Diego A. Flores-Hernández, Luis R. Islas-Estrada, Sergio I. Palomino-Resendiz
Solar tracking systems allow an increase in the use of solar energy for its conversion with photovoltaic technology due to the alignment with the sun. However, there is a compromise between tracking accuracy and the energy required to perform the movement action. Consequently, the wear of the tracker components increases, reducing its useful lifetime and affecting the profitability of these systems. The present research develops a novel tracking strategy based on real-time measurements to increase the lifetime without reducing the energy productivity of the tracking systems. The proposed approach is verified experimentally by implementing the real-time decision-making algorithm and a conventional tracking algorithm in identical tracking systems under the same weather conditions. The proposed strategy reduces energy consumption by 14.18% due to the tracking action, maintaining a practically identical energy generation between both systems. The findings highlight a 53.33% reduction in the movements required for tracking and a 60.77% reduction in operation time, which translates into a 6.8-fold increase in the lifetime of the solar tracking system under the experimental conditions applied. The results are promising, so this research initiates and motivates the development of more complex models to increase the useful life of the tracking systems and their profitability and environmental impact concurrently.
{"title":"A Novel Tracking Strategy Based on Real-Time Monitoring to Increase the Lifetime of Dual-Axis Solar Tracking Systems","authors":"Diego A. Flores-Hernández, Luis R. Islas-Estrada, Sergio I. Palomino-Resendiz","doi":"10.3390/app14188281","DOIUrl":"https://doi.org/10.3390/app14188281","url":null,"abstract":"Solar tracking systems allow an increase in the use of solar energy for its conversion with photovoltaic technology due to the alignment with the sun. However, there is a compromise between tracking accuracy and the energy required to perform the movement action. Consequently, the wear of the tracker components increases, reducing its useful lifetime and affecting the profitability of these systems. The present research develops a novel tracking strategy based on real-time measurements to increase the lifetime without reducing the energy productivity of the tracking systems. The proposed approach is verified experimentally by implementing the real-time decision-making algorithm and a conventional tracking algorithm in identical tracking systems under the same weather conditions. The proposed strategy reduces energy consumption by 14.18% due to the tracking action, maintaining a practically identical energy generation between both systems. The findings highlight a 53.33% reduction in the movements required for tracking and a 60.77% reduction in operation time, which translates into a 6.8-fold increase in the lifetime of the solar tracking system under the experimental conditions applied. The results are promising, so this research initiates and motivates the development of more complex models to increase the useful life of the tracking systems and their profitability and environmental impact concurrently.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Gorakshnath Takalkar, Sayma Akhter, Rahul R. Bhosale
In this investigation, an effort was made to introduce Ag into the CeO2 fluorite crystal lattice to form Ce0.99Ag0.01O2-δ (ACRM) using an ammonium hydroxide-assisted co-precipitation method. The resulting powder obtained after the co-precipitation reaction, filtration, and drying was annealed at 800 °C in a muffle furnace to obtain crystalline ACRM. The phase composition and microstructure of the synthesized ACRM were analyzed using a powder X-ray diffractometer (PXRD) and a scanning electron microscope (SEM). The characterized ACRM powder was then subjected to multiple thermochemical thermal reduction (TR) and CO2 splitting (CDS) cycles using a high-temperature thermogravimetric analyzer (TGA). The TR step was conducted using Ar gas as an inert atmosphere, maintaining the temperature at 1400 °C for 60 min. Subsequently, the same powder was subjected to the CDS step by treating it with a gaseous mixture of 50% CO2 and Ar gas at 1000 °C for 30 min. ACRM displayed stable redox reactivity towards thermochemical CDS cycles by generating an average of 50.9 μmol of O2/g·cycle and 101.6 μmol of CO/g·cycle, respectively, over 10 thermochemical cycles.
