1. A hybrid GMDH neural network and logistic regression framework for state parameter-based liquefaction evaluation
- Author
-
Duan, Wei, Congress, Surya Sarat Chandra, Cai, Guojun, Liu, Songyu, Dong, Xiaoqiang, Chen, Ruifeng, and Liu, Xuening
- Subjects
Soil liquefaction -- Models ,Logistic regression -- Usage ,Neural networks -- Usage ,Neural network ,Earth sciences - Abstract
The cyclic stress or liquefaction behavior of granular materials is strongly affected by the relative density and confining pressure of the soil. In this study, the state parameter accounting for both relative density and effective stress was used to evaluate soil liquefaction potential. Based on case histories along with the cone penetration test (CPT) database, models for calculating the state parameter using a group method of data handling (GMDH) neural network were developed and recommended according to their performance. The state parameter was then used to develop a state parameter- based probabilistic liquefaction evaluation method using a logistic regression model. From a conservative point of view, the boundary curve of 20% probability of liquefaction was suggested as a deterministic criterion for state parameter- based liquefaction evaluation. Subsequently, a mapping function relating the calculated factor of safety ([F.sub.S]) to the probability of liquefaction ([P.sub.L]) was proposed based on the compiled CPT database. Based on the developed [P.sub.L]- [F.sub.S] function, a new risk criterion associated with the state parameter-based design chart was proposed. Finally, a flowchart of state-based probabilistic liquefaction evaluation and quality control for ground-improvement projects was presented for the benefit of practitioners. Key words: group method of data handling (GMDH), logistic regression, state parameter, liquefaction, cone penetration test. La contrainte cyclique ou le comportement de liquefaction des materiaux granulaires est fortement affecte par la densite relative et la pression de confinement du sol. Au cours de cette etude, le parametre d'etat tenant compte a la fois de la densite relative et de la contrainte effective a ete utilise pour evaluer le potentiel de liquefaction du sol. Sur la base d'etudes de cas et de la base de donnees de l'essai de penetration au cone (<< CPT >>), des modeles de calcul du parametre d'etat utilisant un reseau neuronal de la methode de groupe de traitement des donnees (<< GMDH >>) ont ete developpes et recommandes en fonction de leurs performances. Le parametre d'etat a ensuite ete utilise pour developper une methode d'evaluation probabiliste de la liquefaction basee sur le parametre d'etat en utilisant un modele de regression logistique. D'un point de vue conservateur, la courbe limite de 20 % de probabilite de liquefaction a ete proposee comme critere deterministe pour l'evaluation de la liquefaction basee sur les parametres d'etat. Par la suite, une fonction de mise en correspondance entre le facteur de securite ([F.sub.S]) calcule et la probabilite de liquefaction ([P.sub.L]) a ete proposee sur la base de donnees CPT compilee. Sur la base de la fonction [P.sub.L]- [F.sub.S] developpee, un nouveau critere de risque associe a la carte de conception basee sur les parametres d'etat a ete propose. Finalement, un organigramme de l'evaluation probabiliste de la liquefaction basee sur l'etat et du controle de la qualite pour les projets d'amelioration du sol a ete presente a l'intention des praticiens. [Traduit par la Redaction] Mots-cles : methode de groupe de traitement des donnees (<< GMDH >>), regression logistique, parametre d'etat, liquefaction, essai de penetration au cone., 1. Introduction Due to the high cost and difficulty associated with collecting high-quality undisturbed soil samples and testing granular soils, various in situ methods are used by geotechnical engineers to [...]
- Published
- 2021
- Full Text
- View/download PDF