Evaluating Slope Deformation of Earth Dams Due to Earthquake Shaking Using MARS and GMDH Techniques

Abstract

Assessing the behavior of earth dams under dynamic loads is one of the most significant problems with the design of such large structures. The purpose of this study is to provide new models for predicting dam dispersion in real earthquake conditions. In the first phase, 103 real cases of deformation in earth dams were collected and analyzed due to earthquakes that occurred over recent years. Using nonlinear and machine learning techniques, i.e., group method of data handling (GMDH) and multivariate adaptive regression splines (MARS), two models for prediction of the slope deformation in earth dams under the various types of earthquakes were applied and developed. The main parameters used in these simulation techniques were earthquake magnitude (Mw), fundamental period ratio (Td/Tp), yield acceleration ratio (ay/amax) as inputs and value of slope deformation (Dave) as output. Finally, in order to check the accuracy of the results of the new models, a comparison was made with the previous relations and models in seismic conditions for the slope deformation in earth dams. The results showed that the MARS model, which is able to provide a mathematical equation, has a better result than the GMDH model. These new models are recommended to be used for future analyses based on their flexible capabilities.

Reliability Analysis of Pile Foundation Using Soft Computing Techniques: A Comparative Study

“…Through quadratic polynomials, each neuron layer has many data points connected to each other, and new neurons are created in the process [39,40]. The general structure of the model is depicted in Figures 4 and 5 by Cai et al [28]. Input parameters are entered in the first layer and some functions are defined.…”

“…The use of emotions in machine learning is being extensively researched in recent past [24][25][26][27]. GMDH is self-organizing type neural networks (NN) which has been found to be reliable computational method to replace classical methods [28][29][30]. ANFIS has been applied in various fields in literature [31][32][33], and in the current study its developed for pile bearing capacity detection.…”

Review on Dynamic Behaviour of Earth Dam and Embankment During an Earthquake

“…In terms of reinforcement, some techniques have been used to improve the structural behaviour during an earthquake. It is also important to mention that among all available techniques for solving slope and dam problems, artificial intelligence has been successfully used and suggested in this area and geotechnical engineering (Singh et al 2004(Singh et al , 2005Verma and Singh 2013;Gordan et al 2016b;Mahdiyar et al 2017;Koopialipoor et al 2019;Zhou et al 2015Zhou et al , 2016aZhou et al , b, 2019Cai et al 2020;Tang et al 2020).…”

A Novel Feature Selection Approach Based on Tree Models for Evaluating the Punching Shear Capacity of Steel Fiber-Reinforced Concrete Flat Slabs

“…In the light of the above discussion, over the last decades, the applications of artificial intelligent (AI) and machine learning (ML) techniques have rapidly grown, and new intelligent models have been developed to solve problems in science and engineering (especially civil engineering) [ 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 ]. Hoang [ 62 ] studied the sophisticated data analysis approach for estimating shear punching capacity.…”