Show simple item record

dc.contributor Tuzcu, Ilhan en
dc.contributor.advisor Akihiko, Kumagai en
dc.contributor.author Hunduma, Henok en
dc.date.accessioned 2013-07-08T17:18:18Z en
dc.date.available 2013-07-08T17:18:18Z en
dc.date.issued 2013-07-08 en
dc.date.submitted 2013-05-07 en
dc.identifier.uri http://hdl.handle.net/10211.9/2100 en
dc.description Thesis (M.S., Mechanical Engineering)--California State University, Sacramento, 2013. en
dc.description.abstract The purpose of this thesis is to develop Artificial Intelligence Models to predict the 28-days compressive strength of Portland cement (CCS). Two models, Artificial Neural Network and Fuzzy Logic were created using 4 input parameters of Portland cement that comprise both the physical and chemical characteristics. C3S, C2S, Alkali, and Cement fineness, were used as input variables to predict one outcome of compressive strength. Early strength prediction in the production process instead of waiting 28 days for the test to be completed could significantly improve the quality of the cement and reduce the cost associated with the waiting period. Data collected from literature was applied to predict the compressive strength of Portland cement. A rectangular mold of cement and water was created and kept in a temperature of 20℃ with 90% relative humidity for 24 hours. The cured sample was then stored in a water bath for 27 days and 6 identical bars were tested. The original data had twenty input parameters of cement with one output of compressive strength. The four most significant input parameters were selected for this particular revision. Out of the 150 generated points 100 were used to train the models while 50 data points were applied in the testing of the system. The average percentage errors achieved were 4.2% and 5.8 % for the fuzzy logic model and ANN model respectively. The results indicated that Artificial Intelligence (AI) could be a useful tool for the prediction of cement strength, and through the application of fuzzy logic algorithms, a more user friendly and more explicit model than the ANN could be produced within successful low error margins. en
dc.description.sponsorship Mechanical Engineering en
dc.language.iso en_US en
dc.subject Artificial intelligence estimation en
dc.title Estimating the compressive strength of Portland cement using artificial neural network en
dc.type Thesis en

Files in this item


This item appears in the following Collection(s)

Show simple item record

Search DSpace

My Account

RSS Feeds