Modeling Self-Healing of Concrete Using Hybrid Genetic Algorithm–Artificial Neural Network

Abstract

This paper presents an approach to predicting the intrinsic self-healing in concrete using a hybrid genetic algorithm-artificial neural network (GA-ANN). A genetic algorithm was implemented in the network as a stochastic optimizing tool for the initial optimal weights and biases. This approach can assist the network in achieving a global optimum and avoid the possibility of the network getting trapped at local optima. The proposed model was trained and validated using an especially built database using various experimental studies retrieved from the open literature. The model inputs include the cement content, water-to-cement ratio (w/c), type and dosage of supplementary cementitious materials, bio-healing materials, and both expansive and crystalline additives. Self-healing indicated by means of crack width is the model output. The results showed that the proposed GA-ANN model is capable of capturing the complex effects of various self-healing agents (e.g., biochemical material, silica-based additive, expansive and crystalline components) on the self-healing performance in cement-based materials.

Self-Calibration Algorithm for a Pressure Sensor with a Real-Time Approach Based on an Artificial Neural Network

“…Since there is no commonly accepted optimal method to determine the best architecture of an ANN, a trial and error approach was adopted [ 32 ]. The number of neurons, number of layers, activation functions, training algorithm and the computation requirements are the major characteristics considered during the design.…”

“…ANNs are advantageous because feedforward networks are universal approximators capable of learning continuous functions with any desired degree of accuracy [ 31 ]. In most cases, the ANN model was trained using the back-propagation (BP) algorithm [ 32 ]. ANNs offer a reliable tool that can model and predict complex problems [ 33 ].…”

“…Furthermore, ANNs can be applied to a variety of tasks and problems, such as classification, interpretation, diagnosis, modelling, and control. They are more suitable to problems that are highly complex to solve by mathematical modelling or other classical procedures [ 32 ].…”

The Prediction of Self-Healing Capacity of Bacteria-Based Concrete Using Machine Learning Approaches

“…Even though a large amount of experiments have been conducted, there is no suggestion of the material design in practical use to control the crack width, which is important in civil engineering. The cracking phenomena of materials [Hamdia, Silani, Zhuang et al (2017); Nanthakumar, Lahmer, ; Vu-Bac, Lahmer, ; Zhuang, Huang, Rabczuk et al (2014)] have been investigated widely including efficient remeshing techniques [Areias, Reinoso, Camanho et al (2018); Areias and Rabczuk (2017); Areias, Rabczuk and Msekh (2016); Areias, Msekh and Rabczuk (2016); Areias, Rabczuk and Camanho (2014); Areias, Rabczuk and Dias-da-Costa (2013); Areias and Rabczuk (2013); Anitescu, Hossain and Rabczuk (2018)], phase field methods [Badnava, Msekh, Etemadi et al (2018); Msekh, Cuong, Zi et al (2017)], multiscale methods for fracture [Budarapu, Gracie, Bordas et al (2014); Budarapu, Gracie, Yang et al (2014); Talebi, Silani and Rabczuk (2015); Talebi, Silani, Bordas et al (2014)], peridynamics [Ren, Zhuang and Rabczuk (2017); Ren, Zhuang, Cai et al (2016)], DEM [Zhou, Zhu, Ju et al (2017)], meshfree methods [Amiri, Milan, Shen et al (2014); Amiri, Anitescu, Arroyo et al (2014); Rabczuk, Gracie, Song et al (2010); Rabczuk, Areias and Belytschko (2007); Zhuang, Cai and Augarde (2014); Zhuang, Zhu and Augarde (2014)], the phantom node method [Chau-Dinh, Zi, Lee et al (2012)], the smooth extended finite element method [Chen, Rabczuk, Bordas et al (2012)] as well as other partition of unity based methods like the extended isogeometric analysis [Ghorashi, Valizadeh, Mohammadi et al (2015);Nguyen-Thanh, Valizadeh, Nguyen et al (2015); Nguyen-Xuan, Liu, Bordas et al (2013)]. However, the healing effects of such materials are not well simulated.…”

“…After artificial intelligence system is put forth, it has been widely used to substitute for the traditional regression analysis method, and begins to be used in forecasting works [Das, Biswal, Sivakugan et al (2011); Gordan, Jahed Armaghani, Hajihassani et al (2016); Hoang and Pham (2016)]. Only a few researches use ML in the optimization of the self-healing concrete [Ramadan Suleiman and Nehdi (2017)]. However, a systematic comparison of the available ML algorithms still lacks since the performance differences may be substantial in their application to the optimization of the bacteria-based self-healing concrete.…”