Elsevier eBooks, Journal Year: 2024, Volume and Issue: unknown, P. 27 - 37
Published: Nov. 22, 2024
Language: Английский
Environmental Modelling & Software, Journal Year: 2024, Volume and Issue: 174, P. 105971 - 105971
Published: Feb. 10, 2024
Language: Английский
Citations
84
Published: July 14, 2023
This research aims to develop a Machine Learning model for predicting soil moisture levels, which may be used construct smart irrigation systems. The was evaluated and trained using data from the "Smart Irrigation System Dataset" made publicly available by University of California, Irvine. A transfer-learned ResNet50 is various classification measures like accuracy, recall, precision, area under ROC curve (AUC). proposed has an AUC 0.95, meaning it correctly identifies positive negative samples 95% time. Moreover, model's performance measured against that other famous machine learning models logistic regression, Support Vector Machines (SVM), K-Nearest Neighbours (KNN), random forests, decision trees, naive Bayes, with majority these conventional being outperformed. These findings have ramifications researchers engineers creating intelligent systems precision agriculture.
Language: Английский
Citations
34
Sensors, Journal Year: 2023, Volume and Issue: 23(16), P. 7081 - 7081
Published: Aug. 10, 2023
Fresh dates have a limited shelf life and are susceptible to spoilage, which can lead economic losses for producers suppliers. The problem of accurate estimation fresh is essential various stakeholders involved in the production, supply, consumption dates. Modified atmosphere packaging (MAP) one methods that improves quality increases by reducing rate ripening. Therefore, this study aims apply fast cost-effective non-destructive techniques based on machine learning (ML) predict estimate stored date fruits under different conditions. Predicting estimating scheduling them at right time supply chain benefit from nutritional advantages observed physicochemical attributes fruits, including moisture content, total soluble solids, sugar tannin pH, firmness, during storage vacuum MAP 5 24 ∘C every 7 days determine using approach. TinyML-compatible regression models were employed stages fruit development period. decrease begins when they transition Khalal stage Rutab stage, ends start spoil or ripen Tamr stage. Low-cost Visible-Near-Infrared (VisNIR) spectral sensors (AS7265x-multi-spectral) used capture internal fruit. Regression estimation. findings indicated modified with 20% CO2 N balance efficiently increased 53 44 days, respectively, maintained ∘C. However, decreased 23 room temperature (24 ∘C). Edge Impulse supports training deployment low-cost microcontrollers, be real-time estimations TinyML sensors.
Language: Английский
Citations
29
Journal Of Big Data, Journal Year: 2024, Volume and Issue: 11(1)
Published: May 6, 2024
Abstract The United Nations’ 17 Sustainable Development Goals stress the importance of global and local efforts to address inequalities implement sustainability. Addressing complex, interconnected sustainability challenges requires a systematic, interdisciplinary approach, where technology, AI, data-driven methods offer potential solutions for optimizing resources, integrating different aspects sustainability, informed decision-making. Sustainability research surrounds various local, regional, challenges, emphasizing need identify emerging areas gaps AI models play crucial role. study performs comprehensive literature survey scientometric semantic analyses, categorizes problems, discusses sustainable use big data. outcomes analyses highlight collaborative inclusive that bridges regional differences, interconnection topics, major themes related It further emphasizes significance developing hybrid approaches combining techniques, expert knowledge multi-level, multi-dimensional Furthermore, recognizes necessity addressing ethical concerns ensuring data in research.
Language: Английский
Citations
16
Sensors, Journal Year: 2025, Volume and Issue: 25(4), P. 1199 - 1199
Published: Feb. 16, 2025
Climate change in Eastern Europe requires introducing automated irrigation systems and monitoring agricultural climatic parameters to ensure food security. The automation of irrigation, together with the generation climate reports based on AI (artificial intelligence) using OpenAI models for Internet Things (IoT) data processing, contributes optimization resources by reducing excessive water energy consumption, supporting plant health through proper increasing sustainable productivity providing suggestions statistics streamline process. In this paper, authors present a system that allows continuous collection such as temperature, humidity, soil moisture, detailed information advanced analytics each device area monitored generate predictive recommendations. transmission is performed wirelessly via WebSocket central database. This uses from all devices connected application assess current conditions at national level, identifying trends generating aid adapting extreme events. integration artificial intelligence context areas step forward development agriculture adaptation increasingly aggressive phenomena, replicable framework vulnerable regions.
