Long-term dam deformation is governed by coupled hydrostatic, thermal and aging effects, producing noisy, nonlinear and temporally dependent monitoring series that are difficult to model with conventional statistical or single machine/deep learning approaches. This study proposes an interpretable hybrid stacking framework, RF–DEGWO–LSTM, for uncertainty-aware dam deformation forecasting. Random forest (RF) acts as a nonlinear feature-based learner and provides embedded feature importance; differential evolution grey wolf optimizer (DEGWO) is used to tune key long short-term memory (LSTM) hyperparameters; and a meta-learner fuses RF and DEGWO–LSTM outputs into a unified predictor that explicitly captures time-varying, non-stationary relationships between hydrostatic, thermal and aging factors and displacement. The framework is validated using long-term monitoring data from two dams (a concrete gravity dam and a run-of-the-river dam). Comparative experiments against RF, plain LSTM, recurrent neural networks (RNNs), eXtreme Gradient Boosting (XGBoost) and DEGWO–LSTM show that the RF–DEGWO–LSTM model consistently achieves the highest R2 and the lowest error metrics. In the most challenging scenario, RF–DEGWO–LSTM improves test-set R2 by approximately 5–10% and reduces root mean square error by about 15–20% relative to the best single-model baseline. Residual analyses indicate reduced bias and variance, while interpretability based on RF feature importance and LSTM attention confirms the dominant roles of hydrostatic and seasonal components, with aging effects being secondary. Accurate forecasting of deformation is critical for early identification of damage evolution and integrity loss in large civil infrastructure. The proposed framework thus offers an accurate, robust and interpretable tool for data-driven dam deformation forecasting and early-warning-oriented structural health management.
AdusumilliSBhattDWangH, et al. (2015) A novel hybrid approach utilizing principal component regression and random forest regression to bridge the period of GPS outages. Neurocomputing166: 185–192.
2.
AsifUJavedMFAbuhussainM, et al. (2024) Predicting the mechanical properties of plastic concrete: An optimization method by using genetic programming and ensemble learners. Case Studies in Construction Materials20: e03135.
3.
AzevedoBFRochaAMACPereiraAI (2024) Hybrid approaches to optimization and machine learning methods: A systematic literature review. Machine Learning113(7): 4055–4097.
4.
BabuCNReddyBE (2014) A moving-average filter based hybrid ARIMA–ANN model for forecasting time series data. Applied Soft Computing23: 27–38.
5.
BayırlıABKesginBUytunM, et al. (2026) Segmentation-Based multi-class detection and radiographic charting of periodontal and restorative conditions on bitewing radiographs using deep learning. Diagnostics16(2): 22.
6.
BiauGScornetE (2016) A random forest guided tour. TEST25(2): 197–227.
7.
BreimanL (2001) Random forests. Machine Learning45(1): 5–32.
8.
CaiY-DLiuX-JXuX-b, et al. (2001) Support vector machines for predicting protein structural class. BMC Bioinformatics2(1): 3.
9.
CaiY-DLiuX-JXuX-b, et al. (2002) Prediction of protein structural classes by support vector machines. Computers & Chemistry26(3): 293–296.
10.
CaoEBaoTLiH, et al. (2022) A hybrid feature selection-multidimensional LSTM framework for deformation prediction of super high arch dams. KSCE Journal of Civil Engineering26(11): 4603–4616.
11.
CaoWWenZFengY, et al. (2024) A multi-point joint prediction model for high-arch dam deformation considering spatial and temporal correlation. Water16(10): 1388.
12.
DaiBGuCZhaoE, et al. (2018) Statistical model optimized random forest regression model for concrete dam deformation monitoring. Structural Control and Health Monitoring25(6): e2170.
13.
FanCXiaoFYanC, et al. (2019) A novel methodology to explain and evaluate data-driven building energy performance models based on interpretable machine learning. Applied Energy235: 1551–1560.
14.
FangCJYWangXLuT, et al. (2024) A dam displacement prediction method based on a model combining random forest, a convolutional neural network, and a residual attention informer. Water3687(16(24)): 6 .
15.
Fernández-CentenoMÁAlocénPToledoMÁ (2024) Identification of trends in dam monitoring data series based on machine learning and individual conditional expectation curves. Water16(9): 1239.
16.
GamseSZhouW-HTanF, et al. (2017) Hydrostatic-season-time model updating using Bayesian model class selection. Reliability Engineering & System Safety169: 40–50.
GravesAFernándezSSchmidhuberJ (2005) Bidirectional LSTM networks for improved phoneme classification and recognition. In: Artificial neural networks: formal models and their applications – ICANN 2005. Berlin, Heidelberg: Springer Berlin Heidelberg.
19.
GuCCuiXGuH, et al. (2023) A new hybrid monitoring model for displacement of the concrete dam. Sustainability15(12): 9609.
