Pub Date : 2024-10-07DOI: 10.1109/TLA.2024.10706025
María Laura Montoro;Maria Nadia Pantano;Cecilia Fernandez;Maria Fabiana Sardella;Gustavo Scaglia
This article proposes a novel model for the extraction of pectin in melon peels and seeds. The methodology is based on the extraction of pectin in an acid medium for 180 minutes at 70, 80, and 90C, evaluating the performance of the product at each temperature. The kinetics of pectin extraction from melon peels and seeds, regardless of the working temperature, presents three stages: rapid release, then a plateau, followed by smooth growth until reaching the maximum amount of the product extracted. This process can be assimilated as a sequence of subprocesses, each with its own delay time and constants time. Based on the experimental results, each stage or period is mathematically modeled as a second-order linear with delay time. This dynamic model takes into account the work matrix, as well as the extraction mechanism used. The deviation of the model concerning to the experimental data is minimal, compared to the empirical and mechanistic models found in the literature for pectin extraction. The latter are based on oversimplified assumptions, leading to significant disparities between experimentally obtained and mathematically simulated results.
{"title":"Mathematical Modeling for Pectin Extraction in Melon waste","authors":"María Laura Montoro;Maria Nadia Pantano;Cecilia Fernandez;Maria Fabiana Sardella;Gustavo Scaglia","doi":"10.1109/TLA.2024.10706025","DOIUrl":"https://doi.org/10.1109/TLA.2024.10706025","url":null,"abstract":"This article proposes a novel model for the extraction of pectin in melon peels and seeds. The methodology is based on the extraction of pectin in an acid medium for 180 minutes at 70, 80, and 90C, evaluating the performance of the product at each temperature. The kinetics of pectin extraction from melon peels and seeds, regardless of the working temperature, presents three stages: rapid release, then a plateau, followed by smooth growth until reaching the maximum amount of the product extracted. This process can be assimilated as a sequence of subprocesses, each with its own delay time and constants time. Based on the experimental results, each stage or period is mathematically modeled as a second-order linear with delay time. This dynamic model takes into account the work matrix, as well as the extraction mechanism used. The deviation of the model concerning to the experimental data is minimal, compared to the empirical and mechanistic models found in the literature for pectin extraction. The latter are based on oversimplified assumptions, leading to significant disparities between experimentally obtained and mathematically simulated results.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10706025","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142397507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, a reliability analysis to determinate the most preponderant negative effect between the amplitude and intensity of Extreme Operating Gust (EOG) in back-to-back (BTB) power converter connected to PMSG-based wind turbine is proposed. For this, a 42 factorial design is proposed to analyze the impact of amplitude and intensity of the EOG gust on the electrical variables measured at the BTB power converter such as the current, voltage and active power. Note that for this analysis the La Ventosa wind database allocated in Oaxaca, Mexico is considered. The simulation of this system was developed using the capabilities of the PSIM software. Finally, the results of the reliability analysis are presented, determining the factor with the greatest impact on the reliability of the BTB power converter.
