Engineering Science and Technology-An International Journal-Jestech最新文献

{"title":"Advanced deep reinforcement learning for optimizing 3D printing toolpaths: A framework with enhanced agent architectures, Count-Prioritized Replay, and curriculum learning","authors":"Ahmed Merze ,&nbsp;Fatih Vehbi Çelebi","doi":"10.1016/j.jestch.2025.102205","DOIUrl":"10.1016/j.jestch.2025.102205","url":null,"abstract":"<div><div>Optimizing toolpaths in 3D printing presents a significant challenge for achieving efficient and high-quality prints. In this study, a novel deep reinforcement learning (DRL) framework is proposed to overcome this problem. A core component of this framework is the proposed agent: Count-Prioritized Replay Deep Q-Network (CPR-DQN). The first key contribution is a developed simulation environment called <span>PrintBoardEnv</span>, which has curriculum learning built in. Secondly, a new method is developed for experience replay called Count-Prioritized Replay (CPR). The third key development is the CPR-DQN agent itself that uses a special architecture which includes features like Implicit Quantile Networks (IQN), Munchausen RL, Dueling, and Noisy Networks. Our agent is trained in two stages respectively, which is an offline pre-training, and then an online training. Our CPR-DQN agent is compared with other agents like DQN, Rainbow DQN, and Beyond The Rainbow (BTR). Furthermore, it is demonstrated that the CPR-DQN agent achieves great performance, highlighting the benefits of the proposed framework for toolpath optimization.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"72 ","pages":"Article 102205"},"PeriodicalIF":5.4,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145340870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolution mechanisms of residual stress in roll bending process for large-scale aerospace integral structures","authors":"Yuhang Duan ,&nbsp;Jianfeng Wang ,&nbsp;Shulin Lu ,&nbsp;Lihong Cheng ,&nbsp;Shuwen Huang ,&nbsp;Kun Zhou ,&nbsp;Yaobang Zhao ,&nbsp;Junchen Li ,&nbsp;Xiaohong Zhan","doi":"10.1016/j.jestch.2025.102214","DOIUrl":"10.1016/j.jestch.2025.102214","url":null,"abstract":"<div><div>Controlling residual stress distribution during roll bending is essential for manufacturing high-precision components, such as rocket propellant tanks. However, the mechanisms governing stress evolution in this process remain insufficiently understood, which limits the development of effective control strategies. This study introduces a novel full-scale finite element model of a 4.7 m rocket tank panel, rigorously validated against experimental stress measurements, to reveal the governing mechanisms of residual stress development. Unlike prior studies limited to small-scale or simplified geometries, this work systematically clarifies the coupled influence of roller down displacement and roller spacing on stress evolution. Two novel optimization pathways are identified: (i) intermediate roller displacement (15–20 mm) reduces peak residual stress by 15–20 % through suppressing localized deformation; and (ii) optimal roller spacing (500–550 mm) enhances stress uniformity, reducing equivalent plastic strain by 30–40 %. Moreover, combined parameter optimization (17.5–20 mm displacement with 500–550 mm spacing) lowers critical zone stresses below 30 MPa, significantly improving dimensional stability. These findings not only establish the first validated full-scale stress evolution model for roll bending but also deliver practical, quantitative guidelines for stress control in large aerospace structures.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"72 ","pages":"Article 102214"},"PeriodicalIF":5.4,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145365267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated CFD and Theoretical Modeling of Erosive Wear in Composite Laminates","authors":"Eduardo Hernández Huerta ,&nbsp;Armando Irvin Martínez Pérez ,&nbsp;Rafael Campos Amezcua ,&nbsp;Edgar Ernesto Vera Cárdenas ,&nbsp;Marisa Moreno Ríos","doi":"10.1016/j.jestch.2025.102182","DOIUrl":"10.1016/j.jestch.2025.102182","url":null,"abstract":"<div><div>The use of mathematical models and computational flow dynamics (CFD) in erosive wear studies can provide an approximation of the material performance under different conditions of the temperature, impact angle, and particle velocity. For this reason, the use of these analysis tools is essential for applications in specialized areas such as renewable energy, automotive and aerospace. This research work reports the results of erosive wear using use results of erosive wear using a validated numerical CFD tool based on the behavior and characteristics of the erosive SiC particles on the surface of a laminated composite material. In addition, the use of the Patnaik and Oka models is presented to obtain an approximation of the erosion rate, using the properties of the matrix and the reinforcing