Pub Date : 2025-04-26DOI: 10.1016/j.jestch.2025.102069
Z. Esmaeili , M. Sheikholeslami
Efficient thermal management is crucial for maintaining the safety and lifespan of battery packs, particularly during high-discharge operations. This study investigates an innovative approach to cooling these batteries by modifying traditional rectangular mini-channels with bio-inspired fins, modeled after dolphin dorsal shapes and fish contours. Unlike conventional cooling designs, these structures enhance fluid mixing and heat transfer efficiency when paired with a hybrid nanofluid (Fe3O4–SWCNT in water) under laminar flow conditions. A time-dependent numerical simulation was conducted to replicate unsteady heat generation in battery cells, and the results were validated against established studies. The findings indicated that dolphin dorsal fins and fish-shaped fins reduce total entropy generation by 28.59 % and 14.12 %, respectively, compared to a standard mini-channel. Additionally, the convective heat transfer coefficient improved by 20.18 % with dolphin fins and 43.04 % with fish fins, demonstrating superior thermal regulation. The hydrothermal performance, evaluated using the performance evaluation criterion (PEC), showed that the fish-shaped fins outperformed dolphin fins by 42.87 %, achieving a PEC value of 1.12. These results highlight the effectiveness of bio-inspired fin geometries in optimizing battery cooling systems, offering a promising strategy for improving the efficiency and longevity of batteries in electric vehicles.
{"title":"Enhanced thermal management of lithium-ion batteries using hybrid nanofluids in finned mini-channels: Energy and entropy analyses","authors":"Z. Esmaeili , M. Sheikholeslami","doi":"10.1016/j.jestch.2025.102069","DOIUrl":"10.1016/j.jestch.2025.102069","url":null,"abstract":"<div><div>Efficient thermal management is crucial for maintaining the safety and lifespan of battery packs, particularly during high-discharge operations. This study investigates an innovative approach to cooling these batteries by modifying traditional rectangular mini-channels with bio-inspired fins, modeled after dolphin dorsal shapes and fish contours. Unlike conventional cooling designs, these structures enhance fluid mixing and heat transfer efficiency when paired with a hybrid nanofluid (Fe<sub>3</sub>O<sub>4</sub>–SWCNT in water) under laminar flow conditions. A time-dependent numerical simulation was conducted to replicate unsteady heat generation in battery cells, and the results were validated against established studies. The findings indicated that dolphin dorsal fins and fish-shaped fins reduce total entropy generation by 28.59 % and 14.12 %, respectively, compared to a standard mini-channel. Additionally, the convective heat transfer coefficient improved by 20.18 % with dolphin fins and 43.04 % with fish fins, demonstrating superior thermal regulation. The hydrothermal performance, evaluated using the performance evaluation criterion (PEC), showed that the fish-shaped fins outperformed dolphin fins by 42.87 %, achieving a PEC value of 1.12. These results highlight the effectiveness of bio-inspired fin geometries in optimizing battery cooling systems, offering a promising strategy for improving the efficiency and longevity of batteries in electric vehicles.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102069"},"PeriodicalIF":5.1,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143874476","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}
Pub Date : 2025-04-25DOI: 10.1016/j.jestch.2025.102071
Talal Taha , Sohaib Abdal , Liaqat Ali , Rana Muhammad Zulqarnain , Se-Jin Yook
The paper deals with the analysis of the laminar incompressible flow of a Carreau-Casson-Williamson fluid having magnetohydrodynamics effects with thermophoresis and Brownian motion effects over a wedge surface. Using similarity variables, a set of coupled ordinary differential equations are formulated for the governing equations of fluid flow. The solution process comprises a two-stage calculation. ODEs are first solved numerically using MATLAB’s bvp4c function, a known solver of boundary value problems known to solve complex ODEs within fluid dynamics very effectively. Further optimization and simplification of the analysis are achieved by using an artificial neural networking base Levenberg Marquardt algorithm (ANN-LMA) model. The derived dataset was divided into three parts: training 70%, validation 15%, and testing 15%. MSE metric between the values of and . MSE values can be used to grade the model’s performance. Increased Weissenberg number increases velocity and elasticity by facilitating flow with the boundary layer. On the other hand, raised Casson, Williamson, and magnetic parameters bring down the velocities as there comes the effect of damping and more resistance. Thermophoresis affects the migration rates of the particles by controlling the thermal and concentration gradient in the boundary layer with Brownian motion influencing its diffusion on the viscosity level and stability of fluid while in motion. In general, there is a fundamental interest in wedge flow because this geometry is present in aerodynamics and hydrodynamics, the study of fluids flowing around shapes like airfoils, nozzles, and underwater vehicles.
