Fuzzy Logic Modeling for Strength Prediction of Reactive Powder Concrete

Compressive strength forms the major property which ensures safety and stability in the design of any concrete structure. Addition of admixtures makes concrete of higher strength, which is based on trial-and-error combinations. In the present study, an attempt is made for developing a tool for compressive strength prediction of reactive powder concrete by Mamdani-based fuzzy logic interface system (FIS). The eight main parameters which influencing the strength of concrete were considered as input variables. Database set of 100 data was collected from different literature reviews and worked with trial permutation and combination with different order of material inputs, and 125 rules are set. Twenty-five test results are examined to check the efficiency of the proposed tool and compared with the FIS output by applying various membership functions using both centroid and bisector methods of defuzzification. The predicted results show the potential efficiency of FIS in prediction of compressive strength for reactive powder concrete. The results obtained were satisfactory with high accuracy ranging from 95 to 99%.