Urban waterlogging susceptibility assessment based on a PSO-SVM method using a novel repeatedly random sampling idea to select negative samples

Tang, Xianzhe; Hong, Haoyuan; Shu, Yuqin; Tang, Huijun; Li, Jiufeng; Liu, Wei

doi:10.1016/j.jhydrol.2019.06.058

Cited by 85 publications

(23 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the slope is also an important topographical feature that impacts the occurrence of rainstorm waterlogging [68]. A region with a steeper slope has a lower possibility of being inundated because rainwater can easily flow downslope [69]. Therefore, in this study, we selected elevation and slope to represent the topographic characteristics.…”

Section: Topographymentioning

confidence: 99%

Driving Factors and Risk Assessment of Rainstorm Waterlogging in Urban Agglomeration Areas: A Case Study of the Guangdong-Hong Kong-Macao Greater Bay Area, China

Liu

et al. 2021

Water

View full text Add to dashboard Cite

Understanding the driving factors and assessing the risk of rainstorm waterlogging are crucial in the sustainable development of urban agglomerations. Few studies have focused on rainstorm waterlogging at the scale of urban agglomeration areas. We used the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) of China as a case study. Kernel density estimation (KDE) and spatial autocorrelation analysis were applied to study the spatial distribution characteristics of rainstorm waterlogging spots during 2013–2017. A geographical detector (GD) and geographically weighted regression (GWR) were used to discuss the driving mechanism of rainstorm waterlogging by considering eight driving factors: impervious surface ratio (ISR), mean shape index of impervious surface (Shape_MN), aggregation index of impervious surface (AI), fractional vegetation cover (FVC), elevation, slope, river density, and river distance. The risk of rainstorm waterlogging was assessed using GWR based on principal component analysis (PCA). The results show that the spatial distribution of rainstorm waterlogging in the GBA has the characteristics of multicenter clustering. Land cover characteristic factors are the most important factors influencing rainstorm waterlogging in the GBA and most of the cities within the GBA. The rainstorm waterlogging density increases when ISR, Shape_MN, and AI increase, while it decreases when FVC, elevation, slope, and river distance increase. There is no obvious change rule between rainstorm waterlogging and river density. All of the driving factors enhance the impacts on rainstorm waterlogging through their interactions. The relationships between rainstorm waterlogging and the driving factors have obvious spatial differences because of the differences in the dominant factors affecting rainstorm waterlogging in different spatial positions. Furthermore, the result of the risk assessment of rainstorm waterlogging indicates that the southwest area of Guangzhou and the central area of Shenzhen have the highest risks of rainstorm waterlogging in GBA. These results may provide references for rainstorm waterlogging mitigation through urban renewal planning in urban agglomeration areas.

show abstract

Section: Topographymentioning

confidence: 99%

Driving Factors and Risk Assessment of Rainstorm Waterlogging in Urban Agglomeration Areas: A Case Study of the Guangdong-Hong Kong-Macao Greater Bay Area, China

Liu

et al. 2021

Water

View full text Add to dashboard Cite

show abstract

“…Each particle is adjusted according to the fitness values of themselves and the swarm. And the iteration and optimization are not terminated until all the particles converge to optimal solution [68][69][70][71] (Figure 5(b)). Hence, particles are optimized by constantly updating their speed and position, in which process can be expressed as…”

Section: Particle Swarm Optimizationmentioning

confidence: 99%

A Comparative Study of Landslide Susceptibility Mapping Using SVM and PSO-SVM Models Based on Grid and Slope Units

Zhao

2021

Mathematical Problems in Engineering

View full text Add to dashboard Cite

The main purpose of this study aims to apply and compare the rationality of landslide susceptibility maps using support vector machine (SVM) and particle swarm optimization coupled with support vector machine (PSO-SVM) models in Lueyang County, China, enhance the connection with the natural terrain, and analyze the application of grid units and slope units. A total of 186 landslide locations were identified by earlier reports and field surveys. The landslide inventory was randomly divided into two parts: 70% for training dataset and 30% for validation dataset. Based on the multisource data and geological environment, 16 landslide conditioning factors were selected, including control factors and triggering factors (i.e., altitude, slope angle, slope aspect, plan curvature, profile curvature, SPI, TPI, TRI, lithology, distance to faults, TWI, distance to rivers, NDVI, distance to roads, land use, and rainfall). The susceptibility between each conditioning factor and landslide was deduced using a certainty factor model. Subsequently, combined with grid units and slope units, the landslide susceptibility models were carried out by using SVM and PSO-SVM methods. The precision capability of the landslide susceptibility mapping produced by different models and units was verified through a receiver operating characteristic (ROC) curve. The results showed that the PSO-SVM model based on slope units had the best performance in landslide susceptibility mapping, and the area under the curve (AUC) values of training and validation datasets are 0.945 and 0.9245, respectively. Hence, the machine learning algorithm coupled with slope units can be considered a reliable and effective technique in landslide susceptibility mapping.

show abstract

“…( 5) The sample set is trained through the SVDD algorithm classifier. Figure 6 shows the characteristics of a mechanical fault diagnosis system based on the SVDD algorithm [21].…”

Section: Design Of Mechanical Fault Diagnosis Model For a High Voltage Switchmentioning

confidence: 99%

Mechanical Fault Diagnosis Using a SVDD Intelligent Algorithm in Fabrics

2021

IJOMAM

View full text Add to dashboard Cite

To shorten the time and improve the accuracy in mechanical fault diagnosis, the fault diagnosis mechanism is studied of the high voltage switch in power system, and a new system is proposed as a fast, safe, and accurate mechanical fault diagnosis scheme of high voltage switch. First, the basic structure and common faults of high voltage switches are analyzed, the feature data set has a poor classification effect, and the SVM (Support Vector Machine) algorithm model can process all the feature data for an effective classification and recognition process. Finally, the high voltage switches mechanical faults are extremely scattered, the improved SVDD (Support Vector Data Description) intelligent algorithm can acquire, screen, and classify the mechanical faults data. The results of the improved SVDD intelligent algorithm are less affected by the classification boundary and converge to the average sample value. The new fault diagnosis system is proposed for a high voltage switch based on the improved SVDD intelligent algorithm. Subsequently, the simulation experiment is conducted for a high voltage switch fault diagnosis system based on the SVDD intelligent algorithm. The experimental results show that compared with the traditional mechanical fault diagnosis system, the proposed system can identify new fault types effectively. Meanwhile, the proposed system can quickly and efficiently classify and identify the operation states and fault types of machinery. Consequently, the accuracy of the mechanical fault diagnosis system is improved, providing technical support for the power system operation. Hence, the proposed fault diagnosis system is very suitable for the production of the power system and can greatly promote mechanical fault diagnosis in China.

show abstract

Urban waterlogging susceptibility assessment based on a PSO-SVM method using a novel repeatedly random sampling idea to select negative samples

Cited by 85 publications

References 55 publications

Driving Factors and Risk Assessment of Rainstorm Waterlogging in Urban Agglomeration Areas: A Case Study of the Guangdong-Hong Kong-Macao Greater Bay Area, China

Driving Factors and Risk Assessment of Rainstorm Waterlogging in Urban Agglomeration Areas: A Case Study of the Guangdong-Hong Kong-Macao Greater Bay Area, China

A Comparative Study of Landslide Susceptibility Mapping Using SVM and PSO-SVM Models Based on Grid and Slope Units

Mechanical Fault Diagnosis Using a SVDD Intelligent Algorithm in Fabrics

Contact Info

Product

Resources

About