{"title":"Artificial intelligence approach to magnetohydrodynamic flow of non-Newtonian fluids over a wedge: Thermophoresis and Brownian motion effects","authors":"Talal Taha , Sohaib Abdal , Liaqat Ali , Rana Muhammad Zulqarnain , Se-Jin Yook","doi":"10.1016/j.jestch.2025.102071","DOIUrl":"10.1016/j.jestch.2025.102071","url":null,"abstract":"<div><div>The paper deals with the analysis of the<!--> <!-->laminar incompressible flow of a Carreau-Casson-Williamson fluid having magnetohydrodynamics effects with thermophoresis and Brownian motion effects over a wedge surface. Using similarity variables, a set of coupled ordinary differential equations are formulated for the governing equations of fluid flow. The solution process comprises a two-stage calculation. ODEs are first solved numerically using MATLAB’s bvp4c function, a known solver of boundary value problems known to solve complex ODEs within fluid dynamics very effectively. Further optimization and simplification of the analysis are achieved by using an artificial neural networking base Levenberg Marquardt algorithm (ANN-LMA) model. The derived dataset was divided into three parts: training 70%, validation 15%, and testing 15%. MSE metric between the values of <span><math><mrow><msup><mrow><mn>10</mn></mrow><mrow><mo>-</mo><mn>8</mn></mrow></msup></mrow></math></span> and <span><math><mrow><msup><mrow><mn>10</mn></mrow><mrow><mo>-</mo><mn>10</mn></mrow></msup></mrow></math></span>. MSE values can be used to grade the model’s performance. Increased Weissenberg number increases velocity and elasticity by facilitating flow with the boundary layer. On the other hand, raised Casson, Williamson, and magnetic parameters bring down the velocities as there comes the effect of damping and more resistance. Thermophoresis affects the migration rates of the particles by controlling the thermal and concentration gradient in the boundary layer with Brownian motion influencing its diffusion on the viscosity level and stability of fluid while in motion. In general, there is a fundamental interest in wedge flow because this geometry is present in aerodynamics and hydrodynamics, the study of fluids flowing around shapes like airfoils, nozzles, and underwater vehicles.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102071"},"PeriodicalIF":5.1,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143874475","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}
Pub Date : 2025-04-18DOI: 10.1016/j.jestch.2025.102057
Ayşe Beşkirli
Nowadays, with the demand for renewable energy the interest in wind energy is increasing day by day and the use of wind turbines is becoming widespread. In order to efficiently generate electricity from wind turbines, it is important that the turbines are correctly positioned on wind farms. In this study, a standard wind farm area of 2 km × 2 km was used for wind turbine layout. In addition, different from the literature, a 4 km × 4 km wind farm area was also created. Both wind farming areas were divided into 10 × 10 and 20 × 20 grids as in the literature. In addition, 25 × 25, 40 × 40 and 50 × 50 grids were also created. Thus, the wind farming area is divided into more grids and more flexible positioning of the turbines is aimed. There are three different case scenarios for the layout of the wind turbines. Case A has a constant wind speed of 12 m/s and unidirectional wind, while Case B has a constant wind speed of 12 m/s with a 36-directional 10° angle. Case C has variable wind speeds of 8, 12 and 17 m/s with a 36-directional 10° angle. The HHO algorithm was used to perform all these processes. However, since the wind turbine layout problem is binary, the HHO algorithm is adapted to binary with logic operators. These binary methods are called HHOAND and HHOXOR. The proposed methods achieved competitive results compared to other algorithms in the literature and performed well for all grid structures. Thus, it can be said that the proposed methods are effective for the wind turbine layout problem.
