Using the Bayesian network to predict the remaining useful life of the reinforced concrete decks Under chloride corrosion

Document Type : Original Research

Authors
A. mehdizadeh lima 1
M. Mahmoudi 1
A. Zayeri baghlani nejad 2
1 Faculty of Civil Engineering, Shahid Rajaee Teacher Training University, Tehran, Iran.
2 Faculty of Civil Engineering, Jundi-Shapur University of Technology, Dezful, Iran.
10.22034/24.4.51
Abstract
Structures, including concrete bridges, may be exposed to gradual damage during operation due to environmental conditions such as corrosion, which will reduce their useful life. Knowing the amount of remaining useful life of the structures makes it possible to improve, strengthen or rebuild them at the right time. To determine the remaining useful life of a structure, there are three common methods under the titles of data-driven method, failure physics method and combined method. In this article, the combined method of determining the remaining useful life of structures has been studied. The purpose of this research is to propose a suitable method for predicting the remaining useful life of a bridge structure with a reinforced concrete deck under chloride ion corrosion using a Bayesian network. The remaining useful life of reinforced concrete parts under chloride attack includes two parts of the time related to the initial stage of corrosion and the time related to the release of chlorine ions. To determine the remaining useful life part related to the initial stage, various researches have been done and the American ACI365 committee has proposed a software called Life-365 for this purpose. There is no comprehensive research to determine the second part of the remaining life, which is related to the release stage. Based on the prepared Bayesian network and the formula obtained in this research, the remaining life of the chloride diffusion stage in concrete was estimated to be 9.116 years in the best conditions and 2.73 years in the worst conditions. Meanwhile, the number suggested by the ACI365 committee, in practical work, is usually equal to 6 years for the release stage. This issue clarifies the need for more research in this regard. In this article, using the data available in past researches and reproducing the data and using the Bayesian network, relationships are presented to determine the useful life of the bridge structure in both the initial and release stages.Based on the proposed method, using the Bayesian network, relationships can be obtained for each of the two parts of the remaining useful life of the structure under chloride corrosion, i.e., the corrosion initiation stage and the chloride release stage, in terms of factors affecting the remaining useful life in a specific project. . In these networks, the effect of various factors can be considered, which is one of the advantages of the proposed method.The remaining useful life has an inverse relationship with temperature. When the average temperature increases by 20 degrees, the remaining useful life decreases by an average of 30%.With the help of the proposed relationships, a parametric study was conducted to investigate the effect of different conditions of using pozzolanic compounds on the remaining life of the structure. In this regard, 17 states of different pozzolanic compounds with different concentrations were considered and the average remaining useful life due to different states was calculated. The average life obtained compared to the case where no pozzolan is used in concrete showed a 38% increase in life. In order to evaluate the results of the proposed relationships, the problem of determining the remaining useful life for a numerical model of a concrete bridge and several marine structures located in the Persian Gulf was investigated. The results of this research show that by using the proposed relationships, it is possible to improve the accuracy of estimating the remaining useful life of bridges with concrete decks exposed to chloride ion penetration, relying on the data obtained from the field inspections of the structure.