20.
GuoHZhangJDongY, et al. (2026) LSTM-augmented probability-informed neural network-driven evolution estimation for time-dependent reliability analysis. Structural Safety118: 102652.
21.
HanTJiangDZhaoQ, et al. (2018) Comparison of random forest, artificial neural networks and support vector machine for intelligent diagnosis of rotating machinery. Transactions of the Institute of Measurement and Control40(8): 2681–2693.
22.
HastieTTibshiraniRFriedmanJH (2009) The Elements of Statistical Learning: Data Mining, Inference, and Prediction. New York, NY, USA: Springer.
23.
HeDZhaoJJinZ, et al. (2025) DCAGGCN: A novel method for remaining useful life prediction of bearings. Reliability Engineering & System Safety260: 110978.
24.
HePLiY (2022) A data-driven dam deformation forecasting and interpretation method using the measured prototypical temperature data. Water14(16): 2538.
25.
HouMWeiJShiY, et al. (2024) Hydroformer: Frequency domain enhanced multi-attention transformer for monthly lake level reconstruction with low data input requirements. Water Resources Research60(10): e2024WR037166.
26.
HuangTZhangQTangX, et al. (2022) A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems. Artificial Intelligence Review55(2): 1289–1315.
27.
KalusivalingamAKSharmaAPatelN, et al. (2021) Leveraging deep learning and random forest algorithms for ai-driven genomics in personalized medicine. International Journal of AI and ML2(3).
28.
KarbhariVMAnsariF (2009) Structural Health Monitoring of Civil Infrastructure Systems. Elsevier.
29.
KhanAAChaudhariOChandraR (2024) A review of ensemble learning and data augmentation models for class imbalanced problems: Combination, implementation and evaluation. Expert Systems with Applications244: 122778.
30.
KhattiJGroverKSKimH-J, et al. (2024) Prediction of ultimate bearing capacity of shallow foundations on cohesionless soil using hybrid LSTM and RVM approaches: An extended investigation of multicollinearity. Computers and Geotechnics165: 105912.
31.
KumarSPalSKSinghRP (2018) A novel method based on extreme learning machine to predict heating and cooling load through design and structural attributes. Energy and Buildings176: 275–286.
32.
KuraniADoshiPVakhariaA, et al. (2023) A comprehensive comparative study of artificial neural network (ANN) and support vector machines (SVM) on stock forecasting. Annals of Data Science10(1): 183–208.
33.
LiBYangJHuD (2020) Dam monitoring data analysis methods: A literature review. Structural Control and Health Monitoring27(3): e2501.
34.
LiMLiMRenQ, et al. (2022a) DRLSTM: A dual-stage deep learning approach driven by raw monitoring data for dam displacement prediction. Advanced Engineering Informatics51: 101510.
35.
LiMPanJLiuY, et al. (2022b) Dam deformation forecasting using SVM-DEGWO algorithm based on phase space reconstruction. PLOS ONE17: e0267434.
36.
LiZ (2020) Global sensitivity analysis of the static performance of concrete gravity dam from the viewpoint of structural health monitoring. Archives of Computational Methods in Engineering28: 1611–1646.
37.
LiZKhailahEYLiuX, et al. (2025) Exploring purpose-driven methods and a multifaceted approach in dam health monitoring data utilization. Buildings15(15): 2803.
38.
LiZMaELaiJ, et al. (2024) Tunnel deformation prediction during construction: An explainable hybrid model considering temporal and static factors. Computers & Structures294: 107276.
39.
LinCLiTChenS, et al. (2019) Gaussian Process regression-based forecasting model of dam deformation. Neural Computing and Applications31(12): 8503–8518.
40.
LiuJMaCGuiH, et al. (2022) Transfer learning-based thermal error prediction and control with deep residual LSTM network. Knowledge-Based Systems237: 107704.
41.
LiuJWeiXHuangH (2021) An improved grey wolf optimization algorithm and its application in path planning. IEEE Access9: 121944–121956.
42.
LiuLLiLNianH, et al. (2023a) Enhanced grey wolf optimization algorithm for mobile robot path planning. Electronics12(19): 4026.
43.
LiuMWenZSuH (2024) A multipoint prediction model for the deformation of concrete dams considering climatic features of high-altitude regions. Engineering Structures319: 118845.
44.
LiuXLiZSunL, et al. (2023b) A critical review of statistical model of dam monitoring data. Journal of Building Engineering80: 108106.
45.
LuoSWeiBChenL (2024) Multi-point deformation monitoring model of concrete arch dam based on MVMD and 3D-CNN. Applied Mathematical Modelling125: 812–826.
46.
MengSShiZLiG, et al. (2024) A novel deep learning framework for landslide susceptibility assessment using improved deep belief networks with the intelligent optimization algorithm. Computers and Geotechnics167: 106106.
47.