{"title":"An efficient binary Harris hawks optimization based on logical operators for wind turbine layout according to various wind scenarios","authors":"Ayşe Beşkirli","doi":"10.1016/j.jestch.2025.102057","DOIUrl":"10.1016/j.jestch.2025.102057","url":null,"abstract":"<div><div>Nowadays, with the demand for renewable energy the interest in wind energy is increasing day by day and the use of wind turbines is becoming widespread. In order to efficiently generate electricity from wind turbines, it is important that the turbines are correctly positioned on wind farms. In this study, a standard wind farm area of 2 km × 2 km was used for wind turbine layout. In addition, different from the literature, a 4 km × 4 km wind farm area was also created. Both wind farming areas were divided into 10 × 10 and 20 × 20 grids as in the literature. In addition, 25 × 25, 40 × 40 and 50 × 50 grids were also created. Thus, the wind farming area is divided into more grids and more flexible positioning of the turbines is aimed. There are three different case scenarios for the layout of the wind turbines. Case A has a constant wind speed of 12 m/s and unidirectional wind, while Case B has a constant wind speed of 12 m/s with a 36-directional 10° angle. Case C has variable wind speeds of 8, 12 and 17 m/s with a 36-directional 10° angle. The HHO algorithm was used to perform all these processes. However, since the wind turbine layout problem is binary, the HHO algorithm is adapted to binary with logic operators. These binary methods are called HHO<sub>AND</sub> and HHO<sub>XOR</sub>. The proposed methods achieved competitive results compared to other algorithms in the literature and performed well for all grid structures. Thus, it can be said that the proposed methods are effective for the wind turbine layout problem.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102057"},"PeriodicalIF":5.1,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143843262","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}
With the widespread adoption of Industrial Internet of Things (IIoT) technologies and the rapid advancement of cloud storage, enterprises and users have benefited greatly from the convenience of industrial data sharing. A crucial initial step in this process is the ability to perform fast and efficient keyword searches to help users locate the required data within vast datasets. However, the risk of privacy breaches, the presence of malicious actors, and untrusted cloud servers pose significant threats to both individuals and primary data custodians. Ensuring equitable and precise data searches, while safeguarding sensitive industrial data, is therefore of paramount importance in industrial data-sharing systems. To address these challenges, this paper proposes a dynamic, anonymous, and traceable multi-keyword search scheme that integrates attribute-based encryption (ABE) with blockchain technology. A tracing algorithm is introduced to prevent private key leakage, and a Bloom filter is employed to conceal confidential access policies, preserving user privacy. Blockchain is utilized to generate user identities and record interactions between users and the system, ensuring fair search results, enhancing system security, and maintaining data integrity. To improve practicality, we introduce a fine-grained matching algorithm capable of generating four distinct types of feedback based on user-submitted information. Performance evaluations and security analyses demonstrate the solution’s multifunctionality and high efficiency, highlighting its potential for industrial applications.
{"title":"BA-AMST: A blockchain-assisted anonymous and multi-keyword searchable-traceable data sharing system in Industrial Internet of Things","authors":"Leyou Zhang , Runze Tian , Qing Wu , Fatemeh Rezaeibagha","doi":"10.1016/j.jestch.2025.102055","DOIUrl":"10.1016/j.jestch.2025.102055","url":null,"abstract":"<div><div>With the widespread adoption of Industrial Internet of Things (IIoT) technologies and the rapid advancement of cloud storage, enterprises and users have benefited greatly from the convenience of industrial data sharing. A crucial initial step in this process is the ability to perform fast and efficient keyword searches to help users locate the required data within vast datasets. However, the risk of privacy breaches, the presence of malicious actors, and untrusted cloud servers pose significant threats to both individuals and primary data custodians. Ensuring equitable and precise data searches, while safeguarding sensitive industrial data, is therefore of paramount importance in industrial data-sharing systems. To address these challenges, this paper proposes a dynamic, anonymous, and traceable multi-keyword search scheme that integrates attribute-based encryption (ABE) with blockchain technology. A tracing algorithm is introduced to prevent private key leakage, and a Bloom filter is employed to conceal confidential access policies, preserving user privacy. Blockchain is utilized to generate user identities and record interactions between users and the system, ensuring fair search results, enhancing system security, and maintaining data integrity. To improve practicality, we introduce a fine-grained matching algorithm capable of generating four distinct types of feedback based on user-submitted information. Performance evaluations and security analyses demonstrate the solution’s multifunctionality and high efficiency, highlighting its potential for industrial applications.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102055"},"PeriodicalIF":5.1,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143838074","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}
Pub Date : 2025-04-16DOI: 10.1016/j.jestch.2025.102060
Tao Zhang, Wei Yan, Mengxia Zhou
Traditional passive electromagnetic interference (EMI) filters typically need to use common-mode chokes with high inductance values to ensure that switch-mode power supplies (SMPSs) meet relevant electromagnetic compatibility (EMC) standards. This approach not only increases the weight and size of SMPSs but also reduces their power density. To address these issues, this article proposes a method for modelling and performance evaluation of filtering components for SMPSs based on piezoelectric effect. Piezoelectric filtering components (PFCs) possess low impedance properties at their resonant frequency, while outside of resonance, their impedance is similar to that of a capacitor. As a result, PFCs can not only serve as replacements for traditional interference suppression capacitors (e.g., Y-capacitors) used in passive EMI filters but also effectively suppress interference peaks at specific frequencies. Based on the impedance properties of PFCs, this article proposes a resonant frequency design model. This model can precisely tune the multiple resonant frequencies of PFCs by adjusting their shapes and sizes, thereby leveraging the resonant properties of PFCs to selectively suppress interference peaks at multiple specific frequencies in SMPSs. This article applies the PFCs to the flyback converter and investigates their conducted EMI suppression effectiveness. According to the measurement results, compared to the Y-capacitors, the PFCs reduce the weight and volume of the passive EMI filter by 88.72% and 90.71%, respectively. The experimental results indicate that, based on the model developed in this article, using PFCs as a filtering component can lead to a lighter and more compact filter design.
