Adaptive Bayesian inversion of pore water pressures based on artificial neural network: An earth dam case study

Abstract

Most earth-dam failures are mainly due to seepage, and an accurate assessment of the permeability coefficient provides an indication to avoid a disaster. Parametric uncertainties are encountered in the seepage analysis, and may be reduced by an inverse procedure that calibrates the simulation results to observations on the real system being simulated. This work proposes an adaptive Bayesian inversion method solved using artificial neural network (ANN) based Markov Chain Monte Carlo simulation. The optimized surrogate model achieves a coefficient of determination at 0.98 by ANN with 247 samples, whereby the computational workload can be greatly reduced. It is also significant to balance the accuracy and efficiency of the ANN model by adaptively updating the sample database. The enrichment samples are obtained from the posterior distribution after iteration, which allows a more accurate and rapid manner to the target posterior. The method was then applied to the hydraulic analysis of an earth dam. After calibrating the global permeability coefficient of the earth dam with the pore water pressure at the downstream unsaturated location, it was validated by the pore water pressure monitoring values at the upstream saturated location. In addition, the uncertainty in the permeability coefficient was reduced, from 0.5 to 0.05. It is shown that the provision of adequate prior information is valuable for improving the efficiency of the Bayesian inversion.

References

[1] Talukdar P, Dey A. Hydraulic failures of earthen dams and embankments [J]. Innovative Infrastructure Solutions, 2019, 4(1): 42

[2] Zhang L M, Xu Y, Jia J S. Analysis of earth dam failures: A database approach [J]. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 2009, 3(3): 184-189

[3] Man J-H, Huang H-W, Ai Z-Y, et al.. Stability of complex rock tunnel face under seepage flow conditions using a novel equivalent analytical model [J]. International Journal of Rock Mechanics and Mining Sciences, 2023, 170: 105427

[4] Pan Q-J, Qu X-R, Liu L-L, et al.. A sequential sparse polynomial chaos expansion using Bayesian regression for geotechnical reliability estimations [J]. International Journal for Numerical and Analytical Methods in Geomechanics, 2020, 44(6): 874-889

[5] Sun Y, Huang J-S, Jin W, et al.. Bayesian updating for progressive excavation of high rock slopes using multi-type monitoring data [J]. Engineering Geology, 2019, 252: 1-13

[6] Zhang A-L, Xie H-P, Zhang R, et al.. Mechanical properties and energy characteristics of coal at different depths under cyclic triaxial loading and unloading [J]. International Journal of Rock Mechanics and Mining Sciences, 2023, 161: 105271

[7] Hu J-Z, Zheng J-G, Zhang J, et al.. Bayesian framework for assessing effectiveness of geotechnical site investigation programs [J]. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, 2023, 9(1): 04022054

[8] Qi C-C, Fourie A. A real-time back-analysis technique to infer rheological parameters from field monitoring [J]. Rock Mechanics and Rock Engineering, 2018, 51(10): 3029-3043

[9] Chen S-K, Liu X-F. An investigation of PSO algorithm-based back analysis on the three-dimensional seepage characteristics of an earth dam [J]. Indian Geotechnical Journal, 2019, 49(2): 232-240

[10] Wu C-Z, Hong Y, Chen Q-S, et al.. A modified optimization algorithm for back analysis of properties for coupled stress-seepage field problems [J]. Tunnelling and Underground Space Technology, 2019, 94: 103040

[11] Jiang S-H, Huang J-S, Qi X-H, et al.. Efficient probabilistic back analysis of spatially varying soil parameters for slope reliability assessment [J]. Engineering Geology, 2020, 271: 105597

[12] Rossat D, Baroth J, Briffaut M, et al.. Bayesian updating for nuclear containment buildings using both mechanical and hydraulic monitoring data [J]. Engineering Structures, 2022, 262: 114294

[13] Kim B, Lee J, Park K, et al.. Characterizing coefficient of permeability based on response of groundwater level to river stage using regional database [J]. Environmental Earth Sciences, 2024, 83(3): 88

[14] Cividini A, Jurina L, Gioda G. Some aspects of “characterization” problems in geomechanics [J]. International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 1981, 18(6): 487-503

[15] Pichler B, Lackner R, Mang H A. Back analysis of model parameters in geotechnical engineering by means of soft computing [J]. International Journal for Numerical Methods in Engineering, 2003, 57(14): 1943-1978

[16] Mei X-C, Li C-Q, Cui Z, et al.. Application of metaheuristic optimization algorithms-based three strategies in predicting the energy absorption property of a novel aseismic concrete material [J]. Soil Dynamics and Earthquake Engineering, 2023, 173: 108085

