Practice of Artificial Intelligence in Geotechnics

Introduction

We have extensively performed the application of AI in geotechnics due to the strong capacity of solving non-linear and high-dimensional problem of AI. For example, the optimization and Bayesian-based methods can bridge the gap between advanced constitutive theories and engineering practice; ML-based surrogate model can also be applied in engineering practice as the alternative to experimental and numerical methods, saving expenses for engineering design. Currently, we focuses on the following four topics.


Topic 1: Optimization-based parameters identification and model selection

We enhanced several optimization algorithms (DE, GA…), applied to the parameters identification and model selection based on laboratory tests or engineering measurements, and applied to excavation project with intelligent updating of the soil model with parameters.


  • Jin Y-F, Yin Z-Y. Enhancement of backtracking search algorithm for identifying soil parameters[J]. International Journal for Numerical and Analytical Methods in Geomechanics, 2020, 44(9): 1239-1261.
  • Jin Y-F, Yin Z-Y, Zhou W-H, Liu X. Intelligent model selection with updating parameters during staged excavation using optimization method[J]. Acta Geotechnica, 2020, 15(9): 2473-2491.
  • Jin Y-F, Yin Z-Y, Zhou W-H, Horpibulsuk S. Identifying parameters of advanced soil models using an enhanced transitional Markov chain Monte Carlo method[J]. Acta Geotechnica, 2019, 14(6): 1925-1947
  • Jin Y-F, Yin Z-Y, Zhou W-H, Huang H-W. Multi-objective optimization-based updating of predictions during excavation[J]. Engineering Applications of Artificial Intelligence, 2019, 78: 102-123.
  • Jin Y-F, Yin Z-Y, Wu Z-X, Zhou W-H. Identifying parameters of easily crushable sand and application to offshore pile driving[J]. Ocean Engineering, 2018, 154: 416-429.
  • Yin Z-Y, Jin Y-F, Shen J S, Hicher P-Y. Optimization techniques for identifying soil parameters in geotechnical engineering: Comparative study and enhancement[J]. International Journal for Numerical and Analytical Methods in Geomechanics, 2018, 42(1): 70-94.
  • Yin Z-Y, Jin Y-F, Shen S-L, Huang H-W. An efficient optimization method for identifying parameters of soft structured clay by an enhanced genetic algorithm and elastic–viscoplastic model[J]. Acta Geotechnica, 2017, 12(4): 849-867.
  • Jin Y-F, Wu Z-X, Yin Z-Y, Shen J S. Estimation of critical state-related formula in advanced constitutive modeling of granular material[J]. Acta Geotechnica, 2017, 12(6): 1329-1351.
  • Jin Y-F, Yin Z-Y, Shen S-L, Hicher P-Y. Selection of sand models and identification of parameters using an enhanced genetic algorithm[J]. International Journal for Numerical and Analytical Methods in Geomechanics, 2016, 40(8): 1219-1240.
  • Jin Y-F, Yin Z-Y, Shen S-L, Hicher P-Y. Investigation into MOGA for identifying parameters of a critical-state-based sand model and parameters correlation by factor analysis[J]. Acta Geotechnica, 2016, 11(5): 1131-1145.


Topic 2: Bayesian-based parameters identification and model class selection

We enhanced original TMCMC using DE algorithm, and applied the Parallel-based DE-TMCMC for parameter identification and model class selection.



  • Jin Y-F, Yin Z-Y, Zhou W-H, Horpibulsuk S. Identifying parameters of advanced soil models using an enhanced transitional Markov chain Monte Carlo method[J]. Acta Geotechnica, 2019, 14(6): 1925-1947
  • Jin Y-F, Yin Z-Y, Zhou W-H, Shao J-F. Bayesian model selection for sand with generalization ability evaluation[J]. International Journal for Numerical and Analytical Methods in Geomechanics, 2019, 43(14): 2305-2327.


Topic 3: Deep learning based constitutive modelling of soils and SSI

Machine learning (ML) has been back on the stage of research works in all the walks in recent due to its excellent capacity of solving nonlinear problems with desired speed and accuracy. ML can directly learn from the raw data without any assumptions and the application scopes increase with the increasing number of data, thus provides a new methodology to model the complicated mechanical properties and behaviors of soils and may develop a unified model.
We developed long short-term memory (LSTM) neural network charactered by time sequence prediction based constitutive model to capture path-dependent soil behaviours, and developed data-driven surrogate models using deep learning to capture the responses of foundation in soils, such as failure envelope and force-displacement responses of caisson foundation



  • Zhang P, Yin ZY, Jin YF, 2021. State-of-the-Art Review of Machine Learning Applications in Constitutive Modeling of Soils. Archives of Computational Methods in Engineering, https://doi.org/10.1007/s11831-020-09524-z
  • Zhang P, Yin ZY, Jin YF, Ye GL, 2020. An AI-based model for describing cyclic characteristics of granular materials. International Journal for Numerical and Analytical Methods in Geomechanics, 44, 9: 1315-1335
  • Zhang P, Yin ZY, Zheng YY, Gao FP, 2020. A LSTM Surrogate Modelling Approach for Caisson Foundations. Ocean Engineering, 204, 107263
  • Zhang P, Jin YF, Yin ZY, Yang Y, 2020. Random forest based artificial intelligent model for predicting failure envelopes of caisson foundations in sand. Applied Ocean Research, 101, 102223


Topic 4: Machine learning-based correlations for soil properties

We adopted a series of machine learning algorithms (e.g. GP = genetic programming; EPR = evolutionary polynominal regression; SVR = support vector regression; RF = random forest; BPNN = backpropagation neural network; ANN_MCD = artificial neural network and Monte Carlo dropout) to propose prediction models for soil properties (e.g. compressibility index of clay, creep index of clay, undrained shear strength of clay, hydraulic conductivity of clay, soil compaction parameters, air-entry value of compacted soils).




  • Zhang P, Yin ZY, Jin YF, Chan THT, Gao FP, 2021. Intelligent Modelling of Clay Compressibility using Hybrid Meta-Heuristic and Machine Learning Algorithms. Geoscience Frontiers, 12, 1, 441-452
  • Zhang P, Yin ZY, Jin YF, Chan THT, 2020 A Novel Hybrid Surrogate Intelligent Model for Creep Index Prediction based on Particle Swarm Optimization and Random Forest. Engineering Geology. 265, 105328
  • Wang HL#, Yin Z-Y*, Zhang P, Jin YF (2020). Straightforward prediction for air-entry value of compacted soils using machine learning algorithms. Eng. Geol., 279: 105911.
  • Wang HL#, Yin Z-Y* (2020). High performance prediction of soil compaction parameters using multi expression programming. Eng. Geol., 276: 105758.
  • Jin Y-F, Yin Z-Y, Zhou W-H, Yin J-H, Shao J-F. A single-objective EPR based model for creep index of soft clays considering L2 regularization[J]. Engineering Geology, 2019, 248: 242-255.
  • Jin Y-F, Yin Z-Y. An intelligent multi-objective EPR technique with multi-step model selection for correlations of soil properties[J]. Acta Geotechnica, 2020, 15(8): 2053-2073.
  • Yin Z-Y, Jin Y-F, Huang H-W, Shen S-L. Evolutionary polynomial regression based modelling of clay compressibility using an enhanced hybrid real-coded genetic algorithm[J]. Engineering Geology, 2016, 210: 158-167.