{"title":"Modeling and performance evaluation of filtering components for switch-mode power supplies based on piezoelectric effect","authors":"Tao Zhang, Wei Yan, Mengxia Zhou","doi":"10.1016/j.jestch.2025.102060","DOIUrl":"10.1016/j.jestch.2025.102060","url":null,"abstract":"<div><div>Traditional passive electromagnetic interference (EMI) filters typically need to use common-mode chokes with high inductance values to ensure that switch-mode power supplies (SMPSs) meet relevant electromagnetic compatibility (EMC) standards. This approach not only increases the weight and size of SMPSs but also reduces their power density. To address these issues, this article proposes a method for modelling and performance evaluation of filtering components for SMPSs based on piezoelectric effect. Piezoelectric filtering components (PFCs) possess low impedance properties at their resonant frequency, while outside of resonance, their impedance is similar to that of a capacitor. As a result, PFCs can not only serve as replacements for traditional interference suppression capacitors (e.g., Y-capacitors) used in passive EMI filters but also effectively suppress interference peaks at specific frequencies. Based on the impedance properties of PFCs, this article proposes a resonant frequency design model. This model can precisely tune the multiple resonant frequencies of PFCs by adjusting their shapes and sizes, thereby leveraging the resonant properties of PFCs to selectively suppress interference peaks at multiple specific frequencies in SMPSs. This article applies the PFCs to the flyback converter and investigates their conducted EMI suppression effectiveness. According to the measurement results, compared to the Y-capacitors, the PFCs reduce the weight and volume of the passive EMI filter by 88.72% and 90.71%, respectively. The experimental results indicate that, based on the model developed in this article, using PFCs as a filtering component can lead to a lighter and more compact filter design.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102060"},"PeriodicalIF":5.1,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143833594","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}
Based on the overall Field Oriented Control (FOC) control strategy of a Permanent Magnet Synchronous Motor (PMSM), a flexible and efficient control system architecture is developed in this work to achieve superior control performance. Sliding Mode Control (SMC) laws are utilized for both the inner and outer loop, but the typical cascade control characteristic of the system is maintained. Thus, the inner loop (IL) control laws are designed to provide increased flexibility by using fractional order (FO) computation and a response speed that is an order of magnitude higher than that of the outer loop (OL). The optimization of the tuning parameters of these controllers is performed by a computational intelligence (CI) algorithm, more specifically the Improved Grey Wolf Optimizer-Cuckoo Search Optimization (IGWO-CSO). The minimization of the computation time in the implementation of control algorithms is achieved by using a neural network (NN) that estimates the derivative value of the sliding surface in the structure of the SMC type speed controller. A term is added to the control law to cancel global perturbations of the system model, estimated with a Disturbance Observer (DO). Mitigation of the numerical stability problems of the derivative computation is achieved by using a Levant observer tracking differentiator. The use of Multi-Agent Reinforcement Learning (MARL) based on three properly trained Twin-Delayed Deep Deterministic (TD3) RL agents, which provide correction signals overlapping the control signals, contributes to the superior performance of the sensorless control system of the PMSM (SCS-PMSM). These include both parametric robustness to parameter and load torque variations, but also the use of an adaptation law to estimate the stator resistance, which can vary significantly. The superiority of the proposed SCS-PMSM over a benchmark control system based on Proportional Integrator (PI) controllers is demonstrated by following both the Software-in-the-Loop (SIL) and Hardware-in-the-Loop Simulated-Rapid Control Prototyping (HILS-RCP) phases. The realization of an RCP for the proposed RCP SCS-PMSM at different sampling periods corresponding to the implementation in both high performance and low/medium performance Digital Signal Processors (DSPs) is achieved using a SpeedGoat Performance Real-Time Target Machine platform.
