Applications of Computational Intelligence in Concrete Technology

Description

Computational intelligence (CI) in concrete technology has not yet been fully explored worldwide because of some limitations in data sets. This book discusses the selection and separation of data sets, performance evaluation parameters for different types of concrete and related materials, and sensitivity analysis related to various CI techniques. Fundamental concepts and essential analysis for CI techniques such as artificial neural network, fuzzy system, support vector machine, and how they work together for resolving real-life problems, are explained. Features: It is the first book on this fast-growing research field. It discusses the use of various computation intelligence techniques in concrete technology applications. It explains the effectiveness of the methods used and the wide range of available techniques. It integrates a wide range of disciplines from civil engineering, construction technology, and concrete technology to computation intelligence, soft computing, data science, computer science, and so on. It brings together the experiences of contributors from around the world who are doing research in this field and explores the different aspects of their research. The technical content included is beneficial for researchers as well as practicing engineers in the concrete and construction industry. Cover Half Title Series Page Title Page Copyright Page Table of Contents Preface Editors Contributors Chapter 1 Usage of Computational Intelligence Techniques in Concrete Technology 1.1 Introduction 1.2 Computational Intelligence Models for Concrete Technology 1.2.1 Artificial Neural Networks (ANN) 1.2.2 Adaptive Neuro-Fuzzy Inference System (ANFIS) 1.2.3 Genetic Algorithm (GA) 1.2.4 Random Forest (RF) 1.2.5 Random Tree (RT) 1.2.6 Linear Regression (LR) 1.2.7 M5P Model 1.2.8 Support Vector Machine (SVM) 1.3 Predictive Computational Intelligence in Concrete Technology 1.3.1 Prediction of Compressive Strength of the Concrete 1.3.2 Prediction of Ultrasonic Pulse Velocity of the Concrete 1.4 Conclusions References Chapter 2 Developing Random Forest, Random Tree, and Linear Regression Models to Predict Compressive Strength of Concrete Using Glass Fiber 2.1 Introduction 2.2 Modeling Techniques 2.2.1 Random Forest (RF) 2.2.2 Random Tree (RT) 2.2.3 Linear Regression (LR) 2.2.3.1 Methodology and Data Description 2.2.3.2 Performance Evaluation Criteria 2.3 Result Analysis and Discussion 2.3.1 Result of RF and RT Model 2.3.2 Result of LR Model 2.4 Intercomparison among Computing Models 2.5 Sensitivity Analysis 2.6 Conclusion References Chapter 3 Prediction of Compressive Strength at Elevated Temperatures Using Machine Learning Methods 3.1 Introduction 3.2 Dataset and Methodology 3.3 Predictive Modelling 3.4 Analysis of Results and Discussion 3.4.1 Linear Regression Method 3.4.2 Regression Tree Method 3.4.3 Boosting Method 3.4.4 Neural Network Method 3.5 Conclusion References Chapter 4 Implementation of Machine Learning Approaches to Evaluate Flexural Strength of Concrete with Glass Fiber 4.1 Introduction 4.2 Machine Learning Models 4.2.1 Random Forest (RF) 4.2.2 Bagging 4.2.3 Stochastic 4.2.4 M5P Tree 4.3 Methodology and Dataset 4.4 Model Evaluation 4.5 Result and Discussion 4.5.1 Random Forest (RF) Model Evaluation 4.5.2 Bagging Random Forest Model Evaluation (BRF) 4.5.3 Stochastic Random Forest Model Evaluation (SRF) 4.5.4 M5P Tree Model Evaluation 4.6 Results Comparison 4.7 Sensitivity Analysis 4.8 Conclusion References Chapter 5 A Comparative Study Using