Tribological properties of CNT-filled epoxy-carbon fabric composites: Optimization and modelling by machine learning

Journal of Materials Research and Technology, Journal Year: 2023, Volume and Issue: 28, P. 2582 - 2601

Published: Dec. 24, 2023

Polymer matrix composites reinforced with fibers/fillers are extensively used in several tribological components of automotive and boating applications. The mechanical performance polymer improves by incorporating nanofillers as secondary reinforcement. present research work fabricated carbon fabric-reinforced epoxy using the hand layup. were 0.1 wt%, 0.2 0.5 wt% nanotubes (CNT) fillers Tribological properties filled CNT have been carried out a pin‐on‐disc method. Adding significantly behaviour reducing wear rate coefficient friction. large surface area interaction due to higher aspect ratio shows improved adhesion between fabrics. It various characteristics composites—also, an analysis worn surfaces is analyze mechanisms scanning electronic microscopy. employs combination experimental analyses machine learning (ML) techniques explore resistance, hardness, predictive modeling volume loss composites. hyperparameter fine-tuning ML algorithms, including Random Forest (RF), k-Nearest Neighbors (KNN), XGBoost, demonstrates superior capabilities, particularly RF. study bridges material science, ML, practical applications, contributing valuable insights for developing advanced composite materials.

Language: Английский

---

Towards Improving Sustainable Water Management in Geothermal Fields: SVM and RF Land Use Monitoring

Journal of Human Earth and Future, Journal Year: 2024, Volume and Issue: 5(2), P. 216 - 242

Published: June 1, 2024

The management and monitoring of land use in geothermal fields are crucial for the sustainable utilization water resources, as well striking a balance between production renewable energy preservation environment. This study primarily compared Support Vector Machine (SVM) Random Forest (RF) machine learning methods, using satellite imagery from Landsat 8 Sentinel 2 2021 2023, to monitor Patuha area. objective is improve practices by accurately categorizing different cover types. comparative analysis assessed efficacy these techniques upholding sustainability regions. examined application SVM RF techniques, with particular emphasis on parameter refinement model assessment, enhance classification accuracy. By employing Kernlab e1071 algorithm comparison, research sought produce precise Land Use Model Map, which underscores significance advanced analytical environmental management. approach was utmost importance improving reinforcing practices. evaluation methods demonstrates superiority terms accuracy, stability, precision, particularly intricate urban settings, hence establishing it preferred tasks demanding high reliability. areas alignment Sustainable Development Goals (SDGs) 6 15, fosters conservation ecosystems. Doi: 10.28991/HEF-2024-05-02-06 Full Text: PDF

Language: Английский

---

Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan

Ecological Indicators, Journal Year: 2024, Volume and Issue: 159, P. 111670 - 111670

Published: Feb. 1, 2024

Dry land ecosystems extend over 40 % of the Earth, supporting an estimated 3 billion human population. Thus, quantifying LCLU changes in such is essential for achieving sustainable development goals. In this context, research aimed to examine past three decades (1990 – 2020) arid ecosystem Pakistan, i.e., Cholisatn desert. Three remote sensing indices, normalized difference vegetation index (NDVI), barren (NDBaI), and top grain soil (TGSI) are taken as representatives their temporal relationship associated with meteorological drought, e.g. standardized precipitation (SPI). Moreover, machine learning-based random forest (RF) classification followed by change detection techniques was implemented. Results from RF classifier revealed applicability accurately predicting LULC validation overall accuracy 0.99. Output interesting finding where desert experienced significant last decades. The highest expansion (4.4 %) took place 2014 2020 at expense reduction (-6.3 %). Mann-Kendall trend (MK) Sen's slope (SS) analysis showed a (P < 0.001) increasing NDVI (SS = 0.004), SPI 0.01 0.04) decreasing NDBaI TGSI -0.001, −0.005). Interestingly, positive Pearson correlation range (r 0.6–0.8) SPI-1 6, negative 0.5–0.7) indices reveals strong linear between drought. provides substantial implications policy makers stakeholders emphasizing need proactive strategies drought resistant improve maintain ecological health combating impacts climatic change.