本研究采用氢氧化铵辅助共沉淀法，努力将银引入 CeO2 萤石晶格，形成 Ce0.99Ag0.01O2-δ（ACRM）。共沉淀反应、过滤和干燥后得到的粉末在马弗炉中于 800 °C 下退火，得到结晶的 ACRM。使用粉末 X 射线衍射仪（PXRD）和扫描电子显微镜（SEM）分析了合成的 ACRM 的相组成和微观结构。然后使用高温热重分析仪（TGA）对表征后的 ACRM 粉末进行了多次热化学热还原（TR）和二氧化碳分裂（CDS）循环。热化学还原步骤使用氩气作为惰性气氛，将温度保持在 1400 ℃，持续 60 分钟。随后，同样的粉末在 1000 °C 下用 50% CO2 和 Ar 的混合气体处理 30 分钟，进行 CDS 步骤。ACRM 在热化学 CDS 循环中表现出稳定的氧化还原反应活性，在 10 个热化学循环中分别生成了平均 50.9 μmol 的 O2/g-cycle 和 101.6 μmol 的 CO/g-cycle。
{"title":"Co-Precipitation Synthesized Ag-Doped Ceria Redox Material (ACRM) for the Thermochemical Conversion of CO2 into Solar Fuels","authors":"Gorakshnath Takalkar, Sayma Akhter, Rahul R. Bhosale","doi":"10.3390/app14188272","DOIUrl":"https://doi.org/10.3390/app14188272","url":null,"abstract":"In this investigation, an effort was made to introduce Ag into the CeO2 fluorite crystal lattice to form Ce0.99Ag0.01O2-δ (ACRM) using an ammonium hydroxide-assisted co-precipitation method. The resulting powder obtained after the co-precipitation reaction, filtration, and drying was annealed at 800 °C in a muffle furnace to obtain crystalline ACRM. The phase composition and microstructure of the synthesized ACRM were analyzed using a powder X-ray diffractometer (PXRD) and a scanning electron microscope (SEM). The characterized ACRM powder was then subjected to multiple thermochemical thermal reduction (TR) and CO2 splitting (CDS) cycles using a high-temperature thermogravimetric analyzer (TGA). The TR step was conducted using Ar gas as an inert atmosphere, maintaining the temperature at 1400 °C for 60 min. Subsequently, the same powder was subjected to the CDS step by treating it with a gaseous mixture of 50% CO2 and Ar gas at 1000 °C for 30 min. ACRM displayed stable redox reactivity towards thermochemical CDS cycles by generating an average of 50.9 μmol of O2/g·cycle and 101.6 μmol of CO/g·cycle, respectively, over 10 thermochemical cycles.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"145 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fruit quality assessment is vital for ensuring consumer satisfaction and marketability in agriculture. This study explores deep learning techniques for assessing fruit quality, focusing on practical deployment in resource-constrained environments. Two approaches were compared: training a convolutional neural network (CNN) from scratch and fine-tuning a pre-trained MobileNetV2 model through transfer learning. The performance of these models was evaluated using a subset of the Fruits-360 dataset chosen to simulate real-world conditions for small-scale producers. MobileNetV2 was selected for its compact size and efficiency, suitable for devices with limited computational resources. Both approaches achieved high accuracy, with the transfer learning model demonstrating faster convergence and slightly better performance. Feature map visualizations provided insight into the model’s decision-making, highlighting damaged areas of fruits which enhances transparency and trust for end users. This study underscores the potential of deep learning models to modernize fruit quality assessment, offering practical, efficient, and interpretable tools for small-scale farmers.
{"title":"Simplified Deep Learning for Accessible Fruit Quality Assessment in Small Agricultural Operations","authors":"Víctor Zárate, Danilo Cáceres Hernández","doi":"10.3390/app14188243","DOIUrl":"https://doi.org/10.3390/app14188243","url":null,"abstract":"Fruit quality assessment is vital for ensuring consumer satisfaction and marketability in agriculture. This study explores deep learning techniques for assessing fruit quality, focusing on practical deployment in resource-constrained environments. Two approaches were compared: training a convolutional neural network (CNN) from scratch and fine-tuning a pre-trained MobileNetV2 model through transfer learning. The performance of these models was evaluated using a subset of the Fruits-360 dataset chosen to simulate real-world conditions for small-scale producers. MobileNetV2 was selected for its compact size and efficiency, suitable for devices with limited computational resources. Both approaches achieved high accuracy, with the transfer learning model demonstrating faster convergence and slightly better performance. Feature map visualizations provided insight into the model’s decision-making, highlighting damaged areas of fruits which enhances transparency and trust for end users. This study underscores the potential of deep learning models to modernize fruit quality assessment, offering practical, efficient, and interpretable tools for small-scale farmers.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To address the problem of incomplete perception of limited viewpoints of ore blocks in future remote and intelligent shoveling-dominated mining scenarios, a method of using new view generation technology to predict ore blocks with limited view based on a latent diffusion model is proposed. Initially, an ore block image-pose dataset is created. Then, based on prior knowledge, the latent diffusion model undergoes transfer learning to develop an intelligent ore block shape prediction model (IOBSPM) for rock blocks. During training, structural similarity loss is innovatively introduced to constrain the prediction results and solve the issue of discontinuity in generated images. Finally, neural surface reconstruction is performed using the generated multi-view images of rock blocks to obtain a 3D model. Experimental results show that the prediction model, trained on the rock block dataset, produces better morphological and detail generation compared to the original model, with single-view generation time within 5 s. The average PSNR, SSIM, and LPIPS values reach 23.02 dB, 0.754, and 0.268, respectively. The generated views also demonstrate good performance in 3D reconstruction, highlighting significant implications for future research on remote and autonomous shoveling.