{"title":"Impact of the Extreme Operating Gusts on Power Converter Connected to PMSG-based Wind Turbine for Reliability Analysis","authors":"Gregorio Martínez Reyes;Emmanuel Hernández Mayoral;Efraín Dueñas Reyes;Reynaldo Iracheta Cortez;José Rafael Dorrego Portela","doi":"10.1109/TLA.2024.10705994","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705994","url":null,"abstract":"In this paper, a reliability analysis to determinate the most preponderant negative effect between the amplitude and intensity of Extreme Operating Gust (EOG) in back-to-back (BTB) power converter connected to PMSG-based wind turbine is proposed. For this, a 42 factorial design is proposed to analyze the impact of amplitude and intensity of the EOG gust on the electrical variables measured at the BTB power converter such as the current, voltage and active power. Note that for this analysis the La Ventosa wind database allocated in Oaxaca, Mexico is considered. The simulation of this system was developed using the capabilities of the PSIM software. Finally, the results of the reliability analysis are presented, determining the factor with the greatest impact on the reliability of the BTB power converter.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705994","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-04DOI: 10.1109/TLA.2024.10705972
Jose Yael Lopez Hernandez;Enrique Gonzalez;Raul Pena;Antonio Carlos Bento;Sergio Camacho-Leon
In IoT-based supply chain transportation, low rates for data loss, minimizing time to destination, and optimizing energy consumption are paramount. These factors can be influenced by variable parameters, data volume, logging procedures, positioning complexities, and communication hiccups during transit. This study introduces an adaptive data logging algorithm for a cost-effective IoT node, addressing these challenges. This innovation enables real-time data acquisition and remote display via a web interface. Experimental tests demonstrate the prototype's reliability in both controlled indoor and dynamic outdoor environments, particularly in environmental and GPS data collection. Results reveal 5.24% data loss indoors and 23.24% via the web interface. Outdoors, data loss peaks at 55.34%, increasing to 82.76% with the web interface. However, the obtained information is adequate for prototype validation. The algorithm reduces data by 74%, leading to lower data processing and power transmission needs. Moreover, determining the distance from GPS coordinates is essential for predicting travel times and monitoring vehicle velocity to maximize efficiency. The results from this prototype are expected to enhance the development of advanced models, thus enriching future scientific research initiatives that aim to incorporate IoT technology into transportation systems.
{"title":"Implementation of an adaptive data logging algorithm in low-cost IoT nodes for supply chain transport monitoring","authors":"Jose Yael Lopez Hernandez;Enrique Gonzalez;Raul Pena;Antonio Carlos Bento;Sergio Camacho-Leon","doi":"10.1109/TLA.2024.10705972","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705972","url":null,"abstract":"In IoT-based supply chain transportation, low rates for data loss, minimizing time to destination, and optimizing energy consumption are paramount. These factors can be influenced by variable parameters, data volume, logging procedures, positioning complexities, and communication hiccups during transit. This study introduces an adaptive data logging algorithm for a cost-effective IoT node, addressing these challenges. This innovation enables real-time data acquisition and remote display via a web interface. Experimental tests demonstrate the prototype's reliability in both controlled indoor and dynamic outdoor environments, particularly in environmental and GPS data collection. Results reveal 5.24% data loss indoors and 23.24% via the web interface. Outdoors, data loss peaks at 55.34%, increasing to 82.76% with the web interface. However, the obtained information is adequate for prototype validation. The algorithm reduces data by 74%, leading to lower data processing and power transmission needs. Moreover, determining the distance from GPS coordinates is essential for predicting travel times and monitoring vehicle velocity to maximize efficiency. The results from this prototype are expected to enhance the development of advanced models, thus enriching future scientific research initiatives that aim to incorporate IoT technology into transportation systems.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705972","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142377148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-04DOI: 10.1109/TLA.2024.10705971
Ricardo C. Camara de M. Santos;Mateus Coelho;Ricardo Oliveira
Real-time object detection in images is one of the most important areas in computer vision and finds applications in several fields, such as security systems, protection, independent vehicles, and robotics. Many of these applications need to use edge hardware platforms, and it is vital to know the performance of the object detector on these hardware platforms before developing the system. Therefore, in this work, we executed performance benchmark tests of the YOLOv7-tiny model for real-time object detection using a camera and three embedded hardware platforms: Raspberry Pi 4B, Jetson Nano, and Jetson Xavier NX. We tested and analyzed the NVIDIA platforms and their different power modes. The Raspberry Pi 4B achieved an average of 0.9 FPS. The Jetson Xavier NX achieved 30 FPS, the maximum possible FPS rate, in three power modes. In the tests, it was possible to notice that the maximum CPU clock of the Jetson Xavier NX impacts the FPS rate more than the GPU clock itself. The Jetson Nano achieved 7.4 and 5.2 FPS in its two power consumption modes.
