Developing three hybrid machine learning algorithms for predicting the mechanical properties of plastic concrete samples with different geometries

Plastic concrete is an engineering material, which is commonly used for construction of cut-off walls to prevent water seepage under the dam. This type of concrete shows great promise    to satisfy the requirements of the strength, stiffness and permeability for remedial cut-off wall construction. This paper aims to explore three hybrid machine learning algorithms including Artificial Neural Network (ANN), Support Vector Machine (SVM) and Adaptive Neuro-Fuzzy Inference System (ANFIS) optimized with Particle Swarm Optimization (PSO) to predict the compressive and splitting tensile strength of plastic concretes. To this end, data were collected from different sources and data gaps were covered by extra experimental tests and finally, 387 data for compressive strength and 107 data for splitting tensile strength were gathered for modeling. This study shows that ANN-PSO is superior to SVM-PSO and ANFIS-PSO in case of predicting compressive as well as splitting tensile strength   of plastic concretes. The coefficient of determination (R2) in case of ANN-PSO for both training and testing sets is more than 0.95. Results of this study can be used to predict the compressive and splitting tensile strength of plastic concretes with regards to constituent materials and specimen geometry of plastic concrete.