AI-driven spectral analysis of soil heaving for automated surveys in rail transport infrastructure

Abstract

The expansion of rail transport infrastructures necessitates accurate and efficient soil surveys to ensure long-term stability and performance, particularly in regions prone to soil heaving. This study aimed to demonstrate the potential of non-destructive spectral analysis combined with Agentic Artificial Intelligence for automating the identification of soil heaving potential, providing a transformative approach to soil assessment in railway construction. A robust AI-agent was developed to predict soil heaving potential across temperature regimes (ranging from 0°C to -5°C and back), enabling characterization of the relative acoustic compressibility coefficient (β) based on the physical and mechanical properties of the soil. The main objective was to develop a framework that integrated spectral reflectance data with machine learning algorithms to predict soil heaving potential and reduce the reliance on traditional invasive methods. The experimental setup employed digital techniques to process and record longitudinal and transverse acoustic pulse signals reflected from piezoelectric sensors mounted on soil specimens. The processed signals were automatically transferred via a USB adapter to a PC for further analysis by the AI-agent. Acoustic diagnostics of the soils were performed using Fast-Fourier Transform (FFT) Spectral Analysis, followed by correlation of waveform spectra with heaving deformation. The AI-agent utilized a hybrid architecture combining Convolutional Neural Network (CNN), Support Vector Machine (SVM), and Random Forest (RF) algorithms to address the complexities of heterogeneous soil data and multifaceted prediction tasks—including heaving classification and deformation regression—while mitigating overfitting. Soil heaving potential was accurately predicted by the AI agent, with minor variations attributed to equipment sensitivity.

Keywords

Automated soil heaving control / Non-Destructive Spectral Analysis (NDSA) / Fast Fourier Transform (FFT) / AI agent integration / Rail transport infrastructure / Agent-to-Agent protocol (A2A) / Agentic Artificial Intelligence (AgenAI)

References

[1] Abdellatif AD, Abou-Kota MES, Gainzour SK, Allam AS. Soil spectral reflectance behavior related to chemical soil properties and macronutrients using the PLSR model. International Journal of Agricultural Science and Statistical Sciences, 2022, 17(1): 309-316

[2] ASTM D4221-99. (2007). Standard Test Method for Particle-Size Analysis of Soils. ASTM International, West Conshohocken, PA, USA.

[3] ASTM D854-14. (2014). Standard Test Methods for Specific Gravity of Soil Solids by Water Pycnometer. ASTM International, West Conshohocken, PA, USA. https://doi.org/10.1520/D0854-14

[4] ASTM D2845-08. (2015). Standard Test Method for Laboratory Determination of Pulse Velocities and Ultrasonic Elastic Constants of Rock. ASTM International, West Conshohocken, PA, USA.

[5] ASTM D2487-17. (2017). Standard Practice for Classification of Soils for Engineering Purposes (Unified Soil Classification System). ASTM International, West Conshohocken, PA, USA. https://doi.org/10.1520/D2487-17

[6] ASTM D4318-17e1. (2017). Standard Test Methods for Liquid Limit, Plastic Limit, and Plasticity Index of Soils. ASTM International, West Conshohocken, PA, USA. https://doi.org/10.1520/D4318-17E01

[7] ASTM D4428/D4428M-07. (2017). Standard Test Methods for Crosshole Seismic Testing. ASTM International, West Conshohocken, PA, USA.

[8] ASTM D2216-19. (2019). Standard Test Methods for Laboratory Determination of Water (Moisture) Content of Soil and Rock by Mass. ASTM International, West Conshohocken, PA, USA. https://doi.org/10.1520/D2216-19

[9] ASTM D7263-21. (2021). Standard Test Methods for Laboratory Determination of Density (Unit Weight) of Soil Specimens. ASTM International, West Conshohocken, PA, USA. https://doi.org/10.1520/D7263-21

[10] ASTM E494-15. (2015). Standard Practice for Measuring Ultrasonic Velocity in Materials. ASTM International, West Conshohocken, PA, USA. https://doi.org/10.1520/E0494-15

[11] ASTM E317-16. (2016). Standard Practice for Evaluating Performance Characteristics of Ultrasonic Pulse-Echo Testing Instruments and Systems Without the Use of Electronic Measurement Instruments. ASTM International, West Conshohocken, PA, USA. https://doi.org/10.1520/E0317-16

[12] Behrens T, Rossel R, Kerry R, MacMillan R, Schmidt K, Lee J, Zhu A. The relevant range of scales for multi-scale contextual spatial modelling. Scientific Reports, 2019.

