A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Evaluation of Adaptive Neuro-Fuzzy Inference System for Forecasting Soil’s Unconfined Compressive Strength
Adaptive Neuro-Fuzzy Inference System (ANFIS) was used to model the unconfined compressive strength (UCS) of lime and cement-treated soil. It provides advantages of a combination between neural network principles and Takagi–Sugeno Fuzzy logic in one context. In geotechnical research, the benefits of deploying artificial intelligence techniques include the ability to deal with the difficult challenges related to the effective use of construction materials to attain an optimal valuation of geotechnical material’s behavior and long-term engineering project. An ANFIS model was developed using 237 datasets generated by experimental results by varying quantities of lime and cement treatment ranging from 0 to 12%. The constructed model’s input variables were the percentage of sand, silt, clay, gravel, cement, plasticity index, organic content, and lime, and the output variable was standard soaked UCS of 7 days. The model assessment results revealed a coefficient of determination (R2) for testing data of 0.9927, for training data 0.8238, and ANFIS final results 0.952. This demonstrates the functioning of ANFIS to achieve long-term geotechnical materials integration in the built environment.
Evaluation of Adaptive Neuro-Fuzzy Inference System for Forecasting Soil’s Unconfined Compressive Strength
Adaptive Neuro-Fuzzy Inference System (ANFIS) was used to model the unconfined compressive strength (UCS) of lime and cement-treated soil. It provides advantages of a combination between neural network principles and Takagi–Sugeno Fuzzy logic in one context. In geotechnical research, the benefits of deploying artificial intelligence techniques include the ability to deal with the difficult challenges related to the effective use of construction materials to attain an optimal valuation of geotechnical material’s behavior and long-term engineering project. An ANFIS model was developed using 237 datasets generated by experimental results by varying quantities of lime and cement treatment ranging from 0 to 12%. The constructed model’s input variables were the percentage of sand, silt, clay, gravel, cement, plasticity index, organic content, and lime, and the output variable was standard soaked UCS of 7 days. The model assessment results revealed a coefficient of determination (R2) for testing data of 0.9927, for training data 0.8238, and ANFIS final results 0.952. This demonstrates the functioning of ANFIS to achieve long-term geotechnical materials integration in the built environment.
Evaluation of Adaptive Neuro-Fuzzy Inference System for Forecasting Soil’s Unconfined Compressive Strength
Lect.Notes Mechanical Engineering
Venkata Rao, Ravipudi (editor) / Taler, Jan (editor) / Jangir, Himanshu (author) / Rupali, S. (author)
Advanced Engineering Optimization Through Intelligent Techniques ; Chapter: 35 ; 377-387
2023-04-08
11 pages
Article/Chapter (Book)
Electronic Resource
English
| British Library Online Contents | 2017