Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Prediction of Liquefaction of Soils Using Particle Swarm Optimization (PSO)
Prediction of liquefaction potential of soils is significant in order to mitigate risk and major damages to structures. Currently used deterministic methods have drawbacks like mismatch between the assumptions in modelling and the actual in-situ conditions, observational errors. Hence many predictive techniques are being used as an alternative solution to reach a better decision and the neural networking approaches are an ideal one. This paper presents the technique of neural network to develop an Artificial Neural Network (ANN) model optimized by Particle Swarm Optimization (PSO), based on CPT data to predict the liquefaction potential of soils. The database used in this study consists of 235 CPT-based field records from ten major earthquakes over a period of 35 years. Important parameters including normalized peak horizontal acceleration at ground surface, earthquake magnitude, total vertical stress, effective vertical stress, cone resistance and depth of penetration, are selected as the input parameters for the ANN-PSO model. PSO technique is hybridized along with Artificial Neural Network (ANN) to utilize the advantage of both the techniques.
Prediction of Liquefaction of Soils Using Particle Swarm Optimization (PSO)
Prediction of liquefaction potential of soils is significant in order to mitigate risk and major damages to structures. Currently used deterministic methods have drawbacks like mismatch between the assumptions in modelling and the actual in-situ conditions, observational errors. Hence many predictive techniques are being used as an alternative solution to reach a better decision and the neural networking approaches are an ideal one. This paper presents the technique of neural network to develop an Artificial Neural Network (ANN) model optimized by Particle Swarm Optimization (PSO), based on CPT data to predict the liquefaction potential of soils. The database used in this study consists of 235 CPT-based field records from ten major earthquakes over a period of 35 years. Important parameters including normalized peak horizontal acceleration at ground surface, earthquake magnitude, total vertical stress, effective vertical stress, cone resistance and depth of penetration, are selected as the input parameters for the ANN-PSO model. PSO technique is hybridized along with Artificial Neural Network (ANN) to utilize the advantage of both the techniques.
Prediction of Liquefaction of Soils Using Particle Swarm Optimization (PSO)
Lecture Notes in Civil Engineering
Marano, Giuseppe Carlo (Herausgeber:in) / Ray Chaudhuri, Samit (Herausgeber:in) / Unni Kartha, G. (Herausgeber:in) / Kavitha, P. E. (Herausgeber:in) / Prasad, Reshma (Herausgeber:in) / Achison, Rinu J. (Herausgeber:in) / Anitta Justin, C. (Autor:in) / Sankar, N. (Autor:in)
International Conference on Structural Engineering and Construction Management ; 2021
04.09.2021
8 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Assessment of Seismic Liquefaction of Soils Using Swarm-Assisted Optimization Algorithm
| Springer Verlag | 2021
Identification of hydraulic parameters for unsaturated soils using particle swarm optimization
| British Library Conference Proceedings | 2008
Model prediction of cyclic liquefaction resistance of gassy soils
| Springer Verlag | 2024