Genetic Programming to Predict Bridge Pier Scour

Abstract

Bridge pier scouring is a significant problem for the safety of bridges. Extensive laboratory and field studies have been conducted examining the effect of relevant variables. This note presents an alternative to the conventional regression-based equations (HEC-18 and regression equation developed by authors), in the form of artificial neural networks (ANNs) and genetic programming (GP). 398 data sets of field measurements were collected from published literature and used to train the network or evolve the program. The developed network and evolved programs were validated by using the observations that were not involved in training. The performance of GP was found more effective when compared to regression equations and ANNs in predicting the scour depth of bridge piers.

Modelling of upheaval buckling of offshore pipeline buried in clay soil using genetic programming

“…For scoring complexity, the total number of complex functions including arctan, sin, cos, exp, ln were considered. Maximum number of these complex functions (7) is available in equation of GEP2 model while their minimum number (1) has been appeared in GEP9 model. Therefore, scores of 7, 6, 5, 4, 3, 2 and 1 are considered for number of complex functions of 1, 2, 3, 4, 5, 6 and 7 respectively.…”

“…Azamathulla et al [1] have employed GEP for predicting bridge pier scour. Guven and Gunal [10] have used it for Prediction of Local Scour Downstream of Hydraulic Structures.…”

Can we obtain viable alternatives to Manning’s equation using genetic programming?

“…Also, it is worth mentioning the effort of Giustolisi and Savic, [34] where they build upon Giustolisi, [29] and applied symbolic regression to find an explicit polynomial function for the Colebrook-White friction factor (Colebrook & White, 1937); their findings showed that an eleven-term polynomial plus the bias presented the best results. More recently, Jagupilla et al Jagupilla [35] used river flow information and symbolic regression to obtain concentrations of E. Coli in water systems; Fallah-Mehdipour [36] used GP for groundwater modeling, and Azamathulla [37] used it to predict scour under bridge piers; while Ines et al [38] used Genetic Algorithms to estimate parameters of soil hydraulic functions. In general, the application of evolutionary techniques in hydraulics has been centered on parameter optimization and/or parameter estimation (e.g., obtaining the Chezy resistance coefficient [29] -comparable to the GMS's n coefficient), with a fraction of those studies focusing on symbolic regression (e.g.…”

Gene Expression Programing for Estimating Suspended Sediment Yield in Middle Euphrates Basin, Turkey

“…The first is to choose the fitness function [32,33]. For this problem, the fitness f i of an individual program i is defined by the following expression:…”

“…In this problem, two terminal sets are used based on two input sets, namely T 1 consists of single independent variable, i.e., T ¼ {Q}, and T 2 consists of three independent variables, i.e., T ¼ {Q, Q tÀ1 , SSY tÀ1 }. The choice of the appropriate function set is not so obvious, but a good guess can always be done in order to include all the necessary functions [32,33]. In this case, four basic arithmetic operators (þ, À, Ã , /), and some basic functions (H, ln(x), log(x), e x , 10 x , power) are used.…”