Pub Date : 2025-12-20DOI: 10.1016/j.asej.2025.103895
Ziheng Zhao, Elmi Bin Abu Bakar, Norizham Bin Abdul Razak, Mohammad Nishat Akhtar
This paper proposes a classification model based on a corrosion and salt damage dataset of civil structures: ViT is used as the teacher model, while MobileNetV2 and MobileNetV3 are used as student models. The second-best model is obtained by improving the classifier and setting different ratios of fine-tuning layers under frozen versus non-frozen classification layers. This model is added with an improved attention mechanism to get the optimal model. The final results show that using ViT with the second weight and MobileNetV2 gives higher Accuracy and Weighted-f1 value, where heat maps generated by applying Grad-CAM reflect that it can generally identify the damage location. The optimal model obtained by choosing a fine-tuning strategy that freezes the classification layers and retrains 15% of the top feature layers can achieve an Accuracy and Weighted-f1 score of above 0.94, better than many advanced deep learning architectures using pre-trained weights.
{"title":"Knowledge distillation and fine-tuning for corrosion and salt damage classification","authors":"Ziheng Zhao, Elmi Bin Abu Bakar, Norizham Bin Abdul Razak, Mohammad Nishat Akhtar","doi":"10.1016/j.asej.2025.103895","DOIUrl":"10.1016/j.asej.2025.103895","url":null,"abstract":"<div><div>This paper proposes a classification model based on a corrosion and salt damage dataset of civil structures: ViT is used as the teacher model, while MobileNetV2 and MobileNetV3 are used as student models. The second-best model is obtained by improving the classifier and setting different ratios of fine-tuning layers under frozen versus non-frozen classification layers. This model is added with an improved attention mechanism to get the optimal model. The final results show that using ViT with the second weight and MobileNetV2 gives higher Accuracy and Weighted-f1 value, where heat maps generated by applying Grad-CAM reflect that it can generally identify the damage location. The optimal model obtained by choosing a fine-tuning strategy that freezes the classification layers and retrains 15% of the top feature layers can achieve an Accuracy and Weighted-f1 score of above 0.94, better than many advanced deep learning architectures using pre-trained weights.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103895"},"PeriodicalIF":5.9,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790131","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}
This study examines Thailand’s first hydropower and solar PV-assisted smart microgrid system operating in real time at Royal Project Intanon. The smart microgrid system comprises a dual 90 kW hydropower generation unit, a 20 kW solar PV system, and a 100 kW Battery Energy Storage System (BESS), with a conventional grid connection. During smart microgrid operation, first-order and average filter control methods are employed to prioritise the solar PV system and the BESS, given their faster response to grid load demand. Initially, BESS favours stabilising both frequency and voltage in the closed microgrid within 30 s, thereby coordinating the PV systems to resume operation. Following that, hydropower generation is synchronised with the smart microgrid system. Although most load demand is met by hydropower, the initial stage of islanding relies on the battery energy storage system. The most extended power outage occurred in September 2021, lasting 1661 min. On a selective day in September, BESS peak charging and discharging were −36.06 kW and 35.51 kW, respectively, whereas the average solar PV generation was below 2 kW. Throughout the power outage, hydropower generation consistently met load demand, and the BESS initiated charging and discharging processes as required by the load profile. Following that, a selective day in August shows that the BESS initiated charging and discharging 59 times. A year-round performance analysis indicates that the examined commercial microgrid controller avoided 5232 min of grid power outages without using a diesel generator and met the load demand efficiently.