{"title":"Rapid control prototyping of sensorless control system for PMSM based on multi-agent reinforcement learning and fractional order sliding mode control","authors":"Marcel Nicola , Claudiu-Ionel Nicola , Dan Selișteanu , Dorin Șendrescu","doi":"10.1016/j.jestch.2025.102054","DOIUrl":"10.1016/j.jestch.2025.102054","url":null,"abstract":"<div><div>Based on the overall Field Oriented Control (FOC) control strategy of a Permanent Magnet Synchronous Motor (PMSM), a flexible and efficient control system architecture is developed in this work to achieve superior control performance. Sliding Mode Control (SMC) laws are utilized for both the inner and outer loop, but the typical cascade control characteristic of the system is maintained. Thus, the inner loop (IL) control laws are designed to provide increased flexibility by using fractional order (FO) computation and a response speed that is an order of magnitude higher than that of the outer loop (OL). The optimization of the tuning parameters of these controllers is performed by a computational intelligence (CI) algorithm, more specifically the Improved Grey Wolf Optimizer-Cuckoo Search Optimization (IGWO-CSO). The minimization of the computation time in the implementation of control algorithms is achieved by using a neural network (NN) that estimates the derivative value of the sliding surface in the structure of the SMC type speed controller. A term is added to the control law to cancel global perturbations of the system model, estimated with a Disturbance Observer (DO). Mitigation of the numerical stability problems of the derivative computation is achieved by using a Levant observer tracking differentiator. The use of Multi-Agent Reinforcement Learning (MARL) based on three properly trained Twin-Delayed Deep Deterministic (TD3) RL agents, which provide correction signals overlapping the control signals, contributes to the superior performance of the sensorless control system of the PMSM (SCS-PMSM). These include both parametric robustness to parameter and load torque variations, but also the use of an adaptation law to estimate the stator resistance, which can vary significantly. The superiority of the proposed SCS-PMSM over a benchmark control system based on Proportional Integrator (PI) controllers is demonstrated by following both the Software-in-the-Loop (SIL) and Hardware-in-the-Loop Simulated-Rapid Control Prototyping (HILS-RCP) phases. The realization of an RCP for the proposed RCP SCS-PMSM at different sampling periods corresponding to the implementation in both high performance and low/medium performance Digital Signal Processors (DSPs) is achieved using a SpeedGoat Performance Real-Time Target Machine platform.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102054"},"PeriodicalIF":5.1,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143817428","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}
Pub Date : 2025-04-11DOI: 10.1016/S2215-0986(25)00121-1
{"title":"Front Matter 1 - Full Title Page (regular issues)/Special Issue Title page (special issues)","authors":"","doi":"10.1016/S2215-0986(25)00121-1","DOIUrl":"10.1016/S2215-0986(25)00121-1","url":null,"abstract":"","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"65 ","pages":"Article 102066"},"PeriodicalIF":5.1,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143820772","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}
Pub Date : 2025-04-11DOI: 10.1016/j.jestch.2025.102050
Duba Sriveni , Dr.Loganathan R
In this paper, a novel and robust deep spatio-textural acoustic feature ensemble-assisted learning environment is proposed for violence detection in surveillance videos (DestaVNet). As the name indicates, the proposed DestaVNet model exploits visual and acoustic features to perform violence detection. Additionally, to ensure the scalability of the solution, it employs an active learning concept that retains optimally sufficient frames for further computation and thus reduces computational costs decisively. More specifically, the DestaVNet model initially splits input surveillance footage into acoustic and video frames, followed by multi-constraints active learning based on the most representative frame selection. It applied the least confidence (LC), entropy margin (EM), and margin sampling (MS) criteria to retain the optimal frames for further feature extraction. The DestaVNet model executes pre-processing and feature extraction separately over the frames and corresponding acoustic signals. It performs intensity equalization, histogram equalization, resizing and z-score normalization as pre-processing task, which is followed by deep spatio-textural feature extraction by using gray level co-occurrence matrix (GLCM), ResNet101 and SqueezeNet deep networks. On the other hand, the different acoustic features, including mel-frequency cepstral coefficient (MFCC), gammatone cepstral coefficient (GTCC), , harmonic to noise ratio (HNR), spectral features and pitch were obtained. These acoustic and spatio-textural features were fused to yield a composite audio-visual feature set, which was later processed for principal component analysis (PCA) to minimize redundancy, and k-NN as part of an ensemble classifier to enhance prediction accuracy, achieving superior performance. The z-score normalization was performed to alleviate the over-fitting problem. Finally, the retained feature sets were processed for two-class classification by using a heterogeneous ensemble learning model, embodying SVM, DT, k-NN, NB, and RF classifiers. Simulation results confirmed that the proposed DestaVNet model outperforms other existing violence prediction methods, where its superiority was affirmed in terms of the (99.92%), precision (99.67%), recall (99.29%) and F-Measure (0.992).