ANFIS and ANN for Determining the Compressive Strength of Concrete 5.1 Introduction 5.2 Soft Computing Techniques 5.3 ANN 5.4 ANFIS 5.5 Performance Assessment Indices 5.6 Results and Discussion 5.6.1 Results of ANFIS Training-Based Model 5.6.2 Results of ANFIS Testing-Based Model 5.6.3 Results of ANN-Based Model 5.7 Comparison of Models 5.8 Conclusion References Chapter 6 Prediction of Concrete Mix Compressive Strength Using Waste Marble Powder: A Comparison of ANN, RF, RT, and LR Models 6.1 Introduction 6.2 Conventional Models 6.3 Soft Computing Techniques 6.3.1 Artificial Neural Network (ANN) 6.3.2 Random Forest (RF) 6.3.3 Random Tree (RT) 6.3.4 Linear Regression (LR) 6.4 Methodology and Dataset 6.4.1 Dataset 6.4.2 Model Evaluation 6.5 Result Analysis 6.5.1 Assessment of Empirical Formula 6.5.2 Assessment of ANN Based Model 6.5.3 Assessment of RF Based Model 6.5.4 Assessment of RT Based Model 6.5.5 Assessment of LR Based Model 6.5.6 Comparison among Best Developed Models 6.6 Conclusion References Chapter 7 Using GA to Predict the Compressive Strength of Concrete Containing Nano-Silica 7.1 Introduction 7.2 Genetic Algorithm (GA) 7.3 Database 7.4 Function Approximation 7.5 Optimization Using Genetic Algorithm Technique 7.6 Availability Optimization Using Genetic Algorithm 7.7 Conclusions References Chapter 8 Evaluation of Models by Soft Computing Techniques for the Prediction of Compressive Strength of Concrete Using Steel Fibre 8.1 Introduction 8.1.1 Objectives of the Study 8.2 Soft Computing Techniques 8.2.1 Artificial Neural Network (ANN) 8.2.2 Artificial Neural Network–Cross-Validation 8.2.3 Linear Regression (LR) 8.3 Methodology and Dataset 8.3.1 Dataset 8.3.2 Model Evaluation 8.4 Result Analysis 8.4.1 Assessment of ANN-Based Model 8.4.2 Assessment of ANN–Cross-Validation (Ten-Fold)-Based Model 8.4.3 Assessment of LR-Based Model 8.4.4 Comparison among Best Developed Models 8.5 Conclusion 8.6 The Interest of Conflict Statement References Chapter 9 Using Regression Model to Estimate the Splitting Tensile Strength for the Concrete with Basalt Fiber Reinforced Concrete 9.1 Introduction 9.2 Soft Computing Technique Review 9.2.1 Gaussian Process Regression 9.2.2 Support Vector Machines (SVM) 9.2.3 Multiple Linear Regression (MLR) 9.2.4 Performance Evaluation Indices 9.2.4.1 Correlation Coefficient (CC) 9.2.4.2 Root Mean Square Error (RMSE) 9.2.4.3 Mean Absolute Error 9.3 Data Set 9.4 Material Methodology 9.5 Results and Discussion 9.5.1 Sensitivity Analysis 9.6 Conclusion References Chapter 10 Prediction of Compressive Strength of Self-Compacting Concrete Containing Silica’s Using Soft Computing Techniques 10.1 Introduction 10.2 Soft Computing Techniques 10.2.1 Artificial Neural Network (ANN) 10.2.2 Linear Regression (LR) 10.2.3 Support Vector Machine (SVM) 10.2.4 Random Forest (RF) 10.2.5 Bagging 10.3 Data and Analysis 10.3.1 Data Set 10.3.2 Evaluation Parameters 10.4 Results and Discussion 10.4.1 ANN and Ensemble ANN Model 10.4.2 LR and Ensemble LR Model 10.4.3 SVM and Ensemble SVM Model 10.4.4 RF and Ensemble RF Model 10.4.5 Inter-Comparison between Applied Models 10.4.6 Sensitivity Analysis 10.4.7 Experimental Work 10.5 Conclusions References Chapter 11 Using Soft Computing Techniques to Predict the Values of Compressive Strength of Concrete with Basalt Fiber Reinforced Concrete 11.1 Introduction 11.2 Review of Regression and Soft Computing Techniques 11.2.1 Artificial Neural Networks 11.2.2 Random Forest 11.2.3 M5P Model 11.2.4 Stochastic 11.2.5 Random Tree Model 11.2.6 Performance Evaluation Indices 11.2.6.1 Correlation