{"title":"Intelligent Prediction of Ore Block Shapes Based on Novel View Synthesis Technology","authors":"Lin Bi, Dewei Bai, Boxun Chen","doi":"10.3390/app14188273","DOIUrl":"https://doi.org/10.3390/app14188273","url":null,"abstract":"To address the problem of incomplete perception of limited viewpoints of ore blocks in future remote and intelligent shoveling-dominated mining scenarios, a method of using new view generation technology to predict ore blocks with limited view based on a latent diffusion model is proposed. Initially, an ore block image-pose dataset is created. Then, based on prior knowledge, the latent diffusion model undergoes transfer learning to develop an intelligent ore block shape prediction model (IOBSPM) for rock blocks. During training, structural similarity loss is innovatively introduced to constrain the prediction results and solve the issue of discontinuity in generated images. Finally, neural surface reconstruction is performed using the generated multi-view images of rock blocks to obtain a 3D model. Experimental results show that the prediction model, trained on the rock block dataset, produces better morphological and detail generation compared to the original model, with single-view generation time within 5 s. The average PSNR, SSIM, and LPIPS values reach 23.02 dB, 0.754, and 0.268, respectively. The generated views also demonstrate good performance in 3D reconstruction, highlighting significant implications for future research on remote and autonomous shoveling.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181124","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This study presents a numerical simulation approach used to investigate seismic-wave propagation in specific geological structures. Using the LS-DYNA software, the simulation incorporated a TNT explosion model to simulate seismic energy released during earthquakes. It provides a new method to investigate the propagation characteristics of seismic-waves within geological structures. Firstly, the measurement conditions and geological settings of the seismic event on 18 February 2012 in Northeast China are presented. Subsequently, a numerical simulation model of seismic-wave propagation is developed. The simulation result validates it by comparing it with recorded data from seismic stations, demonstrating a promising correspondence between the simulated and observed data. Additionally, the simulation simulates the seismic-wave propagation within water and layered geological structures, validating the numerical simulation model. The numerical model is an effective tool for simulating the propagation of seismic waves in geological structures. This study is important for evaluating seismic-wave propagation using the simulation method.
