Evaluation of Compressive Strength of Sustainable Concrete Using Genetic Algorithm Assisted Artificial Neural Networks

Abstract

Sustainable concrete which contains fly ash and slag is increasingly used in modern construction practices. This study presents a genetic algorithm (GA) assisted artificial neural network (ANN) model for evaluating the compressive strength of sustainable concrete. 425 mixtures are used for making the prediction system. Genetic algorithm (GA) is used to generate the initial values of the weight matrix and bias of ANN. The input parameter of GA assisted ANN is water-to-binder ratio, fly ash or slag replacement ratio, sand ratio, and water contents. The output result is compressive strength. The correlation coefficients for single ANN and GA assisted ANN model are 0.88 and 0.911, respectively. GA assisted ANN model has a strong prediction ability for the strength of sustainable concrete.

The Optimization of Distribution Path of Fresh Cold Chain Logistics Based on Genetic Algorithm

“…(2) Multisolution: it is precisely because the genetic algorithm draws on the genetic laws of the biological world, so the algorithm first randomly selects the initial solution space of a certain scale and then optimizes it through operations such as crossover, variation, and selection and selects the parent chromosomes with high adaptability from the solution space of the previous generation because the operations such as crossover, variation, and selection are accompanied by randomness. So, each iteration of a new descendant may be repeated or similar, so there will be multiple feasible solutions [ 11 ]. (3) Strong applicability: because the chromosome retention of the genetic algorithm is determined by the fitness function given in advance, the algorithm is only related to the fitness function, and the continuity of the objective function does not affect the implementation of the algorithm, so the application scope of the algorithm is more extensive.…”