Coefficient (CC) 11.2.6.2 Root Mean Square Error (RMSE) 11.2.6.3 Mean Absolute Error 11.2.6.4 Nash Sutcliffe Model Efficiency 11.3 Materials and Methodology 11.3.1 Data Set 11.4 Results and Discussion 11.4.1 Results of ANN Technique 11.4.2 Results of the Tree and Forest-Based Models 11.4.3 Comparison among ANN and Soft Computing-Based Models 11.4.4 Sensitivity Analysis 11.5 Conclusion References Chapter 12 Soft Computing-Based Prediction of Compressive Strength of High Strength Concrete 12.1 Introduction 12.2 Soft Computing Techniques Theory 12.2.1 GP 12.2.2 SVM 12.3 Data Representation Superplasticizer and Description 12.4 Results and Discussion 12.4.1 Prediction of CS by GP 12.4.2 Prediction of CS by SVM 12.4.3 Prediction of CS by LR 12.4.4 Comparison of Results 12.5 Conclusion References Chapter 13 Forecasting Compressive Strength of Concrete Containing Nano-Silica Using Particle Swarm Optimization Algorithm and Genetic Algorithm 13.1 Introduction 13.2 Problem Formulation 13.2.1 Database 13.2.2 Function Approximation 13.2.2.1 Function Approximation: Algorithm 13.2.2.2 Function Approximation: Model 13.2.2.3 Function Optimization: Model 13.3 Methodology 13.3.1 Genetic Algorithm (GA) 13.3.2 Particle Swarm Optimization 13.4 Results 13.4.1 GA Technique 13.4.2 Optimization Using PSO Technique 13.4.3 Comparison of the Results from GA and PSO Techniques 13.5 Conclusions 13.6 Conflict of Interest References Chapter 14 Prediction of Ultrasonic Pulse Velocity of Concrete 14.1 Introduction 14.2 Details of Modeling Approaches Used 14.2.1 M5P Tree (M5PT) 14.2.2 M5 Rule (M5PR) 14.2.3 Random Forest Regression (Ran-For) 14.3 Experiments Performed and Data Set 14.3.1 Ultrasonic Pulse Velocity Test (UPV) 14.3.2 Core Tests 14.3.3 Half-Cell Potential Test (HCP) 14.3.4 Carbonation Test 14.3.5 Chloride Test 14.3.6 Data Set and Analysis 14.4 Results and Discussion 14.4.1 Results from M5PR 14.4.2 M5PT 14.4.3 Result from Ran-Forest 14.4.4 Comparison of Results 14.5 Conclusion References Chapter 15 Evaluation of ANN and Tree-Based Techniques for Predicting the Compressive Strength of Granite Powder Reinforced Concrete 15.1 Introduction 15.2 Soft Computing Techniques 15.2.1 Artificial Neural Network (ANN) 15.2.2 Random Forest (RF) 15.2.3 Random Tree (RT) 15.2.4 Reduced Error Pruning (REP) Tree 15.2.5 Performance Assessment Parameters 15.2.6 Data Set 15.3 Result and Discussion 15.3.1 Results of ANN-Based Models 15.3.2 Results of RF-Based Models 15.3.3 Results of RT-Based Models 15.3.4 Results of REP Tree-Based Models 15.4 Assessment or Comparison among Soft Computing-Based Applied Models 15.5 Sensitivity Analysis 15.6 Conclusion References Chapter 16 Predicting Recycled Aggregates Compressive Strength in High-Performance Concrete Using Artificial Neural Networks 16.1 Introduction 16.2 Artificial Neural Networks (ANN) (Yadollahi 2016) 16.3 Design of Neural Networks Models 16.4 Experimental Program 16.5 Data Sets 16.6 Analytic Study Using 16.7 Result and Discussion 16.8 Conclusion Acknowledgment References Chapter 17 Compressive Strength Prediction and Analysis of Concrete Using Hybrid Artificial Neural Networks 17.1 Introduction 17.2 Material 17.2.1 Training and Testing Neural Networks with Exemplar Data 17.3 Methods 17.3.1 Data Pre-processing 17.3.2 Training Parameters and Neural Networks Structure 17.3.3 ANN training and Determination Using the BP Algorithm 17.3.4 Using GA to Evolve Neural Networks’ Initial Weights and Biases and Then Training Them Using the BP Technique 17.4 Statistical Analysis 17.5 Results 17.6 Discussion 17.7 Conclusion References Index