Geological Hazard Susceptibility Analysis Based on RF, SVM, and NB Models, Using the Puge Section of the Zemu River Valley as an Example

Sustainability, Journal Year: 2023, Volume and Issue: 15(14), P. 11228 - 11228

Published: July 19, 2023

The purpose of this study was to construct a geological hazard susceptibility evaluation and analysis model using three types machine learning models, namely, random forest (RF), support vector (SVM), naive Bayes (NB), evaluate the landslides, Puge section Zemu River valley in Liangshan Yi Autonomous Prefecture as area. First, 89 shallow landslide debris flow locations were recognized through field surveys remote sensing interpretation. A total eight hazard-causing factors, slope, aspect, rock group, land cover, distance road, river, fault, normalized difference vegetation index (NDVI), selected spatial relationship with occurrence. As result analysis, results weighting factors indicate that two elements group river contribute most creation hazards. After comparing all indices had higher correct area under ROC curve (AUC) value 0.87, root mean squared error (RMSE) 0.118, absolute (MAE) 0.045. SVM highest sensitivity prediction matched actual situation area, effects good. assessment models are able provide help for prevention control hazards same type areas.

Language: Английский

---

Deep Learning-Based Recognition and Classification of Soiled Photovoltaic Modules Using HALCON Software for Solar Cleaning Robots

Sensors, Journal Year: 2025, Volume and Issue: 25(5), P. 1295 - 1295

Published: Feb. 20, 2025

The global installation capacity of solar photovoltaic (PV) systems is exponentially increasing. However, the accumulation soil and debris on panels significantly reduces their efficiency, necessitating frequent cleaning to maintain optimal energy output. This study presents a deep learning-based approach for recognition classification soiled PV images, aimed at enhancing capabilities robots through HALCON software framework. Using EANN CNN architecture along with advanced image processing techniques, proposed system achieves precise detection soiling patterns. framework facilitates acquisition, preprocessing, segmentation, deployment trained models robotic control. demonstrate exceptional accuracy, achieving precision 99.87% 99.91%, respectively. Experimental results highlight system’s potential improve automation strategies, reduce unnecessary cycles, enhance overall performance panels. research underscores transformative role intelligent visual analysis in optimizing maintenance practices renewable applications.

Language: Английский

---

Euclidean Distance-Based Tree Algorithm for Fault Detection and Diagnosis in Photovoltaic Systems

Energies, Journal Year: 2025, Volume and Issue: 18(7), P. 1773 - 1773

Published: April 1, 2025

In this paper, a new methodology for fault detection and diagnosis in photovoltaic systems is proposed. This method employs novel Euclidean distance-based tree algorithm to classify various considered faults. Unlike the decision tree, which requires use of Gini index split data, mainly relies on computing distances between an arbitrary point space entire dataset. Then, minimum maximum each class are extracted ordered ascending order. The proposed four attributes: Solar irradiance, temperature, coordinates power (Impp, Vmpp). developed procedure implemented applied dataset comprising seven distinct classes: normal operation, string disconnection, short circuit three modules, ten cases with 25%, 50%, 75% partial shading. obtained results demonstrate high efficiency effectiveness methodology, classification accuracy reaching 97.33%. A comparison study Support Vector Machine, Decision Tree, Random Forest, K-Nearest Neighbors algorithms conducted. shows performance against other terms accuracy, precision, recall, F1-score.

Language: Английский

---

Performance enhancement through increased mass transport in a modified novel flow pattern for vanadium redox flow battery

Ionics, Journal Year: 2025, Volume and Issue: unknown

Published: April 7, 2025

Language: Английский

---

Various types of A.C conduction mechanism models for solid polymer electrolytes (SPE): A review

Journal of Power Sources, Journal Year: 2025, Volume and Issue: 645, P. 237217 - 237217

Published: May 1, 2025

Language: Английский

---