material that constitute the laminated composite evaluated, as well as the erosion conditions. The numerical results obtained compared to the experimental results had an approximation in of 94% in the erosion rate, 93.5% in the mass loss and 96.1% in the wear zone depth. These findings not only validate the combined use of CFD simulation and theoretical models, as an accurate alternative to traditional experimental testing, also accelerate the materials evaluation process, reduce development costs and improve the design of components with greater resistance to erosive wear. Therefore, these results provide a solid basis for integrating erosive wear resistance criteria into the structural design of composite components, optimizing their service life and performance under severe operating conditions.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102182"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Front Matter 1 - Full Title Page (regular issues)/Special Issue Title page (special issues)","authors":"","doi":"10.1016/j.jestch.2025.102221","DOIUrl":"10.1016/j.jestch.2025.102221","url":null,"abstract":"","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102221"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Soiling identification and forecasting in urban environment","authors":"Ivana Radonjić ,&nbsp;M. Asim Amin ,&nbsp;Milutin Petronijević ,&nbsp;Plamen Tsankov ,&nbsp;Martin Ćalasan","doi":"10.1016/j.jestch.2025.102188","DOIUrl":"10.1016/j.jestch.2025.102188","url":null,"abstract":"<div><div>Urban Photovoltaics (PVs) plays an important role in harnessing power generation for sustainable development, but its performance is impaired under an increasing level of environmental pollutants. This paper presents the development and verification of a Stacked Ensemble Learning (SEL) and Decision Trees (DTs) based algorithm for forecasting PV output under conditions of significant air pollution. It uses a small dataset for training and validation, including data on the production of an urban PV system and meteorological data commonly available from local weather services. The developed algorithm was tested on data collected from an urban location in Niš, Serbia, which is exposed to significant air pollution during the regular heating season, for three different panel mounting configurations. The SEL model achieves a near-zero error metric in predicting PV yield for vertical monocrystalline panels, maintaining high accuracy even under soiling, which is a key feature for real-world applications. Similar conclusions can be drawn for the cases of horizontal and optimally positioned panels. Using the developed SEL algorithm, it is shown that the soiling ratio obtained from the estimation of PV production of clean and soiled panels can be used as a reliable indicator for scheduling the optimal period for cleaning PV panels.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102188"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design adaptation of an electronically tunable oscillator using a low performance linearized CMOS operational transconductance amplifier","authors":"Roman Sotner ,&nbsp;Ladislav Polak ,&nbsp;Lukas Langhammer ,&nbsp;Darius Andriukaitis","doi":"10.1016/j.jestch.2025.102178","DOIUrl":"10.1016/j.jestch.2025.102178","url":null,"abstract":"<div><div>This paper presents the implementation of commercially available CMOS devices with unfavorable properties, such as low output resistance, in an application designed to mitigate these limitations. By<!--> <!-->employing a specific topology and considering key design parameters, the proposed approach minimizes the adverse effects of low output resistance. This design focuses on a linearized operational transconductance amplifier (OTA) based on CMOS transistors, featuring with very low output resistance. This OTA is further integrated into an LC oscillator, where the associated disadvantages are suppressed through a specialized topology and careful selection of parameter values that are unaffected by the low OTA output resistance. The operational verification targets a frequency range of several hundred kHz and a linearly processed voltage range of several hundred mV. The linearized OTA-based low-gain amplifier/attenuator offers a linearity error within −7% (±500 mV). The proposed OTA implementation in the oscillator introduces highly simplified method for adjusting the oscillation condition using a single grounded element while minimizing the adverse effects of low output resistance of OTA. Additionally, the tunability of the oscillator using varactor diodes achieving a range from 120 <!--> <!-->kHz to 273 kHz for a voltage varying from 0 V to 5 V.