A novel hybrid model for bridge dynamic early warning using LSTM-EM-GMM

Abstract

Early warning of existing bridges is now predominated by deterministic methods. However, these methods face challenges in expressing uncertain factors (such as wind load, temperature load, and other variables, etc.). These problems directly impact the timeliness and accuracy of bridge early warning. This study develops an innovative method for bridge dynamic early warning with high versatility and accuracy. Long short-term memory network model (LSTM), expectation maximization (EM) and Gaussian mixture model (GMM) were employed in the proposed method. Firstly, the LSTM model is used to predict the measured monitoring data (such as deflection, strain, cable force, etc.) in real time to obtain the predicted results. Next, the number of clusters for the EM-GMM model is determined using the Calinski-Harabasz (CH) index. The method aims to comprehensively consider the internal cohesion of the clustering, ensuring accurate and reliable clustering results. Then, the EM-GMM model is used to cluster the random influence error and the predicted value, which can get the probabilistic prediction result of each corresponding random influence error. On this basis, the dynamic early warning interval under 95% confidence level is constructed. This facilitates early warning and decision-making for potential structural abnormalities. Finally, the accuracy and practicability of the method are verified by the comparison of engineering applications and existing specifications. The results demonstrate that the probabilistic early warning method considering the uncertain factors in the complex service environment can accurately achieve the dynamic early warning of bridges.

Keywords

Bridge / Structural health monitoring / Probabilistic method / Dynamic early warning / Uncertain factors

References

[] Asad AT, Kim B, Cho S, Sim SH. Prediction model for long-term bridge bearing displacement using artificial neural network and Bayesian optimization. Struct Control Hlth, 2023, 2023: 6664981,

[] Buckley T, Pakrashi V, Ghosh B. A dynamic harmonic regression approach for bridge structural health monitoring. Struct Health Monit, 2021, 20(6): 3150-3181,

[] Cao SP, Hu ZJ, Luo XH, Wang HW. Research on fault diagnosis technology of centrifugal pump blade crack based on PCA and GMM. Measurement, 2021, 173: 108558 ,

[] Daneshvar MH, Sarmadi H, Yuen KV (2023) A locally unsupervised hybrid learning method for removing environmental effects under different measurement periods. Measurement 208:112465. https://doi.org/10.1016/j.measurement.2023.112465

[] Diez A, Khoa NLD, Alamdari MM, Wang Y, Chen F, Runcie P. A clustering approach for structural health monitoring on bridges. J Civ Struct Health, 2016, 6: 429-445,

[] Entezami A, Sarmadi H, Behkamal B. A novel double-hybrid learning method for modal frequency-based damage assessment of bridge structures under different environmental variation patterns. Mech Syst Signal Pr, 2023, 201: 110676,

[] Entezami A, Sarmadi H, Behkamal B, Michele CD (2023) On continuous health monitoring of bridges under serious environmental variability by an innovative multi-task unsupervised learning method. Struct Infrastruct E 1–19. https://doi.org/10.1080/15732479.2023.2166538

[] Fan ZY, Huang Q, Ren Y, Xu X, Zhu ZY. Real-time dynamic warning on deflection abnormity of cable-stayed bridges considering operational environment variations. J Perform Constr Fac, 2021, 35(1): 04020123,

[] García S, Luengo J, Herrera F. . Data preprocessing in data mining, 2015,

[] Gu B, Hu H, Zhao J, Zhang HT, Liu XY. Short-term wind power forecasting and uncertainty analysis based on FCM–WOA–ELM–GMM. Energy Rep, 2023, 9: 807-819,

[] Hu HZ, Liu JX, Zhang XP, Fang MG. An effective and adaptable K-means algorithm for cluster analysis. Pattern Recogn, 2023, 139: 109404,

[] Khosravi A, Nahavandi S, Creighton D. Prediction intervals for short-term wind farm power generation forecasts. Ieee T Sustain Energ, 2013, 4(3): 602-610,

[] Lei W, Wang J, Ji TY, Li PF. Dam deformation early warning model based on cluster analysis and spatiotemporal data fusion. Measurement, 2022, 204: 112109,

[] Li SW, Laima SJ, Li H. Cluster analysis of winds and wind-induced vibrations on a long-span bridge based on long-term field monitoring data. Eng Struct, 2017, 138: 245-259,

[] Li G, Ran RJ, Fang J, Peng H, Wang SM. Early warning for the construction safety risk of bridge projects using a RS-SSA-LSSVM model. Adv Civ Eng, 2021, 2021: 1-14,

[] Li YF, Wu HP, Liu H. Multi-step wind speed forecasting using EWT decomposition, LSTM principal computing, RELM subordinate computing and IEWT reconstruction. Energ Convers Manage, 2018, 167: 203-219,

[] Li SJ, Xin JZ, Jiang Y, Wang CW, Zhou JT, Yang XY. Temperature-induced deflection separation based on bridge deflection data using the TVFEMD-PE-KLD method. J Civ Struct Health, 2023, 13: 781-797,

[] Li DL, Yang DH, Yi TH, Zhang SH, Li WJ, Li C. Anomaly diagnosis of stay cables based on vehicle-induced cable force sums. Eng Struct, 2023, 289,

