings thanks to textual data related to patients with serious illnesses. Considering the class imbalance of the dataset, various resampling techniques are investigated. The approach is complemented by an explainable model, LIME, to understand the shortcomings of the classification results. The results highlight the superior performance of the BERT and LSTM models with an F1-score of 89%
A Fake Profile Detection Model Using Multistage Stacked Ensemble Classification
Get PDFFake profile identification on social media platforms is essential for preserving a reliable online community. Previous studies have primarily used conventional classifiers for fake account identification on social networking sites, neglecting feature selection and class balancing to enhance performance. This study introduces a novel multistage stacked ensemble classification model to enhance fake profile detection accuracy, especially in imbalanced datasets. The model comprises three phases: feature selection, base learning, and meta-learning for classification. The novelty of the work lies in utilizing chi-squared feature-class association-based feature selection, combining stacked ensemble and cost-sensitive learning. The research findings indicate that the proposed model significantly enhances fake profile detection efficiency. Employing cost-sensitive learning enhances accuracy on the Facebook, Instagram, and Twitter spam datasets with 95%, 98.20%, and 81% precision, outperforming conventional and advanced classifiers. It is demonstrated that the proposed model has the potential to enhance the security and reliability of online social networks, compared with existing models
Billet Size Optimization for Hot Forging of AISI 1045 Medium Carbon Steel Using Zener-Hollomon and Cingara-McQueen Model
Get PDFThis study investigates the effects of initial billet size variations on material flow behavior in hot forging processes, aiming to optimize the forging process using validated predictive models. Material and high-temperature compressive tests inform mathematical models, while simulations are conducted via the finite element method (FEM). Results align with the Zener-Hollomon and Cingara-McQueen approaches. The Arrhenius model predicts AISI 1045 steel flow stress with an R2 of 0.968 and an average absolute relative error (AARE) of 7.079%. The Cingara-McQueen equation achieves an R2 of 0.997 and an AARE of 2.960%. Reducing billets size from 260 mm to 230 mm decreases the material usage by up to 11.5%, while maintaining workpiece integrity. Experimental and simulated loads exhibit an AARE of about 2.69%, thereby indicating potential cost and efficiency improvements in hot forging processes
A Self-Repairing Natural Rubber as a Novel Material Pad to Develop an Electro-Surgical Training Prototype
Get PDFThis work aims to develop a self-repairing natural rubber sheet and use it in a new design electro-surgical training prototype. The self-repairing material is prepared via controlled crosslinking with varying curing time and temperature and applied as a material pad. The electrical circuit board in the prototype is created to measure the depth of the surgical blade through a material pad. The completely modified control crosslinking of the rubber sheet is confirmed by the changing chemical structure of rubber latex via FT-IR spectra resulting in the hardening of swelling affected by high crosslinking density. The self-repairing of natural rubber sheets occurred at the cut part and the tensile strength at break increases with the increase in self-repairing time. The prototype testing shows that when the scalpel blade is cut into the rubber sheet at the setting dept, the electrical circuit is activated, making it suitable for medical practice
Application of Genetic Algorithm and Analytical Method to Determine the Appropriate Locations and Capacities for Distributed Energy System
Get PDFIn this study, the genetic algorithm (GA) and an analytical technique are used to properly connect the distributed energy system (DES) to the distribution network of the Federal Capital Territory (FCT). A power flow solution is used to obtain the losses and voltages assigned to the chromosomes as the fitness value for the GA to determine the best locations for the DES. Subsequently, the analytical method is used to calculate the capacities of the DES, corresponding to each location obtained using the GA. The effectiveness of the technique is examined on IEEE 33 and 69 buses, and the results demonstrate a loss reduction of 69.19%, the least voltage of 0.975 pu for the 33-node, and a 70.22% loss reduction with the least voltage of 0.985 pu for the 69-node. The suggested technique is applied to the FCT distribution network, and the results show a 70% voltage improvement and 14.05% loss reduction