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102178"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lateral and longitudinal human–machine co-driving of intelligent vehicle based on driving authorities game","authors":"Bin Tang,&nbsp;Jufang Yao,&nbsp;Haobin Jiang,&nbsp;Wei Mi","doi":"10.1016/j.jestch.2025.102187","DOIUrl":"10.1016/j.jestch.2025.102187","url":null,"abstract":"<div><div>Due to the current limitations of autonomous driving technology, human–machine co-driving is viewed as a viable and practical intermediary solution to bridge the gap between assisted driving and fully automated driving. To address challenges related to the allocation of lateral and longitudinal driving authority, as well as the interaction between the driver and the autonomous system, this paper proposes a game-sharing control strategy that operates at both the decision-making and control levels. At the decision level, a bargaining game-based approach is employed to allocate lateral and longitudinal driving authority dynamically, adjusting the distribution in accordance with a benefit function. At the control level, a lateral controller based on an extended game is developed to compute the optimal control manoeuvres by combining the driver’s inputs and reference tracking commands. This controller smoothly combines control inputs from both the driver and the automated system to ensure vehicle stability and minimize human–machine conflict. Simulation results show that, under a double lane-change scenario, the proposed strategy reduces yaw rate, lateral velocity, and lateral deviation by 0.35 rad/s, 0.6 m/s, and 0.6 m, respectively, compared with the fuzzy authority allocation method. Additionally, under cornering conditions, lateral deviation is reduced by 0.7 m. Finally, the proposed human–machine co-driving control strategy is verified by the experiment. The results indicate that the proposed strategy not only enhances trajectory tracking accuracy but also improves vehicle stability. Furthermore, by comparing the lateral and longitudinal authority allocation between the proposed and fuzzy strategies, it is evident that the approach significantly alleviates human–machine conflict, especially by reducing the frequent transfer of authority.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102187"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Secure electric vehicle charging station demand forecasting under adversarial and false data injection attacks","authors":"Najmul Alam,&nbsp;M.A. Rahman,&nbsp;Md. Arafat Hossain,&nbsp;Md. Rashidul Islam","doi":"10.1016/j.jestch.2025.102203","DOIUrl":"10.1016/j.jestch.2025.102203","url":null,"abstract":"<div><div>Forecasting the charging demand of electric vehicle charging stations (EVCSs) is critical for urban planning, resource allocation, policy development, and efficient grid management. However, reliance on sensor-collected data transmitted through various communication channels poses significant cybersecurity risks to forecasting models. This study evaluates the vulnerability of several commonly used forecasting models named random forest (RF), convolutional neural network (CNN), convolutional long short-term memory (ConvLSTM), and bidirectional convolutional long short-term memory (BiConvLSTM), under simulated cyber-attacks. Different attack scenarios, including fast gradient sign method (FGSM) and basic iterative method (BIM)-based adversarial attacks and scaling-based false data injection (FDI) attacks, are considered with varying attack volumes and perturbations. Metrics, such as root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE), are employed to assess and compare the accuracies of different models. Results indicate that forecasting models exhibit significant performance degradation under these cyber-attacks. As a countermeasure, adversarial training has been employed to mitigate the impact of such attacks and has proven to be highly effective. The experimental observations in this research elucidate the impact of cyber-attacks along with a defense mechanism to mitigate economic and technical risks to the EVCS, fostering the future development of accurate and cyber-resilient forecasting methodologies essential for advancing both academic and industrial domains of EVCS demand forecasting.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102203"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"HybridNet: Advancing MRI image quality using dense attention and deep learning","authors":"Seshagiri Rao Sugguna,&nbsp;Sumesh E.P","doi":"10.1016/j.jestch.2025.102186","DOIUrl":"10.1016/j.jestch.2025.102186","url":null,"abstract":"<div><div>Magnetic Resonance Imaging (MRI) is widely used for brain disease diagnosis due to its superior tissue contrast, but low-field MRI scanners often generate low-resolution images that hinder accurate interpretation. Extending scan time or upgrading to high-field systems increases cost and patient discomfort, making them impractical solutions. Deep learning-based super-resolution has emerged as a promising alternative; however, conventional CNN and GAN-based models frequently oversmooth details, introduce artifacts, or