Modelling and multi-objective optimization of the fresh and mechanical properties of self-compacting high volume fly ash ECC (HVFA-ECC) using response surface methodology (RSM)
Engineered cementitious composite (ECC) requires a lot of material tailoring for the desired self-compacting properties at fresh state and controlled crack width with strain hardening behavior at hardened state to be attained. Finding a balance between these two important stages (fresh and hardened)...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Elsevier Ltd
2021
|
| Online Access: | http://scholars.utp.edu.my/id/eprint/30305/ https://www.scopus.com/inward/record.uri?eid=2-s2.0-85102480674&doi=10.1016%2fj.cscm.2021.e00525&partnerID=40&md5=589ec79cd726c0fa85ea5ceb22a88340 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Engineered cementitious composite (ECC) requires a lot of material tailoring for the desired self-compacting properties at fresh state and controlled crack width with strain hardening behavior at hardened state to be attained. Finding a balance between these two important stages (fresh and hardened) of the composite without compromise on any of the desirable properties is not an easy task. This research was aimed at utilizing the technique of multi-objective optimization in trying to determine an optimal Polyvinyl alcohol (PVA) fiber volume fraction and HVFA replacement of cement (two of the most important ingredients of ECC) to develop a self-compacting HVFA-ECC (SC-HVFA-ECC) with desirable properties at both fresh and hardened stages without compromise. Using the central composite design (CCD) of RSM, 13 mixes of varying combinations of the input factors (PVA: 1�1.5 , FA: 50�70 ) were developed on which six responses (fresh state: v-funnel time, T500 and slump flow; hardened state: compressive strength, flexural strength and tensile capacity) were investigated. Six response models (5 quadratic and 1 linear) were successfully developed and validated using ANOVA. All the models turned out to have a very high R2 value ranging from 87 to 97 . The multi-objective optimization yielded optimal values of the variables (PVA: 1.67 and FA: 53.61 ) and predicted optimized values for the responses. The predicted values were experimentally validated and found to compare very well with the experimental values at less than 5 error. © 2021 |
|---|