{"title":"Performance analysis of a smart microgrid controller for under real-time operating conditions at Royal Project Intanon","authors":"Xianwen Zhu , Buntoon Wiengmoon , Tawat Suriwong , Chatchai Sirisamphanwong","doi":"10.1016/j.asej.2025.103938","DOIUrl":"10.1016/j.asej.2025.103938","url":null,"abstract":"<div><div>This study examines Thailand’s first hydropower and solar PV-assisted smart microgrid system operating in real time at Royal Project Intanon. The smart microgrid system comprises a dual 90 kW hydropower generation unit, a 20 kW solar PV system, and a 100 kW Battery Energy Storage System (BESS), with a conventional grid connection. During smart microgrid operation, first-order and average filter control methods are employed to prioritise the solar PV system and the BESS, given their faster response to grid load demand. Initially, BESS favours stabilising both frequency and voltage in the closed microgrid within 30 s, thereby coordinating the PV systems to resume operation. Following that, hydropower generation is synchronised with the smart microgrid system. Although most load demand is met by hydropower, the initial stage of islanding relies on the battery energy storage system. The most extended power outage occurred in September 2021, lasting 1661 min. On a selective day in September, BESS peak charging and discharging were −36.06 kW and 35.51 kW, respectively, whereas the average solar PV generation was below 2 kW. Throughout the power outage, hydropower generation consistently met load demand, and the BESS initiated charging and discharging processes as required by the load profile. Following that, a selective day in August shows that the BESS initiated charging and discharging 59 times. A year-round performance analysis indicates that the examined commercial microgrid controller avoided 5232 min of grid power outages without using a diesel generator and met the load demand efficiently.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103938"},"PeriodicalIF":5.9,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789595","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}
This study proposes a heuristic seamless supervisory control scheme (SECS) for a constrained grid-connected microgrid with green hydrogen production and storage system (GHPS). Beside the photovoltaic generator (PVG) and the wind generator (WG), the microgrid has an alkaline electrolyser (AE) for hydrogen production, a high-pressure storage tank with a compressor, and proton-exchange membrane fuel cell (FC). It supplies both DC and AC loads. The SECS coordinates the individual controllers of the microgrid subsystems to achieve stable and economic operation. It needs only the three-phase main grid current and hydrogen tank level as inputs to make real time decisions specifying energy-sharing status for the microgrid players. Also, it maintains voltage quality and enables smooth state transition under all operating conditions. By employing elaborated mathematical models, the study analyses the operational dynamics and working efficiency of main components for various scenarios. Two proposed configurations of the microgrid are considered. The entire electrical–mechanical dynamic model is simulated in Matlab environment. Obtained results prove the scheme efficacy and verify the seamless performance of the proposed configurations.
{"title":"Seamless supervisory control scheme of a constrained grid-connected microgrid with hybrid resources and hydrogen production unit","authors":"Safwan Edris , Akram Elmitwally , Mohamed Elgohary , Abdelhady Ghanem","doi":"10.1016/j.asej.2025.103945","DOIUrl":"10.1016/j.asej.2025.103945","url":null,"abstract":"<div><div>This study proposes a heuristic seamless supervisory control scheme (SECS) for a constrained grid-connected microgrid with green hydrogen production and storage system (GHPS). Beside the photovoltaic generator (PVG) and the wind generator (WG), the microgrid has an alkaline electrolyser (AE) for hydrogen production, a high-pressure storage tank with a compressor, and proton-exchange membrane fuel cell (FC). It supplies both DC and AC loads. The SECS coordinates the individual controllers of the microgrid subsystems to achieve stable and economic operation. It needs only the three-phase main grid current and hydrogen tank level as inputs to make real time decisions specifying energy-sharing status for the microgrid players. Also, it maintains voltage quality and enables smooth state transition under all operating conditions. By employing elaborated mathematical models, the study analyses the operational dynamics and working efficiency of main components for various scenarios. Two proposed configurations of the microgrid are considered. The entire electrical–mechanical dynamic model is simulated in Matlab environment. Obtained results prove the scheme efficacy and verify the seamless performance of the proposed configurations.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103945"},"PeriodicalIF":5.9,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789596","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 : 2025-12-17DOI: 10.1016/j.asej.2025.103899
Zhenglin Zhang , Muhammad Adil Khan , Jamroz Khan , Shah Faisal , Xuewu Zuo , Mohammed Kbiri Alaoui
Over the past decade, the Hermite-Hadamard inequality has attracted significant attention from mathematicians, leading to the development of various extensions and generalizations involving different fractional operators, stochastic processes, geometrical interpretations, and applications in image processing. This study focuses on addressing problems related to the Hermite-Hadamard-Jensen-Mercer inequality by incorporating weighted arithmetic means instead of unweighted means within the framework of conformable fractional integral operators. A key objective is to estimate the difference between the derived inequalities. To this end, a novel integral identity is established. Utilizing this identity, several bounds for the difference of the inequalities are obtained by applying convexity, the Hölder inequality, and power-mean inequality. The results are further supported by applications to various means. Finally, the validity and accuracy of the derived results are demonstrated through two-dimensional and three-dimensional graphical representations.