{"title":"An active learning driven deep spatio-textural acoustic feature ensemble assisted learning environment for violence detection in surveillance videos","authors":"Duba Sriveni , Dr.Loganathan R","doi":"10.1016/j.jestch.2025.102050","DOIUrl":"10.1016/j.jestch.2025.102050","url":null,"abstract":"<div><div>In this paper, a novel and robust deep spatio-textural acoustic feature ensemble-assisted learning environment is proposed for violence detection in surveillance videos (DestaVNet). As the name indicates, the proposed DestaVNet model exploits visual and acoustic features to perform violence detection. Additionally, to ensure the scalability of the solution, it employs an active learning concept that retains optimally sufficient frames for further computation and thus reduces computational costs decisively. More specifically, the DestaVNet model initially splits input surveillance footage into acoustic and video frames, followed by multi-constraints active learning based on the most representative frame selection. It applied the least confidence (LC), entropy margin (EM), and margin sampling (MS) criteria to retain the optimal frames for further feature extraction. The DestaVNet model executes pre-processing and feature extraction separately over the frames and corresponding acoustic signals. It performs intensity equalization, histogram equalization, resizing and z-score normalization as pre-processing task, which is followed by deep spatio-textural feature extraction by using gray level co-occurrence matrix (GLCM), ResNet101 and SqueezeNet deep networks. On the other hand, the different acoustic features, including mel-frequency cepstral coefficient (MFCC), gammatone cepstral coefficient (GTCC), <span><math><mrow><mi>GTCC</mi><mo>-</mo><mi>Δ</mi></mrow></math></span>, harmonic to noise ratio (HNR), spectral features and pitch were obtained. These acoustic and spatio-textural features were fused to yield a composite audio-visual feature set, which was later processed for principal component analysis (PCA) to minimize redundancy, and k-NN as part of an ensemble classifier to enhance prediction accuracy, achieving superior performance. The z-score normalization was performed to alleviate the over-fitting problem. Finally, the retained feature sets were processed for two-class classification by using a heterogeneous ensemble learning model, embodying SVM, DT, k-NN, NB, and RF classifiers. Simulation results confirmed that the proposed DestaVNet model outperforms other existing violence prediction methods, where its superiority was affirmed in terms of the (99.92%), precision (99.67%), recall (99.29%) and F-Measure (0.992).</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102050"},"PeriodicalIF":5.1,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143817426","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}
Pub Date : 2025-04-10DOI: 10.1016/j.jestch.2025.102049
Suhaila ZeinElabideen Omer , Fazirulhisyam Hashim , Aduwati Sali , Faisul Arif Ahmad
The data and size of networks have grown substantially due to the rapid development of the Internet and other communication techniques. This has led to the development of numerous new types of attacks, making it harder for network security to detect intrusions accurately. The goal of a Denial of Service (DoS) attack is to overwhelm a target with malicious traffic, exhausting its processing power and network bandwidth. Traditional DoS attacks rely on brute force techniques, making them easier to detect, whereas low-rate and slow attacks pose a greater threat due to their stealthy nature. These attacks target application or server resources with a prolonged trickle of traffic, requiring minimal bandwidth yet making mitigation challenging. Their low resource footprint allows them to degrade or deny service to legitimate users while remaining undetected for extended periods. This research introduces an advanced Intrusion Detection System (IDS) that utilizes a hybrid Long Short-Term Memory Feedforward (LSTM-FF) Neural Network to tackle existing challenges in detecting low-rate DoS (LR-DoS) attacks. Unlike previous models, our approach combines temporal sequence learning with feature refinement, thereby improving the detection of LR-DoS. Additionally, we incorporate automated feature selection using Random Forest, which optimizes efficiency while maintaining interpretability. For model training and evaluation, we use the CIC-DOS2017 dataset, which includes eight distinct types of LR-DoS attacks. To enhance generalizability, we also utilize the CSE-CIC-IDS2018 dataset and the newly introduced LR-HR-DDOS2024 dataset, specifically designed for Software-Defined Networking (SDN)-based environments. To address the class imbalance, we implement a stratified k-fold cross-validation strategy, ensuring robust performance across various attack scenarios. To thoroughly evaluate model performance, we adopt a comprehensive set of metrics, including accuracy, precision, recall, F1-score, specificity, False Alarm Rate (FAR), and ROC-AUC. This ensures a well-rounded validation of our approach. The model surpassed all previous state-of-the-art models with an impressive accuracy of 99.70%, precision of 99.47%, specificity of 99.97%, and an F1-score of 97.52%, all while retaining a low FAR of roughly 0.03%. The LSTM-FF approach also worked well in multi-class classification, with a 99.54% accuracy rate, 93.19% precision, 99.59% specificity, 90.28% F1 score, and 0.40% FAR.