{"title":"Numerical Simulation of Seismic-Wave Propagation in Specific Layered Geological Structures","authors":"Chunyue Hao, Zhoupeng Gu, Kai Li, Xianqian Wu","doi":"10.3390/app14188278","DOIUrl":"https://doi.org/10.3390/app14188278","url":null,"abstract":"This study presents a numerical simulation approach used to investigate seismic-wave propagation in specific geological structures. Using the LS-DYNA software, the simulation incorporated a TNT explosion model to simulate seismic energy released during earthquakes. It provides a new method to investigate the propagation characteristics of seismic-waves within geological structures. Firstly, the measurement conditions and geological settings of the seismic event on 18 February 2012 in Northeast China are presented. Subsequently, a numerical simulation model of seismic-wave propagation is developed. The simulation result validates it by comparing it with recorded data from seismic stations, demonstrating a promising correspondence between the simulated and observed data. Additionally, the simulation simulates the seismic-wave propagation within water and layered geological structures, validating the numerical simulation model. The numerical model is an effective tool for simulating the propagation of seismic waves in geological structures. This study is important for evaluating seismic-wave propagation using the simulation method.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"115 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Vibration frequency and vibration duration are disturbance factors for the structural properties of clay. This study investigates how the vibration frequency and vibration duration from construction disturbances affect the mechanical properties of Zhanjiang Formation structural clay. An electric, frequency-adjustable vibration table applied varying frequencies and durations of vibration to undisturbed soil, creating structural clay samples with different disturbance degrees. Unconfined compressive strength tests and one-dimensional consolidation compression tests were conducted on these samples to obtain disturbance degrees RDq and RDS, defined by strength loss values and compression deformation characteristics, respectively. Orthogonal experiments and grey correlation analysis were used to explore the effects of vibration frequency and vibration duration on the mechanical properties of Zhanjiang Formation structural clay. The results indicated that disturbance degrees RDq and RDS increased linearly with increase in vibration frequency and vibration duration. Range analysis was conducted using two-factor three-level orthogonal experiment of disturbance degrees, and a grey relational analysis model was established to determine the primary and secondary effects of vibration duration and vibration frequency on the mechanical properties of Zhanjiang Formation structural clay. The results demonstrated that the findings from orthogonal experiments and grey relational analysis were consistent, showing that vibration duration had a more significant impact than vibration frequency on the mechanical properties of structural clay. The conclusion suggests that vibration disturbance manifests as a “fatigue damage effect”. Continuous vibration disturbance progressively weakens the cementation bonds between soil particles due to “accumulated” energy, leading to gradual fracture and destruction. With constant vibration frequency, longer durations, or with constant duration, higher frequencies intensify the “fatigue damage effect” of vibration disturbance. Furthermore, during vibration disturbance, Zhanjiang Formation structural clay shows a more pronounced “fatigue damage effect” from vibration duration than from vibration frequency, with cementation bonds between soil particles weakening more effectively due to “accumulated” energy. The research findings enhance the understanding of how vibration frequency and vibration duration from disturbance sources impact the mechanical properties of Zhanjiang Formation structural clay, offer theoretical guidance for using construction vibration machinery, and provide a reference for preventing and controlling soil disturbance.
{"title":"The Influence of Vibration Frequency and Vibration Duration on the Mechanical Properties of Zhanjiang Formation Structural Clay","authors":"Yanhua Xie, Bin Tang, Yansong Shi, Shuaiyu Liu, Jiankun Hu, Binghui Zhang","doi":"10.3390/app14188262","DOIUrl":"https://doi.org/10.3390/app14188262","url":null,"abstract":"Vibration frequency and vibration duration are disturbance factors for the structural properties of clay. This study investigates how the vibration frequency and vibration duration from construction disturbances affect the mechanical properties of Zhanjiang Formation structural clay. An electric, frequency-adjustable vibration table applied varying frequencies and durations of vibration to undisturbed soil, creating structural clay samples with different disturbance degrees. Unconfined compressive strength tests and one-dimensional consolidation compression tests were conducted on these samples to obtain disturbance degrees RDq and RDS, defined by strength loss values and compression deformation characteristics, respectively. Orthogonal experiments and grey correlation analysis were used to explore the effects of vibration frequency and vibration duration on the mechanical properties of Zhanjiang Formation structural clay. The results indicated that disturbance degrees RDq and RDS increased linearly with increase in vibration frequency and vibration duration. Range analysis was conducted using two-factor three-level orthogonal experiment of disturbance degrees, and a grey relational analysis model was established to determine the primary and secondary effects of vibration duration and vibration frequency on the mechanical properties of Zhanjiang Formation structural clay. The results demonstrated that the findings from orthogonal experiments and grey relational analysis were consistent, showing that vibration duration had a more significant impact than vibration frequency on the mechanical properties of structural clay. The conclusion suggests that vibration disturbance manifests as a “fatigue damage effect”. Continuous vibration disturbance progressively weakens the cementation bonds between soil particles due to “accumulated” energy, leading to gradual fracture and destruction. With constant vibration frequency, longer durations, or with constant duration, higher frequencies intensify the “fatigue damage effect” of vibration disturbance. Furthermore, during vibration disturbance, Zhanjiang Formation structural clay shows a more pronounced “fatigue damage effect” from vibration duration than from vibration frequency, with cementation bonds between soil particles weakening more effectively due to “accumulated” energy. The research findings enhance the understanding of how vibration frequency and vibration duration from disturbance sources impact the mechanical properties of Zhanjiang Formation structural clay, offer theoretical guidance for using construction vibration machinery, and provide a reference for preventing and controlling soil disturbance.