{"title":"A version of Hermite-Hadamard-Mercer inequality and associated results","authors":"Zhenglin Zhang , Muhammad Adil Khan , Jamroz Khan , Shah Faisal , Xuewu Zuo , Mohammed Kbiri Alaoui","doi":"10.1016/j.asej.2025.103899","DOIUrl":"10.1016/j.asej.2025.103899","url":null,"abstract":"<div><div>Over the past decade, the Hermite-Hadamard inequality has attracted significant attention from mathematicians, leading to the development of various extensions and generalizations involving different fractional operators, stochastic processes, geometrical interpretations, and applications in image processing. This study focuses on addressing problems related to the Hermite-Hadamard-Jensen-Mercer inequality by incorporating weighted arithmetic means instead of unweighted means within the framework of conformable fractional integral operators. A key objective is to estimate the difference between the derived inequalities. To this end, a novel integral identity is established. Utilizing this identity, several bounds for the difference of the inequalities are obtained by applying convexity, the Hölder inequality, and power-mean inequality. The results are further supported by applications to various means. Finally, the validity and accuracy of the derived results are demonstrated through two-dimensional and three-dimensional graphical representations.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103899"},"PeriodicalIF":5.9,"publicationDate":"2025-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790132","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 : 2025-12-15DOI: 10.1016/j.asej.2025.103915
Dumitru Baleanu , Babak Shiri
This study focuses on fractional differential equations defined by the Neumann series operator. The inverse of the fractional integral operator is associated with a MABC operator. The underlying fractional Neumann equation is proven to be equivalent to weakly singular integral equations. A new collocation method is proposed for the numerical solution. The method separates the solution into regular and non-regular components. Convergence, super-convergence, and stability of the method are obtained. Unlike finite difference methods, the order of convergence is not reduced on uniform meshes. Numerical examples are provided to validate the theoretical findings.
本文主要研究由诺伊曼级数算子定义的分数阶微分方程。分数阶积分算子（1−α）I0+α i α的逆与MABC算子相关。证明了分数阶诺伊曼方程等价于弱奇异积分方程。提出了一种新的数值解的配点法。该方法将溶液分为规则组分和非规则组分。得到了该方法的收敛性、超收敛性和稳定性。与有限差分法不同，在均匀网格上，收敛阶不会降低。数值算例验证了理论结果。
{"title":"Numerical solution for fractional Neumann series equations","authors":"Dumitru Baleanu , Babak Shiri","doi":"10.1016/j.asej.2025.103915","DOIUrl":"10.1016/j.asej.2025.103915","url":null,"abstract":"<div><div>This study focuses on fractional differential equations defined by the Neumann series operator. The inverse of the fractional integral operator <span><math><mo>(</mo><mn>1</mn><mo>−</mo><mi>α</mi><mo>)</mo><msup><mrow><mi>I</mi></mrow><mrow><mn>0</mn></mrow></msup><mo>+</mo><mi>α</mi><msup><mrow><mi>I</mi></mrow><mrow><mi>α</mi></mrow></msup></math></span> is associated with a MABC operator. The underlying fractional Neumann equation is proven to be equivalent to weakly singular integral equations. A new collocation method is proposed for the numerical solution. The method separates the solution into regular and non-regular components. Convergence, super-convergence, and stability of the method are obtained. Unlike finite difference methods, the order of convergence is not reduced on uniform meshes. Numerical examples are provided to validate the theoretical findings.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103915"},"PeriodicalIF":5.9,"publicationDate":"2025-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789598","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 : 2025-12-15DOI: 10.1016/j.asej.2025.103928
Teng Ran, Jianxing Wu, Qing Tao
