Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
The cost optimization of structures is one of the options employers consider when designing and analyzing them. Shear walls are one of the factors that influence the total cost of a structure. In this paper, an optimization model based on the BP Neural Network optimized by the genetic algorithm is proposed to optimize the dimensions and location of shear walls, which greatly improves the performance of the BP network and the accuracy can reach more than 96%. At first, ETABS software was used to design and analyze a nine-story concrete building based on ACI318-11 regulations for different shear wall positions and dimensions. In the following, the software outputs have been used for pseudo-neural training to develop a suitable model for estimating seismic performance and shear wall costs of reinforced concrete frames. The cost of the shear wall is considered as a function of wall width, wall length, wall height, concrete characteristic strength, steel yield stress, transverse reinforcement compression ratio, and longitudinal reinforcement compression ratio. To determine the optimal parameters for the shear wall, including its length and position, the genetic algorithm has been used after training and testing the neural network. The results of the present study show that with the proposed method, the presented design can reduce shear wall costs by about 40%. The outcomes of the presented method indicate that using the mentioned method decreases the wall length for walls 1 to 4 with the values of 27%, 11%, 40%, and 19.3%, respectively.
The cost optimization of structures is one of the options employers consider when designing and analyzing them. Shear walls are one of the factors that influence the total cost of a structure. In this paper, an optimization model based on the BP Neural Network optimized by the genetic algorithm is proposed to optimize the dimensions and location of shear walls, which greatly improves the performance of the BP network and the accuracy can reach more than 96%. At first, ETABS software was used to design and analyze a nine-story concrete building based on ACI318-11 regulations for different shear wall positions and dimensions. In the following, the software outputs have been used for pseudo-neural training to develop a suitable model for estimating seismic performance and shear wall costs of reinforced concrete frames. The cost of the shear wall is considered as a function of wall width, wall length, wall height, concrete characteristic strength, steel yield stress, transverse reinforcement compression ratio, and longitudinal reinforcement compression ratio. To determine the optimal parameters for the shear wall, including its length and position, the genetic algorithm has been used after training and testing the neural network. The results of the present study show that with the proposed method, the presented design can reduce shear wall costs by about 40%. The outcomes of the presented method indicate that using the mentioned method decreases the wall length for walls 1 to 4 with the values of 27%, 11%, 40%, and 19.3%, respectively.
Application of artificial neural networks and genetic algorithm in optimization of concrete shear wall design
Int J Interact Des Manuf
LI, LI (Autor:in)
01.09.2024
11 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
| BASE | 2009
A concrete mix proportion design algorithm based on artificial neural networks
| Online Contents | 2006
A concrete mix proportion design algorithm based on artificial neural networks
| British Library Online Contents | 2006