[] Liu B, Liu CF, Zhou Y, Wang DH, Dun YC. An unsupervised chatter detection method based on AE and merging GMM and K-means. Mech Syst Signal Pr, 2023, 186: 109861,

[] Lu YK, Tian Z, Peng P, Niu JD, Li WC, Zhang HJ. GMM clustering for heating load patterns in-depth identification and prediction model accuracy improvement of district heating system. Energ Building, 2019, 190: 49-60,

[] Morgese M, Wang C, Taylor T, Etemadi M, Ansari F. Distributed detection and quantification of cracks in operating large bridges. J Bridge Eng, 2024, 29(1): 04023101,

[] Morgese M, Wang C, Ying Y, Talor T, Ansari F. Stress–strain response of optical fibers in direct tension. J Eng Mech, 2023, 149(7): 04023037,

[] Nassif AB, Shahin I, Hamsa S, Nemmour N, Hirose K. CASA-based speaker identification using cascaded GMM-CNN classifier in noisy and emotional talking conditions. Appl Soft Comput, 2021, 103,

[] Ni YQ, Wang YW, Zhang C. A Bayesian approach for condition assessment and damage alarm of bridge expansion joints using long-term structural health monitoring data. Eng Struct, 2020, 212,

[] Panda SK, Jana PK. Efficient task scheduling algorithms for heterogeneous multi-cloud environment. J Supercomput, 2015, 71: 1505-1533,

[] Sarmadi H, Entezami A. Application of supervised learning to validation of damage detection. Arch Appl Mech, 2021, 91(1): 393-410,

[] Sarmadi H, Entezami A, Salar M, Michele CD. Bridge health monitoring in environmental variability by new clustering and threshold estimation methods. J Civ Struct Health, 2021, 11: 629-644,

[] Silva DS, Júnior MD, Junior VL, Brennan M. Structural damage detection by fuzzy clustering. Mech Syst Signal Pr, 2008, 22(7): 1636-1649,

[] Tang QZ, Xin JZ, Jiang Y, Zhou JT, Li SP, Fu L. Fast identification of random loads using the transmissibility of power spectral density and improved adaptive multiplicative regularization. J Sound Vib, 2022, 534: 117033,

[] Tong K, Zhang H, Zhao RQ, Zhou JT, Ying HJ. Investigation of SMFL monitoring technique for evaluating the load-bearing capacity of RC bridges. Eng Struct, 2023, 293: 116667,

[] Wang CW, Ansari F, Wu B, Li SJ, Morgese M, Zhou JT. LSTM approach for condition assessment of suspension bridges based on time-series deflection and temperature data. Adv Struct Eng, 2022, 25(16): 3450-3463,

[] Wang Y, Hu QH, Meng DY, Zhu PF. Deterministic and probabilistic wind power forecasting using a variational Bayesian-based adaptive robust multi-kernel regression model. Appl Energ, 2017, 208: 1097-1112,

[] Wu HJ, Yuan GJ, Wang T, Qu HR, Li MS. Threshold setting and application based on structural response characteristics of small and medium span beam bridge. J Chongqing Jiaotong Univ (Nat Sci), 2020, 39(09): 54-58, in Chinese)

[] Xin JZ, Jiang Y, Zhou JT, Peng LL, Liu SY, Tang QZ. Bridge deformation prediction based on SHM data using improved VMD and conditional KDE. Eng Struct, 2022, 261: 114285,

[] Xin JZ, Zhou CY, Jiang Y, Tang QZ, Yang XY, Zhou JT. A signal recovery method for bridge monitoring system using TVFEMD and encoder-decoder aided LSTM. Measurement, 2023, 214(15): 112797,

[] Yang CX, Liu Y, Sun YQ. Damage detection of bridges considering environmental temperature effect by using cluster analysis. Procedia Eng, 2016, 161: 577-582,

[] Ye XW, Sun Z, Lu J. Prediction and early warning of wind-induced girder and tower vibration in cable-stayed bridges with machine learning-based approach. Eng Struct, 2023, 275,

[] Zhao HW, Ding YL, Nagarajaiah S, Li AQ. Behavior analysis and early warning of girder deflections of a steel-truss arch railway bridge under the effects of temperature and trains: case study. J Bridge Eng, 2019, 24(1): 05018013,

[] Zou Q, Deng L, Jiang C. Predicting the bounds of vehicle-induced bridge responses using the interval analysis method. J Bridge Eng, 2016, 21(9): 04016046,

Funding

National Natural Science Foundation of China(Grant Nos. 52278292, 52108475, 52108435); Chongqing Outstanding Youth Science Foundation(Grant No. CSTB2023NSCQ-JQX0029); Chongqing Transportation Science and Technology Project(Grant No. 2022-01); China Postdoctoral Science Foundation(Grant No. 2023M730431); Special Funding of Chongqing Postdoctoral Research Project(Grant No. 2022CQBSHTB2053); Science and Technology Project of Guizhou Department of Transportation(Grant No. 2023-122-001); Chongqing Jiaotong University Postgraduate Research and Innovation Project(CYB23246)