{"title":"Real-time Object Detection Performance Analysis Using YOLOv7 on Edge Devices","authors":"Ricardo C. Camara de M. Santos;Mateus Coelho;Ricardo Oliveira","doi":"10.1109/TLA.2024.10705971","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705971","url":null,"abstract":"Real-time object detection in images is one of the most important areas in computer vision and finds applications in several fields, such as security systems, protection, independent vehicles, and robotics. Many of these applications need to use edge hardware platforms, and it is vital to know the performance of the object detector on these hardware platforms before developing the system. Therefore, in this work, we executed performance benchmark tests of the YOLOv7-tiny model for real-time object detection using a camera and three embedded hardware platforms: Raspberry Pi 4B, Jetson Nano, and Jetson Xavier NX. We tested and analyzed the NVIDIA platforms and their different power modes. The Raspberry Pi 4B achieved an average of 0.9 FPS. The Jetson Xavier NX achieved 30 FPS, the maximum possible FPS rate, in three power modes. In the tests, it was possible to notice that the maximum CPU clock of the Jetson Xavier NX impacts the FPS rate more than the GPU clock itself. The Jetson Nano achieved 7.4 and 5.2 FPS in its two power consumption modes.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705971","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376789","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-04DOI: 10.1109/TLA.2024.10705973
Nestor Gonzalez-Cabrera;Daniel Ernesto Hernandez Reyes;Vicente Torres García
Transmission expansion planning aims to establish when and where to install new infrastructure such as transmission lines, cables, generators and transformers in the electrical power system. The planning must be motivated mainly to satisfy the increase in demand, consequently increase the reliability of the system and provide non-discriminatory access for generators and consumers to the electrical grid. In this sense, this work aims to propose a methodology to handle demand uncertainty by reducing scenarios through the K-means clustering algorithm, which is used to construct representative demand curves that allow using a static model of stochastic linear optimization with less computational effort, which seeks to minimize the investment and operating costs of the electrical system, meeting the total demand of the system. The global demand and nodal demand approach of the system is compared, observing the behaviour of investment and operating costs, as well as their advantages. The results demonstrate that the formulation can be estimate the number of scenarios through mathematical metrics and the global demand approach has the advantage of only needing data on the behavior of the total demand of the system.
{"title":"Transmission Network Expansion Planning Considering Uncertainty in Demand with Global and Nodal Approach","authors":"Nestor Gonzalez-Cabrera;Daniel Ernesto Hernandez Reyes;Vicente Torres García","doi":"10.1109/TLA.2024.10705973","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705973","url":null,"abstract":"Transmission expansion planning aims to establish when and where to install new infrastructure such as transmission lines, cables, generators and transformers in the electrical power system. The planning must be motivated mainly to satisfy the increase in demand, consequently increase the reliability of the system and provide non-discriminatory access for generators and consumers to the electrical grid. In this sense, this work aims to propose a methodology to handle demand uncertainty by reducing scenarios through the K-means clustering algorithm, which is used to construct representative demand curves that allow using a static model of stochastic linear optimization with less computational effort, which seeks to minimize the investment and operating costs of the electrical system, meeting the total demand of the system. The global demand and nodal demand approach of the system is compared, observing the behaviour of investment and operating costs, as well as their advantages. The results demonstrate that the formulation can be estimate the number of scenarios through mathematical metrics and the global demand approach has the advantage of only needing data on the behavior of the total demand of the system.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705973","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-04DOI: 10.1109/TLA.2024.10705968
{"title":"Table of Contents October 2024","authors":"","doi":"10.1109/TLA.2024.10705968","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705968","url":null,"abstract":"","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705968","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The use of machine learning in healthcare has transformed the way diseases are diagnosed and treatments are optimized. However, medical databases often lack balanced data due to challenges in data collection caused by privacy regulations. Certain health conditions are under represented, which hampers machine learning performance. To address this problem, a hybrid approach has been proposed that combines the Synthetic Minority Oversampling Technique (SMOTE) with under sampling and uses two specific techniques tailored for imbalanced datasets. Comparative evaluations were conducted using various thresholds to reduce one class and employingBalanced Accuracy to mitigate bias toward the majority class, with popular machine learning methods. The results showed that Balanced Bagging and Balanced Random Forest consistently outperformed other methods, performing the best with an average ranking of 1.42 and 3.58 out of 32 configurations in the two datasets, respectively. Tree-based approaches such as Random Forest and Gradient Boosting demonstrated similar effectiveness, emphasizing the power of aggregating predictions from multiple trees to reduce bias. Notably, under sampling andSMOTE proved advantageous for non-tree-based models likeKNN, SVM, and Logistic Regression showcasing their usefulness across different algorithms. This study provides a robust solution for handling imbalanced datasets in healthcare, which could potentially optimize healthcare interventions and improve patient outcomes and care.