Low-light conditions degrade image information and affect the performance of visual perception tasks. Most low-light image enhancement methods currently rely on costly paired datasets for training. Meanwhile, many unsupervised models often face challenges in effectively recovering details, spatial structure, and color information. To address these issues, this paper proposes a multi-scale spatial structure restoration architecture, which is trained in an unsupervised manner using SSIM loss and Smoothness loss functions. To simultaneously capture global information and preserve details, we use the Haar Wavelet Downsample to extract image features at different scales. We introduce the Global-Contextual Relay Aggregation module that enhances and restores global and local features. Additionally, we designed the Dual Semantic-Spatial Attention Module with a dual-branch structure. It extracts global structural information and enhances the semantic understanding. We introduced High-Low Frequency Decomposition. It uses the Haar wavelet inverse transform to promote effective fusion of features at different scales. Experimental results show that the proposed method outperforms existing unsupervised approaches on the LOL-v1 (PSNR: 20.0 dB, MAE: 0.0961, DeltaE: 12.3623). The results surpassed supervised models like PairLIE (PSNR: 18.47 dB, MAE: 0.1153, DeltaE: 14.2984). It also demonstrates superior performance in terms of brightness uniformity, detail restoration, and color accuracy on LOL-v2-Real and LOL-v2-Synthetic.
{"title":"Multi-scale spatial-structure restoration network based on unsupervised learning for low-light image enhancement","authors":"Teng Ran, Jianxing Wu, Qing Tao","doi":"10.1016/j.asej.2025.103928","DOIUrl":"10.1016/j.asej.2025.103928","url":null,"abstract":"<div><div>Low-light conditions degrade image information and affect the performance of visual perception tasks. Most low-light image enhancement methods currently rely on costly paired datasets for training. Meanwhile, many unsupervised models often face challenges in effectively recovering details, spatial structure, and color information. To address these issues, this paper proposes a multi-scale spatial structure restoration architecture, which is trained in an unsupervised manner using SSIM loss and Smoothness loss functions. To simultaneously capture global information and preserve details, we use the Haar Wavelet Downsample to extract image features at different scales. We introduce the Global-Contextual Relay Aggregation module that enhances and restores global and local features. Additionally, we designed the Dual Semantic-Spatial Attention Module with a dual-branch structure. It extracts global structural information and enhances the semantic understanding. We introduced High-Low Frequency Decomposition. It uses the Haar wavelet inverse transform to promote effective fusion of features at different scales. Experimental results show that the proposed method outperforms existing unsupervised approaches on the LOL-v1 (PSNR: 20.0 dB, MAE: 0.0961, DeltaE: 12.3623). The results surpassed supervised models like PairLIE (PSNR: 18.47 dB, MAE: 0.1153, DeltaE: 14.2984). It also demonstrates superior performance in terms of brightness uniformity, detail restoration, and color accuracy on LOL-v2-Real and LOL-v2-Synthetic.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103928"},"PeriodicalIF":5.9,"publicationDate":"2025-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789597","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 : 2025-12-15DOI: 10.1016/j.asej.2025.103902
Wang Ben , Chen Hongchu
Existing container security policies struggle to adapt to environmental changes and lack an effective evaluation mechanism. This paper proposes a dynamic security policy generation algorithm using artificial intelligence (AI) and integrates an evaluation mechanism to ensure real-time effectiveness. The model employs Deep Q-Network (DQN) to generate policies, using environmental features like resource usage, network traffic patterns, and threat scores as inputs. Particle Swarm Optimization (PSO) is applied to resolve conflicts and optimize execution efficiency and consistency. A multi-objective regression evaluation mechanism is used to assess policy effectiveness based on detection efficiency, resource usage, and protection accuracy. Experimental results show that the dynamic strategy reduces generation time by 70%, boosts response efficiency by 62.7%, achieves 94.2% task completion, and improves threat detection and protection accuracy by 23.5% and 18.9%, respectively. This method enhances the efficiency and reliability of container security policies in dynamic environments.
