Prediction of the module of elasticity of green concretes containing ground granulated blast furnace slag using hybridized multi-objective ANN and Salp swarm algorithm

Abstract

Ground granulated blast furnave slag (GGBFS) once used in concrete derives both technical and economic advantages. Energy consumption and greenhouse gas emissions can be reduced significantly if cement is replaced by GGBFS in concrete mixtures. However, it is necessary to develop a detailed model in order to evaluate the elastic modulus of the concretes containing GGBFS because of important role of its value as a parameter in different design codes. In addition, it can save energy, cost, and time in comparison to direct laboratory-based measurements. In this research, to develop a model for the estimation of the elastic modulus of concretes containing GGBFS, Artificial neural network (ANN) was used. A multi-objective optimization method titled multi-objective slap swarm algorithm (MOSSA) was proposed to optimize the error and complexity of the developed ANN models. To develop predictive models of elastic modulus. Besides, one of the most used classification techniques to solve engineering problems that is The M5P model tree algorithm was used in order to develop predictive models of elastic modulus. The efficiency of the proposed model developed based on the ANN algorithm was compared with that of the model developed based on the M5P model tree technique with the help of several error measures. The result of this research is that it is possible using the M5P model tree and the proposed ANN model for the purpose of provide predictive tools for estimating the elastic modulus of concretes containing GGBFS and these would have 13.36% and 1.79% mean absolute percentage error (MAPE), respectively. It is understood from these values that the proposed model based on ANN algorithm is much more effective than the one developed using M5P model tree.