Language: Английский
Citations
1
Sustainability, Journal Year: 2023, Volume and Issue: 15(17), P. 12871 - 12871
Published: Aug. 25, 2023
The safety and quality of fresh fruits deserve the greatest attention, are a priority for producers consumers alike. Modern technologies crucial to accurately estimating predicting fruits’ shelf life, optimize supply chain management. Modified atmosphere packaging (MAP) is an essential method that maintains parameters increases life by reducing their ripening rates. This study aimed develop cost-effective, non-destructive technique using tiny machine learning (TinyML) multispectral sensor predict/estimate packaged dates under natural (Control), vacuum-sealed bags (VSBs), MAP with different gas combinations: 20% CO2 + N balance (MAP1), 10% O2 (MAP2). (pH, total soluble solids (TSSs), sugar content (SC), moisture (MC), tannin (TC)) were evaluated storage temperatures times. A (AS7265x) was utilized correlate fruit spectrum analysis same conditions, prepare dataset train prediction models. models trained in Edge Impulse Platform, deployed Arduino Nano 33 BLE sense microcontroller unit (MCU) inference. findings indicated vacuum MAP1 efficiently increased maintained 43 ± 2.39 39 3.34 days, respectively, at 5 °C. optimal neural network consisted input layer 20 nodes (the type, temperature, 18 channels spectral data 410 940 nm wavelengths), two hidden layers 12 nodes, output one node target parameter or life. These predicted R2 = 0.951, pH 0.854, TSSs 0.893, SC 0.881, MC 0.941, TC 0.909. evaluation developed each condition these serve as powerful tools parameters, dates.
Language: Английский
Citations
14
Electronics, Journal Year: 2024, Volume and Issue: 13(22), P. 4362 - 4362
Published: Nov. 7, 2024
This article explores the transformative potential of artificial intelligence (AI) tools across agricultural value chain, highlighting their applications, benefits, challenges, and future prospects. With global food demand projected to increase by 70% 2050, AI technologies—including machine learning, big data analytics, Internet things (IoT)—offer critical solutions for enhancing productivity, sustainability, resource efficiency. The study provides a comprehensive review applications at multiple stages including land use planning, crop selection, management, disease detection, yield prediction, market integration. It also discusses significant challenges adoption, such as accessibility, technological infrastructure, need specialized skills. By examining case studies empirical evidence, demonstrates how AI-driven can optimize decision-making operational efficiency in agriculture. findings underscore AI’s pivotal role addressing with implications farmers, agribusinesses, policymakers, researchers. aims advance evolving research discussions on sustainable agriculture, contributing insights that promote adoption technologies influence farming.
Language: Английский
Citations
6
International Journal of Environmental Research, Journal Year: 2024, Volume and Issue: 19(1)
Published: Oct. 24, 2024
Abstract This study aims to explore the application of artificial intelligence (AI) in resolution sustainability challenges, with a specific focus on environmental studies. Given rapidly evolving nature this field, there is an urgent need for more frequent and dynamic reviews keep pace innovative applications AI. Through systematic analysis 191 research articles, we classified AI techniques applied field sustainability. Our review found that 65% studies supervised learning methods, 18% employed unsupervised learning, 17% utilized reinforcement approaches. The highlights neural networks (ANN), are most commonly contexts, accounting 23% reviewed methods. comprehensive overview identifies key trends proposes new avenues address complex issue achieving Sustainable Development Goals (SDGs). Graphic abstract
Language: Английский
Citations
5
Journal of Landscape Ecology, Journal Year: 2024, Volume and Issue: 17(1), P. 16 - 34
Published: March 6, 2024
Abstract Groundwater plays a vital role in global water resources, supporting agricultural, industrial, and domestic supply systems. However, the long-term sustainability of groundwater is increasingly threatened due to widespread adoption irrigation systems especially micro irrigation. Micro agricultural technique that involves application crops through drip sprinkler This method has gained its ability deliver efficiently crops. review paper examines impacts on sustainability, focusing effects quantity, quality, overall sustainability. The findings reveal can significantly contribute conservation by reducing losses. improper management practices, such as over-irrigation or incorrect rates, lead excessive extraction, depletion aquifers, declining tables. Applying fertilizers pesticides may pollution, thereby affecting quality posing risk human health. article emphasizes significance appropriate design, installation, upkeep minimize potential adverse groundwater. Furthermore, regulatory frameworks, policies, educational programs are crucial promoting sustainable practices present highlights adopting balanced use enhancing techniques, implementing relevant regulations ensure utilization resources
Language: Английский
Citations
4
Sustainability, Journal Year: 2023, Volume and Issue: 15(8), P. 6911 - 6911
Published: April 19, 2023
Display refrigerators consume significantly high energy, and improving their efficiency is essential to minimize energy consumption greenhouse gas emissions. Therefore, providing the refrigeration system with a reliable energy-efficient mechanism real challenge. This study aims design evaluate an intelligent control (ICS) using artificial neural networks (ANN) for performance optimization of solar-powered display (SPDRs). The SPDR was operated traditional at fixed frequency 60 Hz then based on variable frequencies ranging from 40 designed ANN-based ICS combined speed drive. A stand-alone PV provided refrigerator required two options. For evaluation, operating conditions after modification its were compared (TCS) target temperatures 1, 3, 5 °C ambient 23, 29, 35 °C. Based controlled by modified (MCS), power, consumption, coefficient (COP) are improved. results show that both mechanisms maintain same cooling temperature, but consumes more (p < 0.05). At increasing temperature decreased COP TCS MCS. average daily varied 2.8 3.83 1.91 2.82 MCS, respectively. comparison refrigerators’ indicated developed saved about 35.5% worked smoother power when high. MCS higher than 26.37%, 26.59%, 24.22% 23 °C, 29 optimized proved be suitable tool industry.
Language: Английский
Citations
10