{"title":"Dynamic security policy generation and evaluation algorithm for trusted containers based on artificial intelligence","authors":"Wang Ben , Chen Hongchu","doi":"10.1016/j.asej.2025.103902","DOIUrl":"10.1016/j.asej.2025.103902","url":null,"abstract":"<div><div>Existing container security policies struggle to adapt to environmental changes and lack an effective evaluation mechanism. This paper proposes a dynamic security policy generation algorithm using artificial intelligence (AI) and integrates an evaluation mechanism to ensure real-time effectiveness. The model employs Deep Q-Network (DQN) to generate policies, using environmental features like resource usage, network traffic patterns, and threat scores as inputs. Particle Swarm Optimization (PSO) is applied to resolve conflicts and optimize execution efficiency and consistency. A multi-objective regression evaluation mechanism is used to assess policy effectiveness based on detection efficiency, resource usage, and protection accuracy. Experimental results show that the dynamic strategy reduces generation time by 70%, boosts response efficiency by 62.7%, achieves 94.2% task completion, and improves threat detection and protection accuracy by 23.5% and 18.9%, respectively. This method enhances the efficiency and reliability of container security policies in dynamic environments.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103902"},"PeriodicalIF":5.9,"publicationDate":"2025-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790127","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}
This paper presents the application of a three-factor linear model for the analysis and optimization of the main cutting resistance during the machining of PTFE (Polytetrafluoroethylene) material, with nominal dimensions of ∅50 × 500 mm. PTFE is known for its unique mechanical and thermal properties, which pose challenges in machining processes. Key machining parameters—specifically cutting depth, spindle speed, and cutting speed—were initially identified and then collectively examined as machining parameters through regression analysis to determine their interactions and impact on cutting resistance. The aim of this research was to optimize these machining parameters through mathematical modeling to reduce cutting resistance, extend tool life, and enhance productivity. The results demonstrated that proper optimization of the machining parameters can significantly reduce tool wear, lower costs, and improve machining efficiency.
{"title":"Optimization of the main cutting force in the machining of polytetrafluoroethylene (PTFE)","authors":"Slavica Prvulovic , Predrag Mosorinski , Ljubisa Josimovic , Jasna Tolmac , Branislava Radisic , Uros Sarenac","doi":"10.1016/j.asej.2025.103927","DOIUrl":"10.1016/j.asej.2025.103927","url":null,"abstract":"<div><div>This paper presents the application of a three-factor linear model for the analysis and optimization of the main cutting resistance during the machining of PTFE (Polytetrafluoroethylene) material, with nominal dimensions of ∅50 × 500 mm. PTFE is known for its unique mechanical and thermal properties, which pose challenges in machining processes. Key machining parameters—specifically cutting depth, spindle speed, and cutting speed—were initially identified and then collectively examined as machining parameters through regression analysis to determine their interactions and impact on cutting resistance. The aim of this research was to optimize these machining parameters through mathematical modeling to reduce cutting resistance, extend tool life, and enhance productivity. The results demonstrated that proper optimization of the machining parameters can significantly reduce tool wear, lower costs, and improve machining efficiency.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103927"},"PeriodicalIF":5.9,"publicationDate":"2025-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736128","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 : 2025-12-12DOI: 10.1016/j.asej.2025.103888
Asif Raza , Zi-Hong Jiang , Yi-Die Ye , Muhammad Punhal Sahto , Ahmed Lotfy Haridy , Said I. Abouzeid , Ghalib Raza , Jibran Hussain
