A Comparative Study on ANFIS and Fuzzy Expert System Models for Concrete Mix Design

The aim of this study is to design ANFIS model and Fuzzy Expert System for determination of concrete mix design and finally compare their results. The datasets which has been loaded into ANFIS contains 552 mix designs and has been based on ACI mix designs. Moreover, in this study, a Fuzzy Expert System has been designed. Input fields of fuzzy expert system are Slump, Maximum Size of Aggregate (Dmax), Concrete Compressive Strength (CCS) and Fineness Modulus (FM). Output fields are quantities of water, Cement, Fine Aggregate (F.A) and Course Aggregate (C.A).In the ANFIS model, there are 4 layers (4 ANFIS models) that first layer takes values of Dmax and Slump and then determines the quantity of Water, second layer takes values of Water (that computed in previous layer) and CCS to measure the value of Cement, third layer takes values of Dmax and F.M to computes the measure of C.A and 4 layer takes values of Water, Cement, C.A (which determined in previous layers) and the measure of concrete density to compute the value of F.A. Comparison of two systems (FIS and ANFIS) results showed that results of ANFIS model are better than Fuzzy Expert System when these systems designed and tested. In ANFIS model, for Water output field, training and average testing errors are 0.86 and 0.8, respectively. In cement field, training error and average testing error are 0.21 and 0.22, respectively. Training and average testing error of C.A are 0.0001 and 0.0004, respectively and finally, F.A’s training and average testing errors are 0.0049 and 0.0063, respectively.

[1]  L J Kaetzel,et al.  EXPERT/KNOWLEDGE-BASED SYSTEMS FOR CEMENT AND CONCRETE: STATE-OF-THE-ART REPORT , 1991 .

[2]  Qingzhong Liu,et al.  Predicting injection profiles using ANFIS , 2007, Inf. Sci..

[3]  Moncef L. Nehdi,et al.  Fuzzy logic approach for estimating durability of concrete , 2009 .

[4]  Mustafa Saridemir,et al.  Prediction of compressive strength of concretes containing metakaolin and silica fume by artificial neural networks , 2009, Adv. Eng. Softw..

[5]  M. C. Nataraja,et al.  A Fuzzy-Neuro Model for Normal Concrete Mix Design , 2006, Eng. Lett..

[6]  Duc Truong Pham,et al.  Computational Intelligence: for Engineering and Manufacturing , 2007 .

[7]  Kun,et al.  FUZZY LOGIC MODEL FOR PREDICTION OF COMPRESSIVE STRENGTH OF LIGHTWEIGHT CONCRETE MADE WITH SCORIA AGGREGATE AND FLY ASH , 2008 .

[8]  Harun Tanyildizi,et al.  Fuzzy logic model for the prediction of bond strength of high-strength lightweight concrete , 2009, Adv. Eng. Softw..

[9]  Mahmut Bilgehan,et al.  A comparative study for the concrete compressive strength estimation using neural network and neuro-fuzzy modelling approaches , 2011 .

[10]  Solomon Tesfamariam,et al.  Adaptive Network — Fuzzy Inferencing to Estimate Concrete Strength Using Mix Design , 2007 .

[11]  Tayfun Uygunoğlu,et al.  A new approach to determination of compressive strength of fly ash concrete using fuzzy logic , 2006 .

[12]  Jing Xu,et al.  Concrete strength evaluation based on fuzzy neural networks , 2004, Proceedings of 2004 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.04EX826).