{"title":"Addressing Class Imbalance in Healthcare Data: Machine Learning Solutions for Age-Related Macular Degeneration and Preeclampsia","authors":"Antonieta Martinez-Velasco;Lourdes Martínez -Villaseñor;Luis Miralles-Pechuán","doi":"10.1109/TLA.2024.10705995","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705995","url":null,"abstract":"The use of machine learning in healthcare has transformed the way diseases are diagnosed and treatments are optimized. However, medical databases often lack balanced data due to challenges in data collection caused by privacy regulations. Certain health conditions are under represented, which hampers machine learning performance. To address this problem, a hybrid approach has been proposed that combines the Synthetic Minority Oversampling Technique (SMOTE) with under sampling and uses two specific techniques tailored for imbalanced datasets. Comparative evaluations were conducted using various thresholds to reduce one class and employingBalanced Accuracy to mitigate bias toward the majority class, with popular machine learning methods. The results showed that Balanced Bagging and Balanced Random Forest consistently outperformed other methods, performing the best with an average ranking of 1.42 and 3.58 out of 32 configurations in the two datasets, respectively. Tree-based approaches such as Random Forest and Gradient Boosting demonstrated similar effectiveness, emphasizing the power of aggregating predictions from multiple trees to reduce bias. Notably, under sampling andSMOTE proved advantageous for non-tree-based models likeKNN, SVM, and Logistic Regression showcasing their usefulness across different algorithms. This study provides a robust solution for handling imbalanced datasets in healthcare, which could potentially optimize healthcare interventions and improve patient outcomes and care.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705995","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376985","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-04DOI: 10.1109/TLA.2024.10705967
Xiaoyou Yu;Tianchu Li;Ziyun Tian;Miao Yu
We propose a novel deep learning (DL) based HBF design for the dual-functional radar-communication (DFRC) system with the millimeter wave (mmWave) massive multiple-in-multiple-output (MIMO) architecture, in which the HBF is formulated as a non-convex optimization problem. First, the DL-based HBF is designed to minimize the sum-MSE of downlink communications while carrying out necessary radar sensing concurrently. Then the synchronization noise is attached to the input channel data to enhance the robustness of the CNN. After that, an attention mechanism is added into the prediction stage to improve the prediction without affecting the accuracy of the prediction results. Finally, the numerical simulation results show significant tradeoff performance improvements between communication and radar sensing can be obtained over existing HBF designs.