[13] Blasch G, Spengler D, Hohmann C, Neumann C, Itzerott S, Kaufmann H. Multitemporal soil pattern analysis with multispectral remote sensing data at the field-scale. Computers and Electronics in Agriculture, 2015, 113: 1-13.

[14] Bogrekci I, Lee WS. Effects of soil moisture content on absorbance spectra of sandy soils in sensing phosphorus concentrations using UV-VIS-NIR spectroscopy. Transactions of the ASABE, 2013, 49(4): 1175-1180.

[15] Cao J, Guo Z, Lv Y, Xu M, Huang C, Liang H. Pollution risk prediction for cadmium in soil from an abandoned mine based on random forest model. International Journal of Environmental Research and Public Health, 2023, 20(6): 5097.

[16] Fan P, Zhang X, Xu L, Shi Y, Shi C, Cheng X. Collapsible characteristics model of remodeled loess. Research Square, 2020.

[17] Gredilla A, Fdez-Ortiz de Vallejuelo S, Elejoste N, de Diego A, Madariaga JM. Non-destructive Spectroscopy combined with chemometrics as a tool for Green Chemical Analysis of environmental samples: A review. TrAC Trends in Analytical Chemistry, 2016, 76: 30-39.

[18] Kovačević M, Bajat B, Gajić B. Soil type classification and estimation of soil properties using support vector machines. Geoderma, 2010, 154(3–4): 340-347.

[19] Li B, Zixuan Y, Kaiwei N, Yuexiang W. Prediction of soil compaction parameters using machine learning models. Applied Sciences, 2024, 14(7): 2716.

[20] Liu J, Qinhuo L. Soil moisture estimate uncertainties from the effect of soil texture on dielectric semiempirical models. Remote Sensing, 2020, 12(14): 2343.

[21] Mao W, Hei L, Yang Y. Advances on the acoustic emission testing for monitoring of granular soils. Measurement, 2021, 185: 110110.

[22] Marçal M, Souza Z, Tavares R, Farhate C, Oliveira S, Galindo F. Predictive models to estimate carbon stocks in agroforestry systems. Forests, 2021, 12(9): 1240.

[23] Max, J., & Lacoume, J.-L. (1996). Méthodes et techniques de traitement du signal et applications aux mesures physiques. Tome 1, Principes généraux et méthodes classiques. Masson, Paris.

[24] Padarian J, Minasny B, McBratney AB. Using deep learning to predict soil properties from regional spectral data. Geoderma Regional, 2019, 16: e00198.

[25] Patel R, Patel D, Meshram D. A review on non-destructive testing (NDT) techniques: Advances, researches and applicability. International Journal of Current Science Research and Review, 2022, 5(4): 1342-1355.

[26] GOST 28622–2012. (2013). Soils: Laboratory method for determination of frost-heave degree. Interstate Council for Standardization, Metrology and Certification (ISC).

[27] U.S. Geological Survey. (2013). Costs associated with expansive soils in the United States. Retrieved September 28, 2024, from https://pubs.usgs.gov.

[28] Volkov DS, Rogova OB, Proskurnin MA. Photoacoustic and photothermal methods in spectroscopy and characterization of soils and soil organic matter. Photoacoustics, 2020, 17: 100151.

[29] Wang Z, Chen S, Lu R, Zhang X, Ma Y, Shi Z. Non-linear memory-based learning for predicting soil properties using a regional vis-NIR spectral library. Geoderma, 2024, 441: 116752.

[30] Yang Y. et al.. Correlation analysis and prediction of the physical and mechanical properties of coastal soft soil in the Jiangdong New District, Haikou, China. Advances in Civil Engineering, 2024, 2024: 1-14.

[31] Zarnaq MH, Omid M, Sarmadian F, Ghasemi-Mobtaker H. A CNN model for predicting soil properties using VIS-NIR spectral data. Environmental Earth Sciences, 2023.

[32] Zhu AX, Liu J, Du F, Zhang SJ, Qin CZ, Burt J, Behrens T, Scholten T. Predictive soil mapping with limited sample data. European Journal of Soil Science, 2015, 66(3): 535-547.

RIGHTS & PERMISSIONS

The Author(s)