[17] Kang F, Li J-J, Xu Q. Structural inverse analysis by hybrid simplex artificial bee colony algorithms [J]. Computers & Structures, 2009, 87(1314): 861-870

[18] Vaezinejad S, Marandi S, Salajegheh E. A hybrid of artificial neural networks and particle swarm optimization algorithm for inverse modeling of leakage in earth dams [J]. Civil Engineering Journal, 2019, 5(9): 2041-2057

[19] Shu Y-K, Shen Z-Z, Xu L-Q, et al.. Inversion analysis of impervious curtain permeability coefficient using calcium leaching model, extreme learning machine, and optimization algorithms [J]. Applied Sciences, 2022, 12(7): 3272

[20] Tong F, Zhang A-N, Yang J, et al.. Inversion analysis of rock mass permeability coefficient of dam engineering based on particle swarm optimization and support vector machine: A case study [J]. Measurement, 2023, 221: 113580

[21] Pan Q J, Li X Z, Wang S Y, et al.. Data-driven estimations of ground deformations induced by tunneling: A Bayesian perspective [J]. Acta Geotechnica, 2024, 19(1): 475-493

[22] Tao M-L, Yin C-C, Liu Y-H, et al.. Transdimensional Bayesian inversion for airborne EM data in sparse domain [J]. Journal of Applied Geophysics, 2021, 189: 104317

[23] Zhang Z-L, Zhang T-T, Li X-Z, et al.. Bayesian ensemble methods for predicting ground deformation due to tunnelling with sparse monitoring data [J]. Underground Space, 2024, 16: 79-93

[24] Zheng G-L, Wang Y, Wang F, et al.. Stochastic back analysis of permeability coefficient using generalized Bayesian method [J]. Water Science and Engineering, 2008, 1(3): 83-92

[25] Kao C Y, Loh C H. Monitoring of long-term static deformation data of Fei-Tsui arch dam using artificial neural network-based approaches [J]. Structural Control and Health Monitoring, 2013, 20(3): 282-303

[26] Ranković V, Grujović N, Divac D, et al.. Development of support vector regression identification model for prediction of dam structural behaviour [J]. Structural Safety, 2014, 48: 33-39

[27] Li B, Yang J, Hu D-X. Dam monitoring data analysis methods: A literature review [J]. Structural Control and Health Monitoring, 2020, 27(3): e2501

[28] Li C-Q, Zhou J, Du K, et al.. Stability prediction of hard rock pillar using support vector machine optimized by three metaheuristic algorithms [J]. International Journal of Mining Science and Technology, 2023, 33(8): 1019-1036

[29] Pan Q-J, Zhang R-F, Ye X-Y, et al.. An efficient method combining polynomial-chaos Kriging and adaptive radial-based importance sampling for reliability analysis [J]. Computers and Geotechnics, 2021, 140: 104434

[30] Pan Q-J, Dias D. An efficient reliability method combining adaptive support vector machine and Monte Carlo simulation [J]. Structural Safety, 2017, 67: 85-95

[31] Li C-Q, Mei X-C. Application of SVR models built with AOA and Chaos mapping for predicting tunnel crown displacement induced by blasting excavation [J]. Applied Soft Computing, 2023, 147: 110808

[32] Yan L, Zhou T. An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems [J]. Communications in Computational Physics, 2020, 28(5): 2180-2205

[33] Raissi M, Perdikaris P, Karniadakis G E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations [J]. Journal of Computational Physics, 2019, 378: 686-707

[34] Li W-X, Lin G. An adaptive importance sampling algorithm for Bayesian inversion with multimodal distributions [J]. Journal of Computational Physics, 2015, 294: 173-190

[35] Rossat D, Baroth J, Briffaut M, et al.. Bayesian inversion using adaptive Polynomial Chaos Kriging within Subset Simulation [J]. Journal of Computational Physics, 2022, 455: 110986

[36] Robert C P. The Bayesian choice: From decision-theoretic foundations to computational implementation [M], 2001 2nd ed New York, Springer

[37] Yang H-Q, Zhang L-L, Xue J-F, et al.. Unsaturated soil slope characterization with Karhunen-Loève and polynomial chaos via Bayesian approach [J]. Engineering with Computers, 2019, 35(1): 337-350

[38] Metropolis N, Ulam S. The Monte Carlo method [J]. Journal of the American Statistical Association, 1949, 44(247): 335-341

[39] Goodman J, Weare J. Ensemble samplers with affine invariance [J]. Communications in Applied Mathematics and Computational Science, 2010, 5(1): 65-80