{"title":"Deep Learning Based Hybrid Beamforming for mmWave Dual-Functional Radar-Communication","authors":"Xiaoyou Yu;Tianchu Li;Ziyun Tian;Miao Yu","doi":"10.1109/TLA.2024.10705967","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705967","url":null,"abstract":"We propose a novel deep learning (DL) based HBF design for the dual-functional radar-communication (DFRC) system with the millimeter wave (mmWave) massive multiple-in-multiple-output (MIMO) architecture, in which the HBF is formulated as a non-convex optimization problem. First, the DL-based HBF is designed to minimize the sum-MSE of downlink communications while carrying out necessary radar sensing concurrently. Then the synchronization noise is attached to the input channel data to enhance the robustness of the CNN. After that, an attention mechanism is added into the prediction stage to improve the prediction without affecting the accuracy of the prediction results. Finally, the numerical simulation results show significant tradeoff performance improvements between communication and radar sensing can be obtained over existing HBF designs.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705967","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-04DOI: 10.1109/TLA.2024.10705969
Baltazar López-Velasco;Agustin Ruiz-Garcia;José Guillermo Cebada-Reyes;Carlos Alberto Villaseñor-Perea
Modernizing the drying process will reduce agricultural product waste and environmental pollution. The aim of this study was to design a monitoring system based on the internet of things (IoT), temperature and relative humidity for a solar dryer. This system consists of a data collection module that gathers data regarding temperature (Ta), external relative humidity (RH) and on/off time of the solar dryer exhaust fans; a communication module that transmits Ta and RH information via LoRa and Wifi to ThingSpeak for monitoring on a mobile device; and a power module providing electrical power for system operation by solar energy. The operation of the IoT monitoring system was evaluated in three drying experiments of Dominican bananas (Musa paradisiaca var sapientum), in which system performance was satisfactory, allowing the user to visualize graphically in a web and mobile interface the behavior of Ta and RH inside the dryer. The data measured by the system were used to predict banana moisture content with an autoregressive model with exogenous variables (ARX) identified online. The mathematical model found predicted the behavior of moisture content with a good goodness of fit, with values of R2 = 0.99, MSE = 1.2910-5 and MAE = -5.0310-6. The solar dryer allowed reducing the moisture content of bananas up to 19.84 % wet basis (w.b.) in a period of 4 days and by 20.03% w.b. for 5 days in the presence of rainfall.
干燥过程的现代化将减少农产品浪费和环境污染。本研究的目的是为太阳能干燥机设计一个基于物联网(IoT)、温度和相对湿度的监控系统。该系统包括一个数据收集模块,用于收集有关温度(Ta)、外部相对湿度(RH)和太阳能烘干机排风扇开/关时间的数据;一个通信模块,用于通过 LoRa 和 Wifi 向 ThingSpeak 传输 Ta 和 RH 信息,以便在移动设备上进行监控;以及一个电源模块,通过太阳能为系统运行提供电力。在对多米尼加香蕉(Musa paradisiaca var sapientum)进行的三次干燥实验中,对物联网监控系统的运行情况进行了评估,系统性能令人满意,用户可以在网络和移动界面上以图形方式直观地看到干燥机内的温度和相对湿度的变化情况。系统测量到的数据被用来预测香蕉的水分含量,该预测是通过在线识别的外生变量自回归模型(ARX)得出的。所发现的数学模型能很好地预测水分含量的变化,拟合度为 R2 = 0.99,MSE = 1.2910-5 和 MAE = -5.0310-6。太阳能干燥机可在 4 天内将香蕉的湿基含水量降低到 19.84%,在降雨的情况下,5 天内可将湿基含水量降低 20.03%。
{"title":"IoT-based Environmental Monitoring and Prediction of Banana Moisture Content in a Solar Greenhouse Dryer","authors":"Baltazar López-Velasco;Agustin Ruiz-Garcia;José Guillermo Cebada-Reyes;Carlos Alberto Villaseñor-Perea","doi":"10.1109/TLA.2024.10705969","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705969","url":null,"abstract":"Modernizing the drying process will reduce agricultural product waste and environmental pollution. The aim of this study was to design a monitoring system based on the internet of things (IoT), temperature and relative humidity for a solar dryer. This system consists of a data collection module that gathers data regarding temperature (Ta), external relative humidity (RH) and on/off time of the solar dryer exhaust fans; a communication module that transmits Ta and RH information via LoRa and Wifi to ThingSpeak for monitoring on a mobile device; and a power module providing electrical power for system operation by solar energy. The operation of the IoT monitoring system was evaluated in three drying experiments of Dominican bananas (Musa paradisiaca var sapientum), in which system performance was satisfactory, allowing the user to visualize graphically in a web and mobile interface the behavior of Ta and RH inside the dryer. The data measured by the system were used to predict banana moisture content with an autoregressive model with exogenous variables (ARX) identified online. The mathematical model found predicted the behavior of moisture content with a good goodness of fit, with values of R2 = 0.99, MSE = 1.2910-5 and MAE = -5.0310-6. The solar dryer allowed reducing the moisture content of bananas up to 19.84 % wet basis (w.b.) in a period of 4 days and by 20.03% w.b. for 5 days in the presence of rainfall.