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"86 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The occurrence of flight risks and accidents is closely related to pilot workload. Effective detection of pilot workload has been a key research area in the aviation industry. However, traditional methods for detecting pilot workload have several shortcomings: firstly, the collection of metrics via contact-based devices can interfere with pilots; secondly, real-time detection of pilot workload is challenging, making it difficult to capture sudden increases in workload; thirdly, the detection accuracy of these models is limited; fourthly, the models lack cross-pilot generalization. To address these challenges, this study proposes a large language model, WorkloadGPT, which utilizes low-interference indicators: eye movement and seat pressure. Specifically, features are extracted in 10 s time windows and input into WorkloadGPT for classification into low, medium, and high workload categories. Additionally, this article presents the design of an appropriate text template to serialize the tabular feature dataset into natural language, incorporating individual difference prompts during instance construction to enhance cross-pilot generalization. Finally, the LoRA algorithm was used to fine-tune the pre-trained large language model ChatGLM3-6B, resulting in WorkloadGPT. During the training process of WorkloadGPT, the GAN-Ensemble algorithm was employed to augment the experimental raw data, constructing a realistic and robust extended dataset for model training. The results show that WorkloadGPT achieved a classification accuracy of 87.3%, with a cross-pilot standard deviation of only 2.1% and a response time of just 1.76 s, overall outperforming existing studies in terms of accuracy, real-time performance, and cross-pilot generalization capability, thereby providing a solid foundation for enhancing flight safety.
{"title":"WorkloadGPT: A Large Language Model Approach to Real-Time Detection of Pilot Workload","authors":"Yijing Gao, Lishengsa Yue, Jiahang Sun, Xiaonian Shan, Yihan Liu, Xuerui Wu","doi":"10.3390/app14188274","DOIUrl":"https://doi.org/10.3390/app14188274","url":null,"abstract":"The occurrence of flight risks and accidents is closely related to pilot workload. Effective detection of pilot workload has been a key research area in the aviation industry. However, traditional methods for detecting pilot workload have several shortcomings: firstly, the collection of metrics via contact-based devices can interfere with pilots; secondly, real-time detection of pilot workload is challenging, making it difficult to capture sudden increases in workload; thirdly, the detection accuracy of these models is limited; fourthly, the models lack cross-pilot generalization. To address these challenges, this study proposes a large language model, WorkloadGPT, which utilizes low-interference indicators: eye movement and seat pressure. Specifically, features are extracted in 10 s time windows and input into WorkloadGPT for classification into low, medium, and high workload categories. Additionally, this article presents the design of an appropriate text template to serialize the tabular feature dataset into natural language, incorporating individual difference prompts during instance construction to enhance cross-pilot generalization. Finally, the LoRA algorithm was used to fine-tune the pre-trained large language model ChatGLM3-6B, resulting in WorkloadGPT. During the training process of WorkloadGPT, the GAN-Ensemble algorithm was employed to augment the experimental raw data, constructing a realistic and robust extended dataset for model training. The results show that WorkloadGPT achieved a classification accuracy of 87.3%, with a cross-pilot standard deviation of only 2.1% and a response time of just 1.76 s, overall outperforming existing studies in terms of accuracy, real-time performance, and cross-pilot generalization capability, thereby providing a solid foundation for enhancing flight safety.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"21 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
George Papageorgiou, Vangelis Sarlis, Manolis Maragoudakis, Christos Tjortjis
Integrating Large Language Models (LLMs) into e-government applications has the potential to improve public service delivery through advanced data processing and automation. This paper explores critical aspects of a modular and reproducible architecture based on Retrieval-Augmented Generation (RAG) for deploying LLM-based assistants within e-government systems. By examining current practices and challenges, we propose a framework ensuring that Artificial Intelligence (AI) systems are modular and reproducible, essential for maintaining scalability, transparency, and ethical standards. Our approach utilizing Haystack demonstrates a complete multi-agent Generative AI (GAI) virtual assistant that facilitates scalability and reproducibility by allowing individual components to be independently scaled. This research focuses on a comprehensive review of the existing literature and presents case study examples to demonstrate how such an architecture can enhance public service operations. This framework provides a valuable case study for researchers, policymakers, and practitioners interested in exploring the integration of advanced computational linguistics and LLMs into e-government services, although it could benefit from further empirical validation.