rely on synthetic downsampling, thereby limiting their clinical reliability. To address these challenges, we propose HybridNet, a novel CNN-based framework that integrates even–odd pixel decomposition with Dense Attention Blocks (DABs) to capture both global structural consistency and fine anatomical details. The novelty of HybridNet lies in its dual-focus strategy pixel decomposition to preserve structural integrity and reduce aliasing, and dense attention integration to highlight diagnostically significant regions, resulting in sharper and more reliable MRI reconstructions. The objectives of this study includes — design a clinically reliable MRI super-resolution framework that minimizes artifacts and oversmoothing, to ensure generalization across different scanners and acquisition settings, and to provide perceptually faithful reconstructions that aid radiological interpretation without increasing scan time or hardware cost. Proposed HybridNet model is trained on real paired MRI datasets from different scanners, The result demonstrates strong robustness and diagnostic relevance in both Quantitative and qualitative aspects. Quantitative evaluations reveal an improvement of 6–15 dB in PSNR and 2%–4% in SSIM compared to state-of-the-art methods such as SRCNN, FSRCNN, VDSR, and EDSR. Ablation studies further confirm the role of pixel decomposition, attention integration, and feature fusion in enhancing performance. Radiologists’ subjective assessments also validate superior perceptual quality, achieving a Mean Opinion Score of 87.3 and a Perceptual Evaluation score of 73.35. Overall, HybridNet offers a cost-effective and clinically practical solution for enhancing MRI image quality.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102186"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145320841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-sensitivity microwave sensor for buffalo milk fat rate and adulteration analysis with branch line coupler","authors":"Mustafa Yıldırım,&nbsp;Mahmut Ahmet Gözel","doi":"10.1016/j.jestch.2025.102172","DOIUrl":"10.1016/j.jestch.2025.102172","url":null,"abstract":"<div><div>In this study, an innovative microwave sensor was designed based on a Complementary Elliptical Electric-LC (CEE-LC) resonator structure compact with a Branch Line Coupler (BLC) that can determine milk fat content and perform adulteration analysis. Firstly, the fat content measurements of the milk used in the study were performed using a Lactoscan brand milk analyzer to determine the proposed sensor performance: buffalo milk (BM) was determined to have 12.19 % fat content, goat milk (GM) 5.02 %, Jersey milk (JM) 6.81 %, and Simmental milk (SM) 4.10 %. Then, adulterated milk samples were created using JM-BM and water-BM mixtures to perform adulteration analysis using the proposed sensor. In the final stage of sample preparation, a common dielectric probe was used to measure the permittivity and electrical conductivity of all milk samples in the 10 MHz to 20 GHz frequency range. The resonance frequency of the S₂₁, which provides transmission-zero (TZ) in the S-parameter of the proposed sensor, was determined to be 7.03 GHz. The CEE-LC resonator structure in the sensor was positioned on the ground surface projection of the transmission line between “port 1” and “port 2”. In the BM-water adulteration analysis, the resonance frequencies of the sensor at parameter S₂₁ were measured as 2.40 GHz for only BM sample, 2.335 GHz for BM with 20 % water mixture, 2.295 GHz for BM with 40 % water, 2.260 GHz for BM with 60 % water, 2.215 GHz for BM with 80 % water, and 2.105 GHz for only water. Similarly, the proposed sensor detected the resonance frequencies of the S₂₁ parameter in the adulteration analysis performed when JM mixed with the BM as 2.395 GHz for 20 % addition, 2.385 GHz for 40 % addition, 2.370 GHz for 60 % addition, 2.325 GHz for 80 % addition, and 2.32 GHz for pure JM. To increase the usability of the sensor in field applications, a portable measurement system was developed by integrating a voltage-controlled RF oscillator and a power detector. In DC voltage-based measurements, the output voltage increased from 1.23 V to 1.56 V as the milk fat content increased, indicating a correlation between the DC output voltage and fat content. In the adulteration analysis, it decreased from 1.55 V to 1.18 V as the water content in buffalo milk increased. During DC analyses, the 2.4 GHz resonant frequency of the BM, which has the TZ value for parameter S₂₁ in the proposed sensor, was used to determine fat content and adulteration. Literature comparisons revealed that the proposed system exhibits superior performance compared to existing methods, especially in detecting water adulteration in buffalo milk.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"71 ","pages":"Article 102172"},"PeriodicalIF":5.4,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}