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705969","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Point cloud videos capture the time-varying environment and are widely used for dynamic scene understanding. Existing methods develop effective networks for point cloud videos but do not fully utilize the prior information uncovered during pre-training. Furthermore, relying on a single supervised task with a large amount of manually labeled data may be insufficient to capture the foundational structures in point cloud videos. In this paper, we propose a pre-training framework Query-CP to learn the representations of point cloud videos through multiple self-supervised pretext tasks. First, tokenlevel contrast is developed to predict future features under the guidance of historical information. Using a position-guided autoregressor with learnable queries, the predictions are directly contrasted with corresponding targets in the high-level feature space to capture fine-grained semantics. Second, performing only contrastive learning fails to fully explore the complementary structures and dynamics information. To alleviate this, a decoupled spatio-temporal prediction task is designed, where we use a spatial branch to predict low-level features and a temporal branch to predict timestamps of the target sequence explicitly. By combining the above self-supervised tasks, multi-level information is captured during the pre-training stage. Finally, the encoder is fine-tuned and evaluated for action recognition and dynamic semantic segmentation on three datasets. The results demonstrate the effectiveness of our Query-CP. Especially, compared with the state-of-the-art methods, the fine-tuning accuracy on action recognition improves by 3.23% for 24-frame point cloud videos, and the mean accuracy increases by 4.21%.
{"title":"Learnable Query Contrast and Spatio-temporal Prediction on Point Cloud Video Pre-training","authors":"Xiaoxiao Sheng;Zhiqiang Shen;Longguang Wang;Gang Xiao","doi":"10.1109/TLA.2024.10705970","DOIUrl":"https://doi.org/10.1109/TLA.2024.10705970","url":null,"abstract":"Point cloud videos capture the time-varying environment and are widely used for dynamic scene understanding. Existing methods develop effective networks for point cloud videos but do not fully utilize the prior information uncovered during pre-training. Furthermore, relying on a single supervised task with a large amount of manually labeled data may be insufficient to capture the foundational structures in point cloud videos. In this paper, we propose a pre-training framework Query-CP to learn the representations of point cloud videos through multiple self-supervised pretext tasks. First, tokenlevel contrast is developed to predict future features under the guidance of historical information. Using a position-guided autoregressor with learnable queries, the predictions are directly contrasted with corresponding targets in the high-level feature space to capture fine-grained semantics. Second, performing only contrastive learning fails to fully explore the complementary structures and dynamics information. To alleviate this, a decoupled spatio-temporal prediction task is designed, where we use a spatial branch to predict low-level features and a temporal branch to predict timestamps of the target sequence explicitly. By combining the above self-supervised tasks, multi-level information is captured during the pre-training stage. Finally, the encoder is fine-tuned and evaluated for action recognition and dynamic semantic segmentation on three datasets. The results demonstrate the effectiveness of our Query-CP. Especially, compared with the state-of-the-art methods, the fine-tuning accuracy on action recognition improves by 3.23% for 24-frame point cloud videos, and the mean accuracy increases by 4.21%.","PeriodicalId":55024,"journal":{"name":"IEEE Latin America Transactions","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705970","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142376682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}