Neural Analysis of Concrete Shrinkage
Results of tests on laboratory specimens were selected from the data bank described in . The tests on concrete plates concerning the shrinkage strains up to 435 days at nearly stationary humidity and temperatures were accepted as a data to compare the experimental results with the Eurocode 2 recommendations . The comparison was made by means of three neural networks: BPNN-Rprop, RBFNN and ANFIS. It was stated that the RBFNN approximation is close to EC curves.
KeywordsShrinkage Strain Laboratory Specimen Gaussian Membership Function Neural Network Toolbox Concrete Plate
Unable to display preview. Download preview PDF.
- Denmuth H, Beale M (1998) Neural network toolbox for use with Matlab, User’s Guide, Version 3. The Math WorksGoogle Scholar
- Denmuth H, Beale M (1998) Fuzzy logic toolbox for use with Matlab, User’s Guide, Version 3. The Math WorksGoogle Scholar
- ENV 1992–1–1: (1992) Eurocode 2, Design of concrete structures, P.1: General rules and rules for buildingsGoogle Scholar
- Ligeza W, (2002) Redistribution of internal forces in braced concrete plate elements (in Polish), Dept. Civil Eng., Cracow Univ. of Technology, Monograph No 277.Google Scholar