Yongfan Guo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Yongfan Guo has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Hao Zhou
School of Urban Construction, Changzhou University, Changzhou 213164, China
2 shared publications Hongwei Jiang
School of Urban Construction, Changzhou University, Changzhou, China
2 shared publications- Hongwei Jiang
School of Urban Construction, Changzhou University, Changzhou 213164, China
2 shared publications Jiayi Wu
School of Urban Construction, Changzhou University, Changzhou, China
2 shared publications- Jiayi Wu
School of Urban Construction, Changzhou University, Changzhou 213164, China
2 shared publications Mengjie Liu
School of Urban Construction, Changzhou University, Changzhou, China
2 shared publications
Latest Publications
- 17 January 2026
- Published by Springer Nature in Carbonates and Evaporites
- Vol. 41 (1) , 16
Ensemble learning for landslide susceptibility mapping: a review of machine learning and hybrid approaches
- 17 January 2026
- Published by Springer Nature in Carbonates and Evaporites
- Vol. 41 (1) , 16
Ensemble learning for landslide susceptibility mapping: a review of machine learning and hybrid approaches
- Open Access
The Application of KNN-Optimized Hybrid Models in Landslide Displacement Prediction
Early warning systems depend heavily on the accuracy of landslide displacement forecasts. This study focuses on the Bazimen landslide located in the Three Gorges Reservoir region and proposes a hybrid prediction approach combining support vector regression (SVR) and long short-term memory (LSTM) networks. These models are optimized via the K-Nearest Neighbor (KNN) algorithm. Initially, cumulative displacement data were separated into trend and cyclic elements using a smoothing approach. SVR and LSTM were then used to predict the components, and KNN was introduced to optimize input factors and classify the results, improving accuracy. The final KNN-optimized SVR-LSTM model effectively integrates static and dynamic features, addressing limitations of traditional models. The results show that LSTM performs better than SVR, with an RMSE and MAPE of 24.73 mm and 1.87% at monitoring point ZG111, compared to 30.71 mm and 2.15% for SVR. The sequential hybrid model based on KNN-optimized SVR and LSTM achieved the best performance, with an RMSE and MAPE of 23.11 mm and 1.68%, respectively. This integrated model, which combines multiple algorithms, offers improved prediction of landslide displacement and practical value for disaster forecasting in the Three Gorges area.
Open AccessThe Application of KNN-Optimized Hybrid Models in Landslide Displacement Prediction
Early warning systems depend heavily on the accuracy of landslide displacement forecasts. This study focuses on the Bazimen landslide located in the Three Gorges Reservoir region and proposes a hybrid prediction approach combining support vector regression (SVR) and long short-term memory (LSTM) networks. These models are optimized via the K-Nearest Neighbor (KNN) algorithm. Initially, cumulative displacement data were separated into trend and cyclic elements using a smoothing approach. SVR and LSTM were then used to predict the components, and KNN was introduced to optimize input factors and classify the results, improving accuracy. The final KNN-optimized SVR-LSTM model effectively integrates static and dynamic features, addressing limitations of traditional models. The results show that LSTM performs better than SVR, with an RMSE and MAPE of 24.73 mm and 1.87% at monitoring point ZG111, compared to 30.71 mm and 2.15% for SVR. The sequential hybrid model based on KNN-optimized SVR and LSTM achieved the best performance, with an RMSE and MAPE of 23.11 mm and 1.68%, respectively. This integrated model, which combines multiple algorithms, offers improved prediction of landslide displacement and practical value for disaster forecasting in the Three Gorges area.
- Open Access
- 9 June 2025
- Published by MDPI AG in Preprints.org
The Application of Machine Learning Coupled Models and Optimization Algorithms in the Field of Landslide Displacement Prediction
Accurate landslide displacement prediction is an essential component of landslide early warning systems. The Bazimen landslide in the Three Gorges reservoir area was taken as a case study. In recent years, the outstanding performance of dynamic and static models has attracted the attention of many scholars. Static machine learning algorithm—support vector regression (SVR) and dy-namic machine learning algorithm—long short-term memory neural network (LSTM) were used to predict landslide displacement in this study. At first, to predict landslide displacement accu-rately, the moving average method was used to decompose the cumulative displacement into two components: trend and periodic terms. Second, by introducing the K-Nearest Neighbor (KNN), the forecast results of SVR and LSTM models were classified with greater accuracy based on the se-lection of input factors. Finally, we proposed a static-dynamic ensemble landslide displacement predic-tion model optimized with KNN to address the dynamic characteristics of landslide evo-lution and the shortcomings of traditional static prediction models. Through the out-put of the KNN model, the results of the static-dynamic coupling model are updated to get the KNN-optimized SVR-LSTM landslide displacement prediction model. Compared with traditional landslide displacement prediction models that output prediction values, this study optimized the prediction results through a classification algorithm. The results show that the LSTM models show better performance than the SVR models: For monitor-ing point ZG111, the values of RMSE and MAPE with the SVR model were 30.71 mm and 2.15%, while the accuracy factors of the LSTM model were 24.73 mm and 1.87% and the values of RMSE and MAPE with the Ensemble model were 23.11 mm and 1.68%. It can be seen that the ensemble model integrates the advantages of static (SVR) and dynamic (LSTM) models, and its prediction performance is better than that of the single SVR model and the LSTM model. This study provides an integrated view of the landslide displace-ment pre-diction model, which can provide a reference for predicting geological disasters in the Three Gorges reservoir area.