Language: Английский

---

Comparative analysis of sensors and classification algorithms for land cover classification in Islamabad, Pakistan

Earth Science Informatics, Journal Year: 2025, Volume and Issue: 18(2)

Published: Jan. 29, 2025

Language: Английский

---

Deep Learning Techniques for Enhanced Mangrove Land use and Land change from Remote Sensing Imagery: A Blue Carbon Perspective

Big Data Research, Journal Year: 2024, Volume and Issue: unknown, P. 100478 - 100478

Published: June 1, 2024

Language: Английский

---

Comprehensive analysis of land use and cover dynamics in djibouti using machine learning technique: A multi-temporal assessment from 1990 to 2023

Environmental Challenges, Journal Year: 2024, Volume and Issue: 15, P. 100920 - 100920

Published: April 1, 2024

Understanding land use and cover (LULC) dynamics in semi-arid regions is vital for unraveling complex environmental processes resource management. This study delves into the intricate interplay of patterns dynamics, offering indispensable insights repercussions these changes. The aims to quantify categories Djibouti's semi-desert region using remote sensing. It analyzes temporal changes evaluates Random Forest (RF) algorithms classification. Through meticulous quantification comprehensive analysis, research contributes significantly sensing science by enhancing understanding informing sustainable management practices. Leveraging machine learning supervised classification on Google Earth Engine (GEE) platform Landsat data spanning four time periods (1990, 2002, 2012, 2023), alongside spectral indices Digital Elevation Model (DEM) data, our achieves unprecedented insights. Our findings reveal a significant landscape transformation, delineating seven major classes: mangroves, bushes, farmland, built-up areas, water bodies, barren land, salt plains. With overall accuracy ranging from 89% 95%, assessments demonstrate types over studied period. Notably, areas witnessed declines, while lands expanded. underscores pivotal role monitoring long-term their ecological impacts. By harnessing technological advancements, empowers stakeholders make informed decisions conservation landscapes.

Language: Английский

---

Deep learning approaches for estimating forest vegetation cover and exploring influential ecosystem factors

Earth Science Informatics, Journal Year: 2024, Volume and Issue: unknown

Published: June 7, 2024

Language: Английский

---

Study on the extraction method of Glycyrrhiza uralensis Fisch. distribution area based on Gaofen-1 remote sensing imagery: a case study of Dengkou county

Frontiers in Plant Science, Journal Year: 2025, Volume and Issue: 16

Published: March 7, 2025

Glycyrrhiza uralensis Fisch., a perennial medicinal plant with robust root system, plays significant role in mitigating land desertification when cultivated extensively. This study investigates Dengkou County, semi-arid region, as the research area. First, reflectance differences of feature types, and importance bands were evaluated by using random forest (RF) algorithm. Second, after constructing G. vegetation index (GUVI), recognition accuracy was compared between RF classification model constructed based on January-December GUVI common indices set support vector machine (SVM) set. Finally, spectral characteristics other types under 2022 analyzed, historical distribution identified mapped. The results demonstrated that blue near-infrared are particularly for distinguishing . Incorporating year-round (January-December) data significantly improved identification accuracy, achieving producer’s 97.26%, an overall 93.00%, Kappa coefficient 91.38%, user’s 97.32%. Spectral analysis revealed distinct different years types. From 2014 to 2022, expanded from northeast County central southwestern regions, transitioning small, scattered patches larger, concentrated areas. highlights effectiveness models identifying , demonstrating superior performance alternative sets or algorithms. However, generalizability may be limited due influence natural anthropogenic factors Therefore, regional adjustments optimization parameters necessary. provides valuable reference employing remote sensing technology accurately map current regions similar environmental conditions.

Language: Английский

---

Multisource remote sensing and ensemble learning for multidimensional monitoring of heavy metals on mine surfaces

Environmental Geochemistry and Health, Journal Year: 2025, Volume and Issue: 47(5)

Published: April 26, 2025

Language: Английский

---

Enhancing Deforestation Detection Through Multi-Domain Adaptation with Uncertainty Estimation

Forests, Journal Year: 2025, Volume and Issue: 16(5), P. 742 - 742

Published: April 26, 2025

Deep learning models have shown great potential in scientific research, particularly remote sensing for monitoring natural resources, environmental changes, land cover, and use. semantic segmentation techniques enable cover classification, change detection, object identification, vegetation health assessment, among other applications. However, their effectiveness relies on large labeled datasets, which are costly time-consuming to obtain. Domain adaptation (DA) address this challenge by transferring knowledge from a source domain one or more unlabeled target domains. While most DA research focuses single-target single-source problems, multi-target multi-source scenarios remain underexplored. This work proposes deep approach that uses Adversarial Neural Networks (DANNs) deforestation detection multi-domain settings. Additionally, an uncertainty estimation phase is introduced guide human review high-uncertainty areas. Our evaluated set of Landsat-8 images the Amazon Brazilian Cerrado biomes. In experiments, single contains data, while samples domains unlabeled. scenarios, multiple used train models, later domain. The results show significant accuracy improvements over lower-bound baselines, as indicated F1-Score values, uncertainty-based showed further enhance performance, reaching upper-bound baselines certain combinations. As our independent network architecture, we believe it opens new perspectives improving generalization capacity learning-based methods. Furthermore, operational point view, has areas around world lack accurate reference data adequately task.

Language: Английский

---