Published June 26, 2023
| Version v.0.0.0
Software
Open
Code, Data and Results for "Data driven design of alkali-activated concrete using sequential learning" (Journal of Cleaner Production)
Creators
- 1. @BAMResearch
- 2. Theseen
- 3. Ghezal Ahmad Jan
- 4. Stefan
- 5. Michael
Description
This repository contains the code, data, and results associated with the paper "Data driven design of Alkali-activated concrete using Sequential Learning". The corresponding author is Christoph Völker (christoph.voelker@bam.de).
Contents
This repository contains the following files and folders:
README.md: This file, providing an overview of the repository.Data_sets/: A directory containing the AAM formulations as tabular data that have been used for the experiments in this publication as csv files. The files are:- 1_DS_cube_100_1_Strength.csv
- 2_DS_cube_100_2_Strength.csv
- 3_DS_cube_100_3_Strength.csv
- 4_DS_cube_100_4_Strength.csv
- 5_DS_cube_100_5_Strength.csv
- 6_DS_cube_150_2_Strength.csv
- 7_DS_cyl_100x200_1_Strength.csv
- 8_DS_cyl_100x200_2_Strength.csv
- 9_DS_cyl_100x200_3_Strength.csv
Results/: A directory containing the results of the experiments conducted in this publication. It has two subfolders:Model_Performance_Baseline_Python/: A directory containing the baseline benchmarking of the machine learning models as a jupyter notebook.SL_Results_Matlab/: A directory containing the results of the Sequential Learning (SL) benchmarking with the respective model pipelines. It has two subfolders:MEI_Results_Exploit/: A directory containing the result files for the SL benchmarking with the exploit strategy, along with a Matlab scriptPlot_resultsMEI.mto read and plot the results.MLI_Results_Explore/: A directory containing the result files for the SL benchmarking with the explorative strategy, along with a Matlab scriptPlot_resultsMLI.mto read and plot the results.
SLAMD_Benchmarking_App/: A directory containing the SL Benchmarking app SLAMD, which has created the SL results. It has been cloned from the repo https://github.com/BAMresearch/SequentialLearningApp and contains all necessary files and a detailedREADME.mdthat explains installation and usage of SLAMD.
Notes
This paper presents a novel approach for developing sustainable building materials through Sequential Learning. Data sets with a total of 1367 formulations of different types of alkali-activated building materials, including fly ash and blast furnace slag-based concrete and their respective compressive strength and CO2-footprint, were compiled from the literature to develop and evaluate this approach. Utilizing this data, a comprehensive computational study was undertaken to evaluate the efficacy of the proposed material design methodologies, simulating laboratory conditions reflective of real-world scenarios. The results indicate a significant reduction in development time and lower research costs enabled through predictions with machine learning. This work challenges common practices in data-driven materials development for building materials. Our results show, training data required for data-driven design may be much less than commonly suggested. Further, it is more important to establish a practical design framework than to choose more accurate models. This approach can be immediately implemented into practical applications and can be translated into significant advances in sustainable building materials development.
Files
BAMcvoelker/Green-building-materials-a-new-frontier-in-data-driven-sustainable-concrete-design-v.0.0.0.zip
Files
(639.4 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:4f83a7e24f2b4f4fb467849821c65ed7
|
639.4 kB | Preview Download |