The load-settlement behaviour of a model footing subjected to vertical eccentric load is simu-lated using the architecture and learning procedure named ANFIS (Adaptive-Network-based Fuzzy Inference System) trained by the experimental results of a series of load tests both in normal and incremented gravity conditions. Issues related to fuzzy systems, membership functions and over-training error are briefly dis-cussed. The numerical simulations compare favourably with the correspondent laboratory tests. New load-settlement curves could be simulated to obtain the collapse interaction diagrams which are currently intro-duced in mathematical models to predict the displacement evolution of a shallow footing. The efforts encour-age further investigations on the subject to describe adequately the overall behaviour of footings under the combined action of inclined and/or eccentric loads.
Modelling of a footing behaviour using an Adaptive Neuro-Fuzzy Inference System.
FERLISI, Settimio
2002
Abstract
The load-settlement behaviour of a model footing subjected to vertical eccentric load is simu-lated using the architecture and learning procedure named ANFIS (Adaptive-Network-based Fuzzy Inference System) trained by the experimental results of a series of load tests both in normal and incremented gravity conditions. Issues related to fuzzy systems, membership functions and over-training error are briefly dis-cussed. The numerical simulations compare favourably with the correspondent laboratory tests. New load-settlement curves could be simulated to obtain the collapse interaction diagrams which are currently intro-duced in mathematical models to predict the displacement evolution of a shallow footing. The efforts encour-age further investigations on the subject to describe adequately the overall behaviour of footings under the combined action of inclined and/or eccentric loads.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