{"title":"Enhancing E-Government Services through State-of-the-Art, Modular, and Reproducible Architecture over Large Language Models","authors":"George Papageorgiou, Vangelis Sarlis, Manolis Maragoudakis, Christos Tjortjis","doi":"10.3390/app14188259","DOIUrl":"https://doi.org/10.3390/app14188259","url":null,"abstract":"Integrating Large Language Models (LLMs) into e-government applications has the potential to improve public service delivery through advanced data processing and automation. This paper explores critical aspects of a modular and reproducible architecture based on Retrieval-Augmented Generation (RAG) for deploying LLM-based assistants within e-government systems. By examining current practices and challenges, we propose a framework ensuring that Artificial Intelligence (AI) systems are modular and reproducible, essential for maintaining scalability, transparency, and ethical standards. Our approach utilizing Haystack demonstrates a complete multi-agent Generative AI (GAI) virtual assistant that facilitates scalability and reproducibility by allowing individual components to be independently scaled. This research focuses on a comprehensive review of the existing literature and presents case study examples to demonstrate how such an architecture can enhance public service operations. This framework provides a valuable case study for researchers, policymakers, and practitioners interested in exploring the integration of advanced computational linguistics and LLMs into e-government services, although it could benefit from further empirical validation.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A sensitivity analysis was conducted to evaluate several factors, including dam height, bank slope gradient, water storage times, and phased panel filling, on concrete-faced rockfill dams (CFRDs). The analysis identified the three most significant factors to examine their impacts on the stress-deformation characteristics of CFRDs. The results show that the order of influence on the dam body’s stress and deformation characteristics is as follows: dam height > bank slope gradient > water storage times > panel phased construction. From the perspective of stress-deformation of the face slab, water storage times predominantly affect tensile stress, while the bank slope gradient exerts the greatest influence on compressive stress. As the bank slope gradient decreases, the panel’s lateral restraint diminishes, leading to a decrease in the panel’s extrusion efficacy. Consequently, there are notable variations in the panel’s compressive stresses. An increase in dam height correlates with escalating stress and deformation in both the dam and face slab. As the bank slope gradient decreases, the deformation of the dam and face slab, as well as the range of tensile stress of the face slab, also increase. In contrast to a single water storage scenario, the face slab has experienced greater stress and deformation during the initial impoundment under multiple impoundment conditions. Therefore, multiple water storage schemes result in reduced deflection, axial horizontal displacement, and tensile stresses both along the slope and axial in the face slab. Furthermore, the tensile area at the bottom of the face slab transitions into a compressive area.