Techno-economic analysis of off-grid hybrid AC-DC microgrids (HMGs) in desert areas has primarily focused on meeting the load demands in residential, household, domestic, and agricultural applications, while the healthcare sector has been comparatively less considered. These analyses contribute to the development of effective financial techniques. This paper presents a techno-economic design for an HMG that combines AC and DC elements such as diesel generators (DG), photovoltaic (PV), wind turbines (WT), batteries (BAT), and power converters (Conv) to satisfy the power demand of a rural health clinic with a daily load of 110 kWh, situated in the desert area of Nubian in Aswan, Egypt. The optimization of HMG is performed through HOMER Pro based on hourly wind speed, solar irradiance, and clinic load data to evaluate the cost of electricity, carbon emissions, loss of power supply probability, and renewable fraction. The results are compared across four different HMG combinations: PV/WT/BAT/Conv, DG/PV/BAT/Conv, DG/PV/WT/BAT/Conv, and WT/DG. The simulation outcomes indicate that the system incorporating the WT/DG/PV/BAT/Conv provides the most efficient techno-economic solution for meeting the clinic’s power demand. The optimal configuration includes 4 kW of DG, 10 kW of WT, 9 kW of PV, 36 batteries, and 16 kW of power converters. This system achieves the lowest 51.76 k$ of net present cost and 0.107 $/kWh of cost of electricity, 3051 kg/year of CO2 emissions, and a significant renewable contribution of 90.7 %. Furthermore, the sensitivity assessment verifies that the system costs are greatly affected by factors including solar radiation, wind speed, discount rate, and diesel cost.
{"title":"Techno-economic feasibility analysis of a hybrid off grid AC-DC microgrid to support a health clinic","authors":"Asif Raza , Zi-Hong Jiang , Yi-Die Ye , Muhammad Punhal Sahto , Ahmed Lotfy Haridy , Said I. Abouzeid , Ghalib Raza , Jibran Hussain","doi":"10.1016/j.asej.2025.103888","DOIUrl":"10.1016/j.asej.2025.103888","url":null,"abstract":"<div><div>Techno-economic analysis of off-grid hybrid AC-DC microgrids (HMGs) in desert areas has primarily focused on meeting the load demands in residential, household, domestic, and agricultural applications, while the healthcare sector has been comparatively less considered. These analyses contribute to the development of effective financial techniques. This paper presents a techno-economic design for an HMG that combines AC and DC elements such as diesel generators (DG), photovoltaic (PV), wind turbines (WT), batteries (BAT), and power converters (Conv) to satisfy the power demand of a rural health clinic with a daily load of 110 kWh, situated in the desert area of Nubian in Aswan, Egypt. The optimization of HMG is performed through HOMER Pro based on hourly wind speed, solar irradiance, and clinic load data to evaluate the cost of electricity, carbon emissions, loss of power supply probability, and renewable fraction. The results are compared across four different HMG combinations: PV/WT/BAT/Conv, DG/PV/BAT/Conv, DG/PV/WT/BAT/Conv, and WT/DG. The simulation outcomes indicate that the system incorporating the WT/DG/PV/BAT/Conv provides the most efficient techno-economic solution for meeting the clinic’s power demand. The optimal configuration includes 4 kW of DG, 10 kW of WT, 9 kW of PV, 36 batteries, and 16 kW of power converters. This system achieves the lowest 51.76 k$ of net present cost and 0.107 $/kWh of cost of electricity, 3051 kg/year of CO<sub>2</sub> emissions, and a significant renewable contribution of 90.7 %. Furthermore, the sensitivity assessment verifies that the system costs are greatly affected by factors including solar radiation, wind speed, discount rate, and diesel cost.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103888"},"PeriodicalIF":5.9,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736132","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 : 2025-12-12DOI: 10.1016/j.asej.2025.103904
Harish M.S , Lokesh S , Sakthivel P , Akshaya B
