Genetic programming approach for predicting service life of tunnel structures subject to chloride-induced corrosion

• Published in: Journal of advanced research Vol. 20; pp. 141 - 152
• Main Authors: Gao, Wei, Chen, Xin, Chen, Dongliang
• Format: Journal Article
• Language: English
• Published: Egypt Elsevier B.V 01.11.2019; Elsevier
• Subjects: Chloride-induced corrosion; Data-driven method; Genetic programming; Original; Prediction; Service life; Tunnel structure; Tunnel structure; Service life; Chloride-induced corrosion; Genetic programming; Prediction; Data-driven method
• ISSN: 2090-1232; 2090-1224
• DOI: 10.1016/j.jare.2019.07.001

[Display omitted] •GP is used to predict service life of tunnel structure subject to chloride-induced corrosion.•This new method can construct an explicit expression of the prediction model.•This new prediction model can take into account 17 main corrosion factors.•The performance of the new model is compared with that of artificial neural network model.•The effects of two main controlling parameters are analyzed detailed. A new method for predicting the service life of tunnel structures subject to chloride-induced corrosion using data from real engineering examples and genetic programming (GP) is proposed. As a data-driven method, the new approach can construct explicit expressions of the prediction model. The new method was verified by comparing it with the chloride-ion diffusion model considering eight corrosion influence factors. Moreover, 25 datasets collected from tunnel engineering examples were used to construct the new prediction model considering 17 corrosion influence factors belonged to just one classification of engineering corrosion factors. In addition, the performance of the new model was verified through a comparative study with an artificial neural network (ANN) model which is frequently used in chloride-induced corrosion prediction for reinforced concrete structures. The comparison revealed that both the computational result and efficiency of the GP method were significantly better than those of the ANN model. Finally, to comprehensively analyze the new prediction model, the effects of the two main controlling parameters (population size and sample size) were analyzed. The results indicated that as both the population size and the sample size increased, their effect on the computation error decreased, and their optimal values were suggested as 300 and 20, respectively.