- 9 June 2025
- Published by MDPI AG in Preprints.org
Open AccessThe Application of Machine Learning Coupled Models and Optimization Algorithms in the Field of Landslide Displacement Prediction
Accurate landslide displacement prediction is an essential component of landslide early warning systems. The Bazimen landslide in the Three Gorges reservoir area was taken as a case study. In recent years, the outstanding performance of dynamic and static models has attracted the attention of many scholars. Static machine learning algorithm—support vector regression (SVR) and dy-namic machine learning algorithm—long short-term memory neural network (LSTM) were used to predict landslide displacement in this study. At first, to predict landslide displacement accu-rately, the moving average method was used to decompose the cumulative displacement into two components: trend and periodic terms. Second, by introducing the K-Nearest Neighbor (KNN), the forecast results of SVR and LSTM models were classified with greater accuracy based on the se-lection of input factors. Finally, we proposed a static-dynamic ensemble landslide displacement predic-tion model optimized with KNN to address the dynamic characteristics of landslide evo-lution and the shortcomings of traditional static prediction models. Through the out-put of the KNN model, the results of the static-dynamic coupling model are updated to get the KNN-optimized SVR-LSTM landslide displacement prediction model. Compared with traditional landslide displacement prediction models that output prediction values, this study optimized the prediction results through a classification algorithm. The results show that the LSTM models show better performance than the SVR models: For monitor-ing point ZG111, the values of RMSE and MAPE with the SVR model were 30.71 mm and 2.15%, while the accuracy factors of the LSTM model were 24.73 mm and 1.87% and the values of RMSE and MAPE with the Ensemble model were 23.11 mm and 1.68%. It can be seen that the ensemble model integrates the advantages of static (SVR) and dynamic (LSTM) models, and its prediction performance is better than that of the single SVR model and the LSTM model. This study provides an integrated view of the landslide displace-ment pre-diction model, which can provide a reference for predicting geological disasters in the Three Gorges reservoir area.
- Open Access
- 9 June 2025
- Published by California Digital Library (CDL) in EarthArXiv
Ensemble methods for landslide susceptibility mapping: A review of machine learning and hybrid approaches
The assessment of landslide susceptibility holds significant importance in disaster risk reduction. This study comprehensively examines the current research on landslide susceptibility from two aspects: the steps involved in landslide susceptibility assessment and modeling methods. Initially, we retrieved pertinent research articles, published between 2014 and 2023, and focused on “Landslide SensitivityAssessment” from the Web of Science database. Subsequently, we identified frequently occurring keywords in landslide susceptibility assessment studies employing ensemble learning methods during the past decade and created analytical charts. The standard methods for landslide inventory, evaluation indicators, and validation techniques were introduced along with their advantages and limitations. The shortcomings of eachmethod were identified, and potential future research directions were outlined. Finally, a detailed analysis of the use of ensemble methods in landslide susceptibility assessment was conducted; this is presented in several sections. The findings indicate that the advancement of ensemble learning methods has facilitated the development of landslide susceptibility assessment, rendering the landslide modeling process more efficient and accurate. In turn, this has enhanced the intelligence of models in landslide susceptibility research. The results of this study can help researchers understand the current conditions of landslide susceptibility research and provide a reference for subsequent research in this field.
- 9 June 2025
- Published by California Digital Library (CDL) in EarthArXiv
Open AccessEnsemble methods for landslide susceptibility mapping: A review of machine learning and hybrid approaches
The assessment of landslide susceptibility holds significant importance in disaster risk reduction. This study comprehensively examines the current research on landslide susceptibility from two aspects: the steps involved in landslide susceptibility assessment and modeling methods. Initially, we retrieved pertinent research articles, published between 2014 and 2023, and focused on “Landslide SensitivityAssessment” from the Web of Science database. Subsequently, we identified frequently occurring keywords in landslide susceptibility assessment studies employing ensemble learning methods during the past decade and created analytical charts. The standard methods for landslide inventory, evaluation indicators, and validation techniques were introduced along with their advantages and limitations. The shortcomings of eachmethod were identified, and potential future research directions were outlined. Finally, a detailed analysis of the use of ensemble methods in landslide susceptibility assessment was conducted; this is presented in several sections. The findings indicate that the advancement of ensemble learning methods has facilitated the development of landslide susceptibility assessment, rendering the landslide modeling process more efficient and accurate. In turn, this has enhanced the intelligence of models in landslide susceptibility research. The results of this study can help researchers understand the current conditions of landslide susceptibility research and provide a reference for subsequent research in this field.
