Improving Predictions of Concrete Strength
Author Information
Author(s): Zou Zhipeng, Peng Bin, Xie Lianghai, Song Shaoxun, Torgal F. Pacheco, Sanjuán Miguel Ángel
Primary Institution: School of Environment and Architecture, University of Shanghai for Science and Technology
Hypothesis
Can an enhanced Gaussian Process model improve the prediction of compressive strength in ultra-high-performance concrete?
Conclusion
The enhanced Gaussian Process model significantly improves the prediction accuracy of compressive strength in ultra-high-performance concrete.
Supporting Evidence
- The enhanced GP model outperforms traditional methods like ANN and regression models.
- Data augmentation techniques improved the model's predictive performance.
- The model quantifies uncertainty, aiding in decision-making for engineering applications.
Takeaway
This study shows a new way to predict how strong a special type of concrete will be, which helps builders make better and cheaper concrete mixes.
Methodology
The study used an enhanced Gaussian Process model with Singular Value Decomposition and Kalman Filtering to predict compressive strength based on various mix proportions.
Potential Biases
Potential bias from excluding uncertain data points could limit the model's ability to generalize.
Limitations
Exclusion of high-uncertainty data points may introduce bias, affecting generalizability.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website