Predicting the Fundamental Period of Light-Frame Wooden Buildings by Employing Bat Algorithm-Based Artificial Neural Network

Abstract

The study utilizes an artificial neural network model for determining the fundamental period of Light-Frame Wooden Buildings, employing the Bat algorithm on a data set of 71 measured periods of wooden buildings. The number of stories, floor area, storey height, maximum length, and maximum width are selected as input parameters to estimate the fundamental period of light-frame wooden buildings. The accuracy and the competitiveness of the developed model were evaluated by comparing it with a similar particle swarm optimization (PSO)- ANN scheme, the formulas provided in the National Building Code of Canada, an equation obtained from the Eureqa software, and a non-linear regression (NLR) model. The results of the research show that the bat-ANN model exhibited the best overall performance with the lowest RMSE and MAE error values and the highest values of the Coefficient of determination, R2, in comparison to the other examined models. Therefore, the proposed Bat-ANN model can be considered as a reliable, robust, and accurate tool for predicting the fundamental period of wooden buildings.