Keywords

Subjects

[1] Seyed Hosseini S.M., Baharshahi M., Shahanqi K. presenting a data-driven model for estimating the remaining useful life using the combination of turbofan sensor data, Journal of Industrial Engineering Research in Production Systems, 2018, serial 15. (In Persian)
[2] Ali JB, Chebel-Morello B, Saidi L, Malinowski S, Fnaiech F. Accurate bearing remaining useful life prediction based on Weibull distribution and artificial neural network. Mechanical Systems and Signal Processing. 2015 May 1; 56:150-72.
[3] Kunche S, Chen C, Pecht M. A review of PHM system’s architectural frameworks. InThe 54th meeting of the society for machinery failure prevention technology, dayton, oh 2012.
[4] Lei Y, Li N, Guo L, Li N, Yan T, Lin J. Machinery health prognostics: A systematic review from data acquisition to RUL prediction. Mechanical systems and signal processing. 2018 May 1; 104:799-834.
[5] Lei Y, Li N, Guo L, Li N, Yan T, Lin J. Machinery health prognostics: A systematic review from data acquisition to RUL prediction. Mechanical systems and signal processing. 2018 May 1; 104:799-834.
[6] Abichou B, Flórez D, Sayed-Mouchaweh M, Toubakh H, François B, Girard N. Fault diagnosis methods for wind turbines health monitoring: a review. InPHM Society European Conference 2014 (Vol. 2, No. 1).
[7] Si XS, Wang W, Hu CH, Zhou DH. Remaining useful life estimation–a review on the statistical data driven approaches. European journal of operational research. 2011 Aug 16;213(1):1-4.
[8] Baur M, Albertelli P, Monno M. A review of prognostics and health management of machine tools. The International Journal of Advanced Manufacturing Technology. 2020 Mar; 107:2843-63.
[9] Xia T, Dong Y, Xiao L, Du S, Pan E, Xi L. Recent advances in prognostics and health management for advanced manufacturing paradigms. Reliability Engineering & System Safety. 2018 Oct 1; 178:255-68.
[10] Cai B, Shao X, Liu Y, Kong X, Wang H, Xu H, Ge W. Remaining useful life estimation of structure systems under the influence of multiple causes: Subsea pipelines as a case study. IEEE Transactions on Industrial Electronics. 2019 Aug 5;67(7):5737-47.
[11] Chun P.J., Inoue T., Seto D., Ohga M. Prediction of Bridge Deterioration Using GIS-Based Markov Transition Matrix, Internet Journal of Society for Social Management Systems. 2012; 8: 1–9.
[12] Morcous G, Lounis Z, Cho Y. An integrated system for bridge management using probabilistic and mechanistic deterioration models: Application to bridge decks. KSCE Journal of Civil Engineering. 2010 Jul; 14:527-37.
[13] Ranjith S, Setunge S, Gravina R, Venkatesan S. Deterioration prediction of timber bridge elements using the Markov chain. Journal of Performance of Constructed Facilities. 2013 Jun 1;27(3):319-25.
[14] Kosgodagan A. High-dimensional dependence modelling using Bayesian networks for the degradation of civil infrastructures and other applications (Doctoral dissertation, University of Bretagne Loire, Bretagne). 2017.
[15] Kosgodagan‐Dalla Torre A, Yeung TG, Morales‐Nápoles O, Castanier B, Maljaars J, Courage W. A two‐dimension dynamic Bayesian network for large‐scale degradation modeling with an application to a bridges network. Computer‐Aided Civil and Infrastructure Engineering. 2017 Aug;32(8):641-56
[16] Srikanth I, Arockiasamy M. Deterioration models for prediction of remaining useful life of timber and concrete bridges: A review. Journal of traffic and transportation engineering (English edition). 2020 Apr 1;7(2):152-73.
[17] Abbasnezhad,K. and Ansari,R. Bayesian networks, a new approach for modeling uncertainties in construction project management. Journal of Engineering and Construction Management, Volume 5, Number 1, Spring 2019.
(In Persian)
[18] Eliasi,M. Seifi,H. and Haghifam,M.,Modeling and evaluation of hybrid power system reliability and reliability-based analysis using Bayesian networks,Iranian Journal of Electrical Engineering and Computer Engineering (series 35), fall 2013, number 2. (In Persian)
[19] Ramezanianpour A.A., Jahangiri E., Moodi F., Ahmadi B. Assessment of the Service Life Design Model Proposed by fib for the Persian Gulf Region. Journal of Oceanography. 2014 Apr 15;5(17):101-112.
[20] Ramzanianpour Ali Akbar, Jahangiri Ehsan, Modi Faramarz, Ahmadi Babak. Evaluation of service life estimation models for reinforced concrete structures in the marine environment of the Persian Gulf. 1389. The second national concrete conference of Iran. (In Persian)
[21] Neapolitan, R. E. (2004). Learning Bayesian networks (Vol. 38). Upper Saddle River: Pearson Prentice Hall.
[22] Lun PY, Zhang XG, Jiang C, Ma YF, Fu L. Modelling of corrosion-induced concrete cover cracking due to chloride attacking. Materials. 2021 Mar 16;14(6):1440.
Modares Civil Engineering journal
Volume 24, Issue 4
September and October 2024
Pages 51-67