[40] Lye A, Cicirello A, Patelli E. An efficient and robust sampler for Bayesian inference: Transitional ensemble Markov chain Monte Carlo [J]. Mechanical Systems and Signal Processing, 2022, 167: 108471

[41] Gordan B, Jahed A D, Hajihassani M, et al.. Prediction of seismic slope stability through combination of particle swarm optimization and neural network [J]. Engineering with Computers, 2016, 32(1): 85-97

[42] Sasmal S K, Behera R N. Prediction of combined static and cyclic load-induced settlement of shallow strip footing on granular soil using artificial neural network [J]. International Journal of Geotechnical Engineering, 2021, 15(7): 834-844

[43] Yan H, Zhang J-X, Zhou N, et al.. Application of hybrid artificial intelligence model to predict coal strength alteration during CO₂ geological sequestration in coal seams [J]. Science of the Total Environment, 2020, 711: 135029

[44] Salazar F, Morán R, Toledo M Á, et al.. Data-based models for the prediction of dam behaviour: A review and some methodological considerations [J]. Archives of Computational Methods in Engineering, 2017, 24(1): 1-21

[45] Li C-Q, Zhou J, Armaghani D J, et al.. Stability analysis of underground mine hard rock pillars via combination of finite difference methods, neural networks, and Monte Carlo simulation techniques [J]. Underground Space, 2021, 6(4): 379-395

[46] Rosenblatt F. The perceptron: A probabilistic model for information storage and organization in the brain [J]. Psychological Review, 1958, 65(6): 386-408

[47] Wei X, Zhang L-L, Yang H-Q, et al.. Machine learning for pore-water pressure time-series prediction: Application of recurrent neural networks [J]. Geoscience Frontiers, 2021, 12(1): 453-467

[48] Press W H. Numerical recipes: The art of scientific computing [M], 2007 3rd ed Cambridge, UK, Cambridge University Press

[49] Rogers S, Girolami M. A first course in machine learning [M], 2016, New York, Chapman and Hall/CRC

[50] Hagan M T, Demuth H B, Beale M. Neural network design [M], 1997, Stillwater, US, PWS Publishing Co.

[51] Zhang Z-L, Pan Q-J, Yang Z-H, et al.. Physics-informed deep learning method for predicting tunnelling-induced ground deformations [J]. Acta Geotechnica, 2023, 18(9): 4957-4972

[52] Zhang J-M, An L, Li C-Q, et al.. Artificial neural network response assessment of a single footing on soft soil reinforced by rigid inclusions [J]. Engineering Structures, 2023, 281: 115753

[53] Salazar F, Morán R, Toledo M Á, et al.. Data-based models for the prediction of dam behaviour: A review and some methodological considerations [J]. Archives of Computational Methods in Engineering, 2017, 24(1): 1-21

[54] Lecun Y, Bengio Y, Hinton G. Deep learning [J]. Nature, 2015, 521(7553): 436-444

[55] Brinkgreve R, Kumarswamy S, Swolfs W, et al.. PLAXIS 2016 [M], 2016, the Netherlands, PLAXIS Bv

[56] Pan H-Y, Li H-N, Li C. Seismic fragility analysis of free-spanning submarine pipelines incorporating soil spatial variability in soil-pipe interaction and offshore motion propagation [J]. Engineering Structures, 2023, 280: 115639

[57] Mei X-C, Sheng Q, Cui Z, et al.. Experimental investigation on the mechanical and damping properties of rubber-sand-concrete prepared with recycled waste tires for aseismic isolation layer [J]. Soil Dynamics and Earthquake Engineering, 2023, 165: 107718

[58] Cho S E. Probabilistic analysis of seepage that considers the spatial variability of permeability for an embankment on soil foundation [J]. Engineering Geology, 2012, 133: 30-39

[59] Mouyeaux A, Carvajal C, Bressolette P, et al.. Probabilistic analysis of pore water pressures of an earth dam using a random finite element approach based on field data [J]. Engineering Geology, 2019, 259: 105190

[60] Schobi R, Sudret B, Wiart J. Polynomial-chaos-based Kriging [J]. International Journal for Uncertainty Quantification, 2015, 5(2): 171-193

[61] Zhang T-T, Baroth J, Dias D. Deterministic and probabilistic basal heave stability analysis of circular shafts against hydraulic uplift [J]. Computers and Geotechnics, 2022, 150: 104922

[62] Mei X-C, Sheng Q, Chen J, et al.. Aseismic performances of constrained damping lining structures made of rubber-sand-concrete [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2024, 16(5): 1522-1537

RIGHTS & PERMISSIONS

Central South University