Recent Intrusion Detection (ID) networks face difficulties in handling the enlarging volume of network traffic and adapting to emerging cyber threats. Handling data traffic and addressing data imbalance are key requirements for identifying recent cyber threats. This paper proposes a novel hybrid ID system designed to mitigate data imbalance issues. The proposed methodology uses advanced deep learning techniques and optimized characteristic fusion models, making it suitable for high-traffic environments. This research conducts a comprehensive experimental study on five standard ID datasets, focusing on network traffic and system behavior data, which are crucial for detecting potential intrusions. In the deep feature extraction phase, multiple features are considered, including statistical information, T-SNE features, and high-level deep learning features. T-SNE features capture similarities between data points, helping preserve the most important features. For feature fusion, optimal weights are identified using the proposed RCMPA. The fused feature set, created from these optimized weights, improves that a more appropriate and discriminative characteristics are utilized for training. A system then employs a “Multi-scale Dilated Deep Hybrid Network with Attention Mechanism” (MDDHN-AM) for intrusion diagnosis. This developed model integrates TCNN and RNN to detain both temporal and spatial dependencies. TCNN processes sequential information to identify temporal patterns, while RNN captures the dynamic nature of network traffic. The attention mechanism prioritizes the most significant features, enabling more accurate intrusion detection. At last, the presentation of MDDHN-AM was compared to traditional and state-of-the-art intrusion detection methods across multiple metrics. The developed model achieved 96.43% detection accuracy and 97.38% precision, illustrating its efficiency in handling diverse digital attacks and data imbalance. An improved performance over traditional methods highlights its potential as a robust solution for secure communication and protection against evolving digital attacks.
{"title":"Hybrid deep learning model for network intrusion detection using optimal feature fusion","authors":"Harish M.S , Lokesh S , Sakthivel P , Akshaya B","doi":"10.1016/j.asej.2025.103904","DOIUrl":"10.1016/j.asej.2025.103904","url":null,"abstract":"<div><div>Recent Intrusion Detection (ID) networks face difficulties in handling the enlarging volume of network traffic and adapting to emerging cyber threats. Handling data traffic and addressing data imbalance are key requirements for identifying recent cyber threats. This paper proposes a novel hybrid ID system designed to mitigate data imbalance issues. The proposed methodology uses advanced deep learning techniques and optimized characteristic fusion models, making it suitable for high-traffic environments. This research conducts a comprehensive experimental study on five standard ID datasets, focusing on network traffic and system behavior data, which are crucial for detecting potential intrusions. In the deep feature extraction phase, multiple features are considered, including statistical information, T-SNE features, and high-level deep learning features. T-SNE features capture similarities between data points, helping preserve the most important features. For feature fusion, optimal weights are identified using the proposed RCMPA. The fused feature set, created from these optimized weights, improves that a more appropriate and discriminative characteristics are utilized for training. A system then employs a “Multi-scale Dilated Deep Hybrid Network with Attention Mechanism” (MDDHN-AM) for intrusion diagnosis. This developed model integrates TCNN and RNN to detain both temporal and spatial dependencies. TCNN processes sequential information to identify temporal patterns, while RNN captures the dynamic nature of network traffic. The attention mechanism prioritizes the most significant features, enabling more accurate intrusion detection. At last, the presentation of MDDHN-AM was compared to traditional and state-of-the-art intrusion detection methods across multiple metrics. The developed model achieved 96.43% detection accuracy and 97.38% precision, illustrating its efficiency in handling diverse digital attacks and data imbalance. An improved performance over traditional methods highlights its potential as a robust solution for secure communication and protection against evolving digital attacks.</div></div>","PeriodicalId":48648,"journal":{"name":"Ain Shams Engineering Journal","volume":"17 1","pages":"Article 103904"},"PeriodicalIF":5.9,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736133","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