| Idea Name: |
Artificial Intelligence |
| Slogan: |
Predicting concrete strength using Artificial Intelligence |
| Supervisor Name: |
Dr. Muhammad Shahid Siddique |
| Supervisor Designation: |
Assistant Professor |
| Supervisor School: |
MCE |
| Supervisor Department: |
Structural Engineering |
| Contact number: |
0332 0629587 |
| Email ID: |
[email protected] |
| Abstract: |
| Self-compacting concrete is a special type of concrete that flows under its own weight into place to completely fill the formwork and compact without separation. There are few methods in the literature that can predict the strength of SCC based on their mix composition. The purpose of this investigation is to demonstrate the applicability of AI to predict 28 day compressive strength of SCC. |
| What is the unmet need in society that your idea will fulfill ? |
| Despite the widespread use of SCC in construction, there is not yet a method available in the literature, which can reliably predict the compressive strength of SCC based on the mix components. This is mainly due to the highly non-linear behavior exhibited by the compressive strength. |
| Who needs it ? How many would benefit ? |
| An accurate prediction of compressive strength of self-compacting concrete will enable designers and contractors to use it in large quantities without compaction and the incorporation of various admixtures in the SCC will reduce the cost and also help the environment. |
| How will the solution works |
| This study will employ Genetic Programming (GP) techniques to formulate different equations. The resulting equations will relate the mixture proportions of SCC with the 28-day compressive strength. The GP models will be trained and validated using the data already available in the literature. After the formulation of different equations, the equations will be compared to check accuracy and practicality, and the most feasible equation will be finalized that will best relate the inputs with output |
| Who are your competitors ? How is your solution different |
| The strength of SCC has been predicted before using AI techniques, but those models don't generate an empirical equation and give no information about the process happening at the back of the models. Also the compressive strength of SCC incorporating multiple admixtures has not been studied before. |
| Status: |
new |
| Entry Date & Time: |
2022-12-12 (2229) |