{"title":"Study on the Effects of Influence Factors on the Stress and Deformation Characteristics of Ultra-High CFRDs","authors":"Hongmei Li, Jianxin Wang, Yanyuan Lv, Chengming Feng","doi":"10.3390/app14188268","DOIUrl":"https://doi.org/10.3390/app14188268","url":null,"abstract":"A sensitivity analysis was conducted to evaluate several factors, including dam height, bank slope gradient, water storage times, and phased panel filling, on concrete-faced rockfill dams (CFRDs). The analysis identified the three most significant factors to examine their impacts on the stress-deformation characteristics of CFRDs. The results show that the order of influence on the dam body’s stress and deformation characteristics is as follows: dam height > bank slope gradient > water storage times > panel phased construction. From the perspective of stress-deformation of the face slab, water storage times predominantly affect tensile stress, while the bank slope gradient exerts the greatest influence on compressive stress. As the bank slope gradient decreases, the panel’s lateral restraint diminishes, leading to a decrease in the panel’s extrusion efficacy. Consequently, there are notable variations in the panel’s compressive stresses. An increase in dam height correlates with escalating stress and deformation in both the dam and face slab. As the bank slope gradient decreases, the deformation of the dam and face slab, as well as the range of tensile stress of the face slab, also increase. In contrast to a single water storage scenario, the face slab has experienced greater stress and deformation during the initial impoundment under multiple impoundment conditions. Therefore, multiple water storage schemes result in reduced deflection, axial horizontal displacement, and tensile stresses both along the slope and axial in the face slab. Furthermore, the tensile area at the bottom of the face slab transitions into a compressive area.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Image pre-processing is crucial for large fleet management. Many traffic videos are collected by closed-circuit television (CCTV), which has a fixed area monitoring for image analysis. This paper adopts the front camera installed in large vehicles to obtain moving traffic images, whereas CCTV is more limited. In practice, fleets often install cameras with different resolutions due to cost considerations. The cameras evaluate the front images with traffic lights. This paper proposes fuzzy enhancement with RGB and CIELAB conversions to handle multiple resolutions. This study provided image pre-processing adjustment comparisons, enabling further model training and analysis. This paper proposed fuzzy enhancement to deal with multiple resolutions. The fuzzy enhancement and fuzzy with brightness adjustment produced images with lower MSE and higher PSNR for the images of the front view. Fuzzy enhancement can also be used to enhance traffic light image adjustments. Moreover, this study employed You Only Look Once Version 9 (YOLOv9) for model training. YOLOv9 with fuzzy enhancement obtained better detection performance. This fuzzy enhancement made more flexible adjustments for pre-processing tasks and provided guidance for fleet managers to perform consistent image-enhancement adjustments for handling multiple resolutions.
图像预处理对于大型车队管理至关重要。许多交通视频都是通过闭路电视（CCTV）采集的，它有一个固定的区域监控来进行图像分析。本文采用安装在大型车辆上的前置摄像头来获取移动的交通图像，而闭路电视的局限性较大。在实际应用中，出于成本考虑，车队通常会安装不同分辨率的摄像头。这些摄像头通过交通信号灯对前方图像进行评估。本文提出使用 RGB 和 CIELAB 转换进行模糊增强，以处理多种分辨率。本研究提供了图像预处理调整比较，以便进一步进行模型训练和分析。本文提出了处理多种分辨率的模糊增强技术。模糊增强和带有亮度调整的模糊处理所生成的图像，正面图像的 MSE 更低，PSNR 更高。模糊增强还可用于交通灯图像的增强调整。此外，本研究还采用了 You Only Look Once Version 9（YOLOv9）进行模型训练。经过模糊增强的 YOLOv9 获得了更好的检测性能。这种模糊增强技术为预处理任务提供了更灵活的调整，并为车队管理人员在处理多种分辨率时执行一致的图像增强调整提供了指导。
{"title":"Enhancing the Image Pre-Processing for Large Fleets Based on a Fuzzy Approach to Handle Multiple Resolutions","authors":"Ching-Yun Mu, Pin Kung","doi":"10.3390/app14188254","DOIUrl":"https://doi.org/10.3390/app14188254","url":null,"abstract":"Image pre-processing is crucial for large fleet management. Many traffic videos are collected by closed-circuit television (CCTV), which has a fixed area monitoring for image analysis. This paper adopts the front camera installed in large vehicles to obtain moving traffic images, whereas CCTV is more limited. In practice, fleets often install cameras with different resolutions due to cost considerations. The cameras evaluate the front images with traffic lights. This paper proposes fuzzy enhancement with RGB and CIELAB conversions to handle multiple resolutions. This study provided image pre-processing adjustment comparisons, enabling further model training and analysis. This paper proposed fuzzy enhancement to deal with multiple resolutions. The fuzzy enhancement and fuzzy with brightness adjustment produced images with lower MSE and higher PSNR for the images of the front view. Fuzzy enhancement can also be used to enhance traffic light image adjustments. Moreover, this study employed You Only Look Once Version 9 (YOLOv9) for model training. YOLOv9 with fuzzy enhancement obtained better detection performance. This fuzzy enhancement made more flexible adjustments for pre-processing tasks and provided guidance for fleet managers to perform consistent image-enhancement adjustments for handling multiple resolutions.","PeriodicalId":8224,"journal":{"name":"Applied Sciences","volume":"54 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142181086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：