MirjaliliSMirjaliliSMLewisA (2014) Grey wolf optimizer. Advances in Engineering Software69: 46–61.
48.
OuBCaiyiZBoX, et al. (2024) Innovative approach to dam deformation analysis: Integration of VMD, fractal theory, and WOA-DELM. Structural Control and Health Monitoring2024(1): 1710019.
49.
PascanuRMikolovTBengioY (2012) On the difficulty of training Recurrent Neural Networks. In: 30th International Conference on Machine Learning, ICML 2013.
50.
PhamA-DNgoN-TNguyenQ-T, et al. (2020) Hybrid machine learning for predicting strength of sustainable concrete. Soft Computing24(19): 14965–14980.
51.
QiuYYangXChenS (2024) An improved gray wolf optimization algorithm solving to functional optimization and engineering design problems. Scientific Reports14(1): 14190.
52.
Román-PortabalesALópez-NoresMPazos-AriasJJ (2021) Systematic review of electricity demand forecast using ANN-based machine learning algorithms. Sensors21(13): 4544.
53.
Safar YousefifardHGhodrati AmiriGDarvishanE, et al. (2025) Intelligent hybrid approaches utilizing time series forecasting error for enhanced structural health monitoring. Mechanical Systems and Signal Processing224: 112177.
54.
SalazarFMoránRToledoMÁ, et al. (2017) Data-Based models for the prediction of dam behaviour: A review and some methodological considerations. Archives of Computational Methods in Engineering24(1): 1–21.
55.
SchoppaLDisseMBachmairS (2020) Evaluating the performance of random forest for large-scale flood discharge simulation. Journal of Hydrology590: 125531.
56.
SevieriGDe FalcoA (2020) Dynamic structural health monitoring for concrete gravity dams based on the Bayesian inference. Journal of Civil Structural Health Monitoring10: 235–250.
57.
ShaheenMAMHasanienHMAlkuhayliA (2021) A novel hybrid GWO-PSO optimization technique for optimal reactive power dispatch problem solution. Ain Shams Engineering Journal12(1): 621–630.
58.
ShaoCGuCYangM, et al. (2018) A novel model of dam displacement based on panel data. Structural Control and Health Monitoring25(1): e2037.
59.
SunZSandovalLCrystal-OrnelasR, et al. (2022) A review of Earth Artificial Intelligence. Computers & Geosciences159: 105034.
60.
ThaiH-T (2022) Machine learning for structural engineering: A state-of-the-art review. Structures38: 448–491.
61.
TuBWangFHuoY, et al. (2023) A hybrid algorithm of grey wolf optimizer and harris hawks optimization for solving global optimization problems with improved convergence performance. Scientific Reports13(1): 22909.
62.
WangBGuoYWangD, et al. (2022) Prediction model of natural gas pipeline crack evolution based on optimized DCNN-LSTM. Mechanical Systems and Signal Processing181: 109557.
63.
WangHLiBGongJ, et al. (2023) Machine learning-based fatigue life prediction of metal materials: Perspectives of physics-informed and data-driven hybrid methods. Engineering Fracture Mechanics284: 109242.
64.
WangSXuCGuC, et al. (2020) Displacement monitoring model of concrete dams using the shape feature clustering-based temperature principal component factor. Structural Control and Health Monitoring27. DOI: 10.1002/stc.2603.
65.
WuWSuHFengY, et al. (2024) A novel artificial intelligence prediction process of concrete dam deformation based on a stacking model fusion method. Water16(13): 1868.
66.
XieHZhangLLimCP (2020) Evolving CNN-LSTM models for time series prediction using enhanced grey wolf optimizer. IEEE Access8: 161519–161541.
67.
XuGLuYJingZ, et al. (2023) IEALL: Dam deformation prediction model based on combination model method. Applied Sciences13(8): 5160.
68.
XuRLiuXWeiJ, et al. (2024) Predicting the deformation of a concrete dam using an integration of long short-term memory (LSTM) networks and Kolmogorov–Arnold networks (KANs) with a dual-stage attention mechanism. Water16(21): 3043.
69.
YuXDuanYCaiZ (2023) Sub-population improved grey wolf optimizer with Gaussian mutation and Lévy flight for parameters identification of photovoltaic models. Expert Systems with Applications232: 120827.
70.
ZhangCZhouJLiC, et al. (2017) A compound structure of ELM based on feature selection and parameter optimization using hybrid backtracking search algorithm for wind speed forecasting. Energy Conversion and Management143: 360–376.
71.
ZhangRChenZChenS, et al. (2019) Deep long short-term memory networks for nonlinear structural seismic response prediction. Computers & Structures220: 55–68.
72.
ZhangW-SYuanYLongM, et al. (2024) Prediction of surface settlement around subway foundation pits based on spatiotemporal characteristics and deep learning models. Computers and Geotechnics168: 106149.