{"title":"Binary classification of Low-Rate DoS attacks using Long Short-Term Memory Feed-Forward (LSTM-FF) Intrusion Detection System (IDS)","authors":"Suhaila ZeinElabideen Omer , Fazirulhisyam Hashim , Aduwati Sali , Faisul Arif Ahmad","doi":"10.1016/j.jestch.2025.102049","DOIUrl":"10.1016/j.jestch.2025.102049","url":null,"abstract":"<div><div>The data and size of networks have grown substantially due to the rapid development of the Internet and other communication techniques. This has led to the development of numerous new types of attacks, making it harder for network security to detect intrusions accurately. The goal of a Denial of Service (DoS) attack is to overwhelm a target with malicious traffic, exhausting its processing power and network bandwidth. Traditional DoS attacks rely on brute force techniques, making them easier to detect, whereas low-rate and slow attacks pose a greater threat due to their stealthy nature. These attacks target application or server resources with a prolonged trickle of traffic, requiring minimal bandwidth yet making mitigation challenging. Their low resource footprint allows them to degrade or deny service to legitimate users while remaining undetected for extended periods. This research introduces an advanced Intrusion Detection System (IDS) that utilizes a hybrid Long Short-Term Memory Feedforward (LSTM-FF) Neural Network to tackle existing challenges in detecting low-rate DoS (LR-DoS) attacks. Unlike previous models, our approach combines temporal sequence learning with feature refinement, thereby improving the detection of LR-DoS. Additionally, we incorporate automated feature selection using Random Forest, which optimizes efficiency while maintaining interpretability. For model training and evaluation, we use the CIC-DOS2017 dataset, which includes eight distinct types of LR-DoS attacks. To enhance generalizability, we also utilize the CSE-CIC-IDS2018 dataset and the newly introduced LR-HR-DDOS2024 dataset, specifically designed for Software-Defined Networking (SDN)-based environments. To address the class imbalance, we implement a stratified k-fold cross-validation strategy, ensuring robust performance across various attack scenarios. To thoroughly evaluate model performance, we adopt a comprehensive set of metrics, including accuracy, precision, recall, F1-score, specificity, False Alarm Rate (FAR), and ROC-AUC. This ensures a well-rounded validation of our approach. The model surpassed all previous state-of-the-art models with an impressive accuracy of 99.70%, precision of 99.47%, specificity of 99.97%, and an F1-score of 97.52%, all while retaining a low FAR of roughly 0.03%. The LSTM-FF approach also worked well in multi-class classification, with a 99.54% accuracy rate, 93.19% precision, 99.59% specificity, 90.28% F1 score, and 0.40% FAR.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102049"},"PeriodicalIF":5.1,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143807384","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}
Pub Date : 2025-04-10DOI: 10.1016/j.jestch.2025.102052
Yu Yang, Jiahui Xu, Qiangbing Zhou, Shichao Kong, Keyi Lin
The prediction model for the Poisson’s ratio of concrete damage is of significant importance in the field of Structural Health Monitoring (SHM). Seeking a concrete damage Poisson’s ratio prediction model that comprehensively reflects the characteristics of concrete while also being simple and accurate is a challenging task. This study proposes a combination of the Kolmogorov-Arnold Network (KAN), which can fit complex nonlinear relationships with high precision, and the Finite Element Method (FEM) to address this challenge. The research first summarizes the influencing factors of the concrete damage Poisson’s ratio model from classical theories, then uses data obtained from measurements and finite element analysis to train the KAN to develop the concrete damage Poisson’s ratio prediction model. Finally, the accuracy of the model is validated on a test set, and its performance is compared with that of Multi-Layer Perceptron (MLP) networks and classical models. The validation results indicate that the formula model trained by KAN achieves a Root Mean Square Error (RMSE) of 0.055 when predicting the damage Poisson’s ratio of actual test specimens, outperforming four classical models (RMSE ≥ 0.176). The novelty of this study lies in the innovative application of KAN in the concrete damage Poisson’s ratio prediction model, as well as the approach of combining a small amount of measured data with FEM to enhance the efficiency of generating training and testing data. This research not only validates the interpretability and accuracy of KAN but also demonstrates the practicality and effectiveness of the KAN and FEM combination method in the application of predicting the concrete damage Poisson’s ratio, making a significant contribution to the field.
