Capture and Prediction of Rainfall-Induced Landslide Warning Signals Using an Attention-Based Temporal Convolutional Neural Network and Entropy Weight Methods

Zhang, Di; Wei, Kai; Yao, Yi; Yang, Jiacheng; Zheng, Guolong; Li, Qing

doi:10.3390/s22166240

Cited by 8 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where Pij is the normalized indicator data after normalization [39], ej is the entropy value of the indicator, and m is the data set, which is the number of data sets being four in this study [40].…”

Section: Weight Calculation Methodsmentioning

confidence: 99%

Spatial and temporal evolution characteristics and influencing factors of coupling coordination degree of urbanization and eco-environment in the Huaihe Eco-economic Belt

Yao

Chen

Sheng

et al. 2023

Preprint

View full text Add to dashboard Cite

The rapid development of urbanization in recent years has brought about economic and social development, but has led to a series of problems in the eco-environment. On the basis of the evaluation model of urbanization and eco-environment constructed for the Huaihe Eco-economic Belt, a comprehensive analysis of the coupling coordination degree between the two was conducted from 2005 to 2020, and the aggregation characteristics in space were explored, as well as the reasons for the uneven development of urbanization and eco-environment were revealed. The results show that: (1) between 2005 and 2020, the development level and the coupling coordination degree of urbanization and eco-environment in the Huaihe Eco-economic Belt all increased steadily. (2) In terms of global spatial autocorrelation, urbanization and coupling coordination showed a strong positive correlation in 2005, but weakened year by year as time progressed, while eco-environment shows a random spatial distribution. In terms of local autocorrelation, in 2005, urbanization local autocorrelation was mainly clustered in Shandong and Jiangsu Province, but both became more and more spatially dispersed as time progressed. (3) Spatial urbanization was the main aspect affecting the uneven level of urbanization in the Huaihe Eco-economic Belt in 2005, and by 2020, population urbanization had become a relatively weak aspect of urbanization in many cities, including Taizhou, Yancheng and Pingdingshan. In terms of eco-environment, between 2005 and 2015, ecological pressure became a problem for many cities, but in 2020, ecological pressure has weakened again, and soon the state of eco-environment has become a new weak point. The development of central cities such as Bengbu and Huai’an should be strengthened, talents should be retained, the urbanization rate of the population should be increased, and relevant environmental protection policies should be implemented to promote the sustainable development of cities within the Huaihe Eco-economic Belt.

show abstract

Section: Weight Calculation Methodsmentioning

confidence: 99%

Spatial and temporal evolution characteristics and influencing factors of coupling coordination degree of urbanization and eco-environment in the Huaihe Eco-economic Belt

Yao

Chen

Sheng

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Figure 7 shows the results of regression analysis in the third threshold interval with different regression models. The results of the regression relationship are shown in Formulas ( 19)- (21). According to the above methods, the regression model was used to evaluate the accuracy of the early warning threshold intervals at different spatial scales with different scales.…”

Section: Accuracy Evaluationmentioning

confidence: 99%

“…The comprehensive evaluation method has fewer requirements for quantitative data. Through mathematical analysis and processing of the relationship between multiple variables, it is possible to obtain a quantitative value that is closer to the actual situation, so this method should be applied to study areas with many variables or difficult quantitative factors [21]. However, the mathematical calculation in this method is relatively complicated, and there is a certain subjectivity for the weights between multiple variables.…”

Section: Introductionmentioning

confidence: 99%

The Method of Segmenting the Early Warning Thresholds Based on Fisher Optimal Segmentation

Lei

Qin

et al. 2023

Land

View full text Add to dashboard Cite

Most slope collapse accidents are indicated by certain signs before their occurrence, and unnecessary losses can be avoided by predicting slope deformation. However, the early warning signs of slope deformation are often misjudged. It is necessary to establish a method to determine the appropriate early warning signs in sliding thresholds. Here, to better understand the impact of different scales on the early warning signs of sliding thresholds, we used the Fisher optimal segmentation method to establish the early warning signs of a sliding threshold model based on deformation speed and deformation acceleration at different spatial scales. Our results indicated that the accuracy of the early warning signs of sliding thresholds at the surface scale was the highest. Among them, the early warning thresholds of the blue, yellow, orange, and red level on a small scale were 369.31 mm, 428.96 mm, 448.41 mm, and 923.7 mm, respectively. The evaluation accuracy of disaster non-occurrence and occurrence was 93.25% and 92.41%, respectively. The early warning thresholds of the blue, yellow, orange, and red level on a large scale were 980.11 mm, 1038.16 mm, 2164.63 mm, and 9492.75 mm, respectively. The evaluation accuracy of disaster non-occurrence and occurrence was 97.22% and 97.44%, respectively. Therefore, it is necessary to choose deformation at the surface scale with a large scale as the sliding threshold. Our results effectively solve the problem of misjudgment of the early warning signs of slope collapse, which is of great significance for ensuring the safe operation of water conservation projects and improving the slope deformation warning capability.

show abstract

“…Predicting rainfall is very important when determining the likelihood of a landslide. The prediction of landslides was determined based on rainfall forecasts utilizing a neural network with entropy weighting (D. Zhang et al, 2022). Another rainfall prediction was performed on weather data from Bangladesh spanning the years 1901 to 2015 (Azmain et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

Performance evaluation of Machine Learning based regression models for rainfall forecasting

Saleh,

Rasel

2024

Preprint

View full text Add to dashboard Cite

Rainfall plays a crucial role in the water cycle, serving as a direct input for agricultural practices and water resource management. However, its patterns vary significantly across different regions, creating challenges for sustainable water use. This study focuses on the Rangpur district in northwestern Bangladesh, where irrigation heavily relies on unpredictable rainfall. To address this, this study employed three machine learning regression methods – Random Forest, Support Vector Machine, and Gradient Boosting Machine – using historical annual rainfall data from 1990 to 2020. The analysis was conducted on Google Colab, an open-source Python environment. Hyperparameter optimization via grid search was conducted on the three models to maximize prediction accuracy. The analysis revealed that the Random Forest model to be the most accurate for rainfall prediction in the Rangpur district. During the testing phase, it achieved an R-squared value of 0.75, indicating a strong correlation between predictions and actual rainfall. Interestingly, Gradient Boosting Machine outperformed Random Forest in the training phase, highlighting the importance of considering both training and testing performance for model selection. Additionally, Random Forest regression confirmed the strong relationship between predicted and observed rainfall by generating highest correlation (97%). This study demonstrates the effectiveness of Random Forest Regression for forecasting rainfall in Rangpur district. This knowledge can contribute to resilient water management strategies, enabling farmers and authorities to adapt irrigation practices and optimize resource allocation in response to predicted precipitation patterns. Future research could involve incorporating additional environmental variables into the model and exploring ensemble learning techniques for potentially further improving prediction accuracy.

show abstract

Capture and Prediction of Rainfall-Induced Landslide Warning Signals Using an Attention-Based Temporal Convolutional Neural Network and Entropy Weight Methods

Cited by 8 publications

References 36 publications

Spatial and temporal evolution characteristics and influencing factors of coupling coordination degree of urbanization and eco-environment in the Huaihe Eco-economic Belt

Spatial and temporal evolution characteristics and influencing factors of coupling coordination degree of urbanization and eco-environment in the Huaihe Eco-economic Belt

The Method of Segmenting the Early Warning Thresholds Based on Fisher Optimal Segmentation

Performance evaluation of Machine Learning based regression models for rainfall forecasting

Contact Info

Product

Resources

About