Using ANFIS and BPNN Methods to Predict the Unfrozen Water Content of Saline Soil in Western Jilin, China

Liu, Yufeng; Wang, Qing; Zhang, Xudong; Song, Shengyuan; Niu, Cencen; Shangguan, Yunlong

doi:10.3390/sym11010016

Cited by 56 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It demonstrates that the unfrozen water content of frozen soil is practically independent of its initial water content. This result has also been observed in other soils (e.g., Liu et al, 2019; Nagare et al, 2012; Watanabe & Wake, 2009). In contrast, other studies found that unfrozen water content depends on initial water content (Suzuki, 2004; Teng et al, 2020; Xu et al, 1987).…”

Section: Theory Of Freezing and Thawing Processes In Soilsupporting

confidence: 83%

Unified Soil Freezing Characteristic for Variably‐Saturated Saline Soils

Zhou

Meng

Wei

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

Freezing and thawing of soil have a significant impact on the thermal‐hydrological processes in cold regions. Freezing point and unfrozen water content are key variables characterizing the freezing and thawing of soil. The testing results show that unfrozen water content is independent of initial water content in nonsaline silt, based on which a unified model for a soil freezing characteristic curve (SFCC) is presented. The SFCC describes the relationship between unfrozen water content and temperature for variably‐saturated saline soils. It is also demonstrated that the SFCC can describe the variation of soil freezing point with respect to water content and salt concentration. The SFCC model is validated by comparing the experimental data to the calculated result for unfrozen water content and freezing point in a variety of conditions. When the initial water content varies in saline soil, the SFCC changes with constant initial salt concentration, but it remains unchanged with constant gravimetric salt content. The SFCC model can be adopted in the simulation of thermal‐hydrological processes in arid and saline lands and coastal regions. Finally, an example of thermal‐hydrological processes was simulated with the new SFCC model to show the application of the presented model.

show abstract

Section: Theory Of Freezing and Thawing Processes In Soilsupporting

confidence: 83%

Unified Soil Freezing Characteristic for Variably‐Saturated Saline Soils

Zhou

Meng

Wei

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…The type of membership function and number of membership functions are decided based on the variation trend of the input and output data. The types of membership functions in the ANFIS model are triangular, trapezoidal, gbell, gauss, gauss2, pi, dsig and psig [32]. In the ANFIS model, the input and output data are connected by rules with the statements by showing the relationship between the input and output data.…”

Section: Adaptive Neuro-fuzzy Interface System Modelling (Anfis)mentioning

confidence: 99%

Artificial Neural Network and Adaptive Neuro-Fuzzy Interface System Modelling to Predict Thermal Performances of Thermoelectric Generator for Waste Heat Recovery

et al. 2020

View full text Add to dashboard Cite

The present study elaborates the suitability of the artificial neural network (ANN) and adaptive neuro-fuzzy interface system (ANFIS) to predict the thermal performances of the thermoelectric generator system for waste heat recovery. Six ANN models and seven ANFIS models are formulated by considering hot gas temperatures and voltage load conditions as the inputs to predict current, power, and thermal efficiency of the thermoelectric generator system for waste heat recovery. The ANN model with the back-propagation algorithm, the Levenberg–Marquardt variant, Tan-Sigmoidal transfer function and 25 number of hidden neurons is found to be an optimum model to accurately predict current, power and thermal efficiency. For current, power and thermal efficiency, the ANFIS model with pi-5 or gauss-5-membership function is recommended as the optimum model when the prediction accuracy is important while the ANFIS model with gbell-3-membership function is suggested as the optimum model when the prediction cost plays a crucial role along with the prediction accuracy. The proposed optimal ANN and ANFIS models present higher prediction accuracy than the coupled numerical approach.

show abstract

“…Shang and Mao (2001) proposed a model based on backpropagation neural network (BPNN) to predict the empirical parameters of the SFCC of Morin Clay under different initial water content, dry density and NaCl concentration. Based on experimental data obtained by nuclear magnetic resonance, Liu et al (2018) constructed two models using adaptive network fuzzy inference system (ANFIS) and BPNN to predict the UWC of saline soil. Wang Q et al (2020) proposed a new model to predict the UWC of saline soil based on the combined weighting method and ANFIS.…”

Section: Introductionmentioning

confidence: 99%

Unfrozen Water Content Estimation: A Comparison between Ensemble and Non-ensemble Machine Learning Models

Li,

Zhou,

Ren

et al. 2024

Preprint

View full text Add to dashboard Cite

Unfrozen water content (UWC) is a key parameter affecting a variety of soil physical-mechanical properties and processes in frozen soil systems. However, traditional estimation models suffer limitations due to oversimplified assumptions or limited applicable conditions. Given that, there is a compelling need to explore alternative modeling approaches that leverage machine learning (ML) algorithms, which have shown increasing potential in engineering fields. To this end, this study evaluated and compared six widely known ML algorithms (i.e., three ensemble models: RF, LightGBM and XGBoost; and three non-ensemble models: KNN, SVR and BPNN) for modeling UWC based on collected experimental datasets. These algorithms were optimized and evaluated using a framework combining Bayesian optimization and cross-validation to ensure model stability and generalization. The results demonstrated that the ensemble tree-based methods, particularly LightGBM and XGBoost, achieved the highest predictive accuracy and superior overall performance. On the other hand, the nonensemble methods exhibited poorer generalization abilities. Interestingly, during 10-fold cross-validation, consistent underperformance was observed for a particular fold, possibly stemming from the challenges of the data distribution in that fold after random shuffling. The present study highlights the effectiveness of ensemble learning approaches, importance of proper hyperparameter tuning and validation strategies, and intrinsic modeling challenges arising from the difference between the freezing and thawing phase change behaviors. This comprehensive ML model comparison and robust training framework provide valuable guidance on selecting suitable data-driven techniques for modeling frozen soil properties for cold regions hydrogeology and engineering practices.

show abstract

Using ANFIS and BPNN Methods to Predict the Unfrozen Water Content of Saline Soil in Western Jilin, China

Cited by 56 publications

References 28 publications

Unified Soil Freezing Characteristic for Variably‐Saturated Saline Soils

Unified Soil Freezing Characteristic for Variably‐Saturated Saline Soils

Artificial Neural Network and Adaptive Neuro-Fuzzy Interface System Modelling to Predict Thermal Performances of Thermoelectric Generator for Waste Heat Recovery

Unfrozen Water Content Estimation: A Comparison between Ensemble and Non-ensemble Machine Learning Models

Contact Info

Product

Resources

About