混凝土损伤泊松比预测模型在结构健康监测（SHM）领域具有重要意义。寻找一个既能全面反映混凝土特性，又简单准确的混凝土损伤泊松比预测模型是一项具有挑战性的任务。本研究提出将可高精度拟合复杂非线性关系的柯尔莫哥洛夫-阿诺德网络（KAN）与有限元法（FEM）相结合来应对这一挑战。研究首先从经典理论中总结了混凝土破坏泊松比模型的影响因素，然后利用测量和有限元分析获得的数据训练 KAN，开发出混凝土破坏泊松比预测模型。最后，在测试集上验证了模型的准确性，并将其性能与多层感知器（MLP）网络和经典模型进行了比较。验证结果表明，由 KAN 训练的公式模型在预测实际试样的损坏泊松比时，均方根误差（RMSE）为 0.055，优于四个经典模型（RMSE ≥ 0.176）。本研究的创新之处在于将 KAN 创新性地应用于混凝土破坏泊松比预测模型，以及将少量测量数据与有限元相结合的方法，以提高生成训练和测试数据的效率。这项研究不仅验证了 KAN 的可解释性和准确性，还证明了 KAN 与有限元相结合的方法在预测混凝土破坏泊松比应用中的实用性和有效性，为该领域做出了重大贡献。
{"title":"Fitting method of concrete damage Poisson’s ratio model based on Kolmogorov-Arnold network","authors":"Yu Yang, Jiahui Xu, Qiangbing Zhou, Shichao Kong, Keyi Lin","doi":"10.1016/j.jestch.2025.102052","DOIUrl":"10.1016/j.jestch.2025.102052","url":null,"abstract":"<div><div>The prediction model for the Poisson’s ratio of concrete damage is of significant importance in the field of Structural Health Monitoring (SHM). Seeking a concrete damage Poisson’s ratio prediction model that comprehensively reflects the characteristics of concrete while also being simple and accurate is a challenging task. This study proposes a combination of the Kolmogorov-Arnold Network (KAN), which can fit complex nonlinear relationships with high precision, and the Finite Element Method (FEM) to address this challenge. The research first summarizes the influencing factors of the concrete damage Poisson’s ratio model from classical theories, then uses data obtained from measurements and finite element analysis to train the KAN to develop the concrete damage Poisson’s ratio prediction model. Finally, the accuracy of the model is validated on a test set, and its performance is compared with that of Multi-Layer Perceptron (MLP) networks and classical models. The validation results indicate that the formula model trained by KAN achieves a Root Mean Square Error (RMSE) of 0.055 when predicting the damage Poisson’s ratio of actual test specimens, outperforming four classical models (RMSE ≥ 0.176). The novelty of this study lies in the innovative application of KAN in the concrete damage Poisson’s ratio prediction model, as well as the approach of combining a small amount of measured data with FEM to enhance the efficiency of generating training and testing data. This research not only validates the interpretability and accuracy of KAN but also demonstrates the practicality and effectiveness of the KAN and FEM combination method in the application of predicting the concrete damage Poisson’s ratio, making a significant contribution to the field.</div></div>","PeriodicalId":48609,"journal":{"name":"Engineering Science and Technology-An International Journal-Jestech","volume":"66 ","pages":"Article 102052"},"PeriodicalIF":5.1,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143817427","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号