A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Prediction of Structural Response Based on Ground Acceleration using Artificial Neural Networks
This study utilizes Artificial Neural Networks to predict the structural responses multi-story reinforced concrete building based on ground acceleration. The strong ground acceleration might cause the catastrophic collapse of the multi-story building which leads to casualties and property damages. Therefore, it is imperative to properly design the multi-story building against the seismic hazard. Seismic-resistant building design process requires structural analysis to be performed to obtain the necessary building responses. Modal response spectrum analysis is performed to simulate ground acceleration and produce structural response data for further use in the ANN. The ANN architecture comprises of 3 layers: an input layer, a hidden layer, and an output layer. Ground acceleration parameters from 34 provinces in Indonesia, soil condition, and building geometry are selected as input parameters, whereas structural responses consisting of acceleration, velocity, and displacement (story drift) are selected as output parameters for the ANN. As many as 6345 datasets are used to train the ANN. From the overall datasets, 4590 data sets (72%) are used for training process, 877 data sets (14%) for the validation process, and 878 data sets (14%) for testing. The trained ANN is capable to predict structural responses based on ground acceleration at (96%) rate of prediction and the calculated Mean-Squared Errors (MSE) as low as 1.2.10−4. The high accuracy of structural response prediction can greatly assist the engineer to identify the building condition rapidly and plan the building maintenance routinely.
Prediction of Structural Response Based on Ground Acceleration using Artificial Neural Networks
This study utilizes Artificial Neural Networks to predict the structural responses multi-story reinforced concrete building based on ground acceleration. The strong ground acceleration might cause the catastrophic collapse of the multi-story building which leads to casualties and property damages. Therefore, it is imperative to properly design the multi-story building against the seismic hazard. Seismic-resistant building design process requires structural analysis to be performed to obtain the necessary building responses. Modal response spectrum analysis is performed to simulate ground acceleration and produce structural response data for further use in the ANN. The ANN architecture comprises of 3 layers: an input layer, a hidden layer, and an output layer. Ground acceleration parameters from 34 provinces in Indonesia, soil condition, and building geometry are selected as input parameters, whereas structural responses consisting of acceleration, velocity, and displacement (story drift) are selected as output parameters for the ANN. As many as 6345 datasets are used to train the ANN. From the overall datasets, 4590 data sets (72%) are used for training process, 877 data sets (14%) for the validation process, and 878 data sets (14%) for testing. The trained ANN is capable to predict structural responses based on ground acceleration at (96%) rate of prediction and the calculated Mean-Squared Errors (MSE) as low as 1.2.10−4. The high accuracy of structural response prediction can greatly assist the engineer to identify the building condition rapidly and plan the building maintenance routinely.
Prediction of Structural Response Based on Ground Acceleration using Artificial Neural Networks
Suryanita, Reni (author) / Maizir, Harnedi (author) / Jingga, Hendra (author)
2017-04-15
oai:zenodo.org:1028074
Article (Journal)
Electronic Resource
English
DDC:
690
Prediction of blast-induced ground vibration using artificial neural networks
| British Library Online Contents | 2011
Prediction of blast-induced ground vibration using artificial neural networks
| Online Contents | 2011
A method for prediction of peak horizontal acceleration by artificial neural networks
| British Library Conference Proceedings | 1996
Prediction of seismic-induced structural damage using artificial neural networks
| Online Contents | 2009