Improved Prediction of Frost Depth Penetration Using Recurrent Neural Networks
Publication: Cold Regions Engineering 2024: Sustainable and Resilient Engineering Solutions for Changing Cold Regions
ABSTRACT
Frost effects, such as frost heave and thaw weakening, can significantly degrade pavements. Effective mitigation relies on accurate prediction of frost penetration depth. Current prediction methods use empirical equations or finite element analysis, with recent progress using machine learning. One potentially appropriate machine learning model may be a recurrent neural network, which takes in present data as input and outputs an estimation of future data, which can then be fed back into the model recursively to make further predictions. Using this method and training data from Hill Air Force Base, Utah, and Air Force Academy, Colorado, we were able to forecast soil parameters including temperature, thermal conductivity, and moisture content for frost susceptible soils, with a deviation from experimental values of no more than 10%, with the most significant contributions to accuracy being the use of gated recurrent unit neurons and the incorporation of multiple soil parameters.
Get full access to this chapter
View all available purchase options and get full access to this chapter.
REFERENCES
Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., . . . Farhan, L. (2021). Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions. Journal of Big Data, 8.
Barooni, M., Ziarati, K., & Barooni, A. (2023). Frost Prediction Using Machine Learning Methods in Fars Province. 28th International Computer Conference. Tehran: Computer Society of Iran.
Bianchini, A., & Gonzalez, C. R. (2012). Pavement-Transportation Computer Assisted Structural Engineering (PCASE) Implementation of the Modified Berggren (ModBerg) Equation for Computing the Frost Penetration Depth within Pavement Structures. Vicksburg: US Army Corps of Engineers.
Choi, H.-J., Kim, S., Kim, Y., & Won, J. (2022). Predicting Frost Depth of Soils in South Korea Using Machine Learning Techniques. Sustainability (14), 9767.
Dayarathne, R., Hawlader, B., Phillips, R., & Robert, D. (2023). Two-dimensional finite element modeling of long-term frost heave beneath chilled gas pipelines. Cold Regions Science and Technology, 208.
Lein, A. W., Slone, S. L., Smith Jr., C. E., Bernier, A. P., & Oren, J. I. (2019). Frost-Depth Penetration and Frost Heave in Frost-Susceptible Soils. Hanover: US Army Corps of Engineers.
Pollard, W. H. (2017). Chapter 15 - Periglacial Processes in Glacial Environments. In J. Menzies, & J. J. van der Meer, Past Glacial Environments (pp. 537-564).Elsevier.
Rajaei, P., & Baladi, G. Y. (2015). Frost Depth: General Prediction Model. Transportation Research Record: Journal of the Transportation Research Board, 2510, 74-80.
Talsma, C. J., Solander, K. C., Mudunuru, M. K., Crawford, B., & Powell, M. R. (2023). Frost prediction using machine learning and deep neural network models. Frontiers in Artificial Intelligence.
Zaytar, M. A., & Amrani, C. E. (2016). Sequence to Sequence Weather Forecasting with Long Short-Term Memory Recurrent Neural Networks. International Journal of Computer Applications, 143(11), 0975-8887.
Information & Authors
Information
Published In
History
Published online: May 9, 2024
ASCE Technical Topics:
- Artificial intelligence and machine learning
- Cold regions engineering
- Computer programming
- Computing in civil engineering
- Engineering fundamentals
- Finite element method
- Freeze and thaw
- Frost
- Geomechanics
- Geotechnical engineering
- Methodology (by type)
- Neural networks
- Numerical methods
- Soil analysis
- Soil mechanics
- Soil properties
- Soil water
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.