- 8 January 2025
- Published by Elsevier in Computers and Geotechnics
- Vol. 179
Thermodynamics-based constitutive modeling of chemically-induced grain degradation for cohesive granular materials
- 8 January 2025
- Published by Elsevier in Computers and Geotechnics
- Vol. 179
Thermodynamics-based constitutive modeling of chemically-induced grain degradation for cohesive granular materials
- 24 November 2023
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 419
A reactive-transport phase-field modelling approach of chemo-assisted cracking in saturated sandstone
- 24 November 2023
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 419
A reactive-transport phase-field modelling approach of chemo-assisted cracking in saturated sandstone
- 1 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 118-125
A Practical Transient Stability Margin Algorithm Based on Information-Driven and Fine Fault Information
At present, the transient stability control adopts the uniform cutting stability control measure and asymmetric faults are treated with three-phase short circuit, called event-driven analysis, which ensures the reliability of the system stability, but the accuracy is not enough. Taking three-phase short circuit as a benchmark, this paper analyzes the relationship of limit cutting time between the three-phase short circuit and various asymmetric faults in different fault positions, presents the angle-time curve of four kinds of short-circuit faults with a unified expression and obtains the limit cutting time directly according to the limit cutting angle, which turns event-driven into information -driven. The single-machine infinite bus model is built in matlab and the analysis of asymmetric faults is verified. It is found that the maximum difference of limit cutting time between asymmetric faults and three-phase short circuit is 500ms. Therefore, it is of great significance to obtain the accurate limit cutting time of asymmetric faults.
- 1 November 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 118-125
A Practical Transient Stability Margin Algorithm Based on Information-Driven and Fine Fault Information
At present, the transient stability control adopts the uniform cutting stability control measure and asymmetric faults are treated with three-phase short circuit, called event-driven analysis, which ensures the reliability of the system stability, but the accuracy is not enough. Taking three-phase short circuit as a benchmark, this paper analyzes the relationship of limit cutting time between the three-phase short circuit and various asymmetric faults in different fault positions, presents the angle-time curve of four kinds of short-circuit faults with a unified expression and obtains the limit cutting time directly according to the limit cutting angle, which turns event-driven into information -driven. The single-machine infinite bus model is built in matlab and the analysis of asymmetric faults is verified. It is found that the maximum difference of limit cutting time between asymmetric faults and three-phase short circuit is 500ms. Therefore, it is of great significance to obtain the accurate limit cutting time of asymmetric faults.
- Open Access
A Geochemical Model of Fluids and Mineral Interactions for Deep Hydrocarbon Reservoirs
A mutual solubility model for CO2-CH4-brine systems is constructed in this work as a fundamental research for applications of deep hydrocarbon exploration and production. The model is validated to be accurate for wide ranges of temperature (0–250°C), pressure (1–1500 bar), and salinity (NaCl molality from 0 to more than 6 mole/KgW). Combining this model with PHREEQC functionalities, CO2-CH4-brine-carbonate-sulfate equilibrium is calculated. From the calculations, we conclude that, for CO2-CH4-brine-carbonate systems, at deeper positions, magnesium is more likely to be dissolved in aqueous phase and calcite can be more stable than dolomite and, for CO2-CH4-brine-sulfate systems, with a presence of CH4, sulfate ions are likely to be reduced to S2−and H2S in gas phase could be released after S2−saturated in the solution. The hydrocarbon “souring” process could be reproduced from geochemical calculations in this work.
Open AccessA Geochemical Model of Fluids and Mineral Interactions for Deep Hydrocarbon Reservoirs
A mutual solubility model for CO2-CH4-brine systems is constructed in this work as a fundamental research for applications of deep hydrocarbon exploration and production. The model is validated to be accurate for wide ranges of temperature (0–250°C), pressure (1–1500 bar), and salinity (NaCl molality from 0 to more than 6 mole/KgW). Combining this model with PHREEQC functionalities, CO2-CH4-brine-carbonate-sulfate equilibrium is calculated. From the calculations, we conclude that, for CO2-CH4-brine-carbonate systems, at deeper positions, magnesium is more likely to be dissolved in aqueous phase and calcite can be more stable than dolomite and, for CO2-CH4-brine-sulfate systems, with a presence of CH4, sulfate ions are likely to be reduced to S2−and H2S in gas phase could be released after S2−saturated in the solution. The hydrocarbon “souring” process could be reproduced from geochemical calculations in this work.