Developing a forecasting model of concrete compressive strength using relevance vector machines

 
 
 
  • Abstract
  • Keywords
  • References
  • PDF
  • Abstract


    We analyze results of two experiments that tested effect of adding Silica on the compressive strength of concrete at early stage and after long period. The two experiments evaluated different silica/cement ratios for different mixing periods. Adding Silica to concrete mix produce high early strength material which is highly desirable in airports and highways.

    More than 90 samples of different silica/cement ratios are tested for compressive strength at 3 and 28 days. Test results showed high early up to 60 MPa. Strength increase is proportional with the increase of silica/cement ratio and mixing time with maximum at ratio of 15/100 and 30 minutes mixing time.

    A relevance Vector Machine (RVM) model is developed to predict concrete compressive strength using concrete mixture inputs information. RVM model predictions matched experimental data closely. The developed model can be used to predict compressive strength in future periods based on initial information related to cement mixture.

    Keywords: Relevance Vector Machine, Silicate Percent, Prediction Model, Milling Time, Compressive Strength, Concrete.


  • References


      Khalil, A., McKee, M., Kemblowski, M., and Asefa, T., 2005. Sparse Bayesian learning machine for real-time management of reservoir releases. Water Resources Research. Vol. 41 (11), W11401.
    1. Khalil, A., McKee, M., Kemblowski, M., Asefa, T., and Bastidas, L., 2006. Multiobjective analysis of chaotic dynamic systems with sparse learning machines. Advances of Water Resources. Vol. 29(1), pp.7288.
    2. Mati? D., F. Kuli?, M. Pineda-Snchez, and I. Kamenko, Support vector machine classifier for diagnosis in electrical machines: Application to broken bar, Expert Syst. Appl., vol. 39, no. 10, pp. 8681-8689, 2012.
    3. MacKay D. J., Bayesian methods for adaptive models, Ph.D. Thesis, California Institute of Technology, 1992. Online at: http://resolver.caltech.edu/CaltechETD:etd-01042007-131447.
    4. Suetani H., Ideta A. M., and J. Morimoto, Nonlinear structure of escape-times to falls for a passive dynamic walker on an irregular slope: Anomaly detection using multi-class support vector machine and latent state extraction by canonical correlation analysis, Proceeding IEEE Int. Conference on Intelligent Robots and Sys., pp. 2715-2722, 2011.
    5. Suliman M.R. and M. Awwad, 2000. Utilizing of Silica in Early High Strength Concrete. Cement and Concrete Technology in the 2000s, Second International Symposium, 6-10 September, 2000, Istanbul, Turkey.
    6. Suliman K. M.R. And M. T. Awwad, 2005. Virgin Silica improves the durability of Portland cement concrete. Journal of Engineering Scuence, Assiut University. Vol. 33, No. 1, pp 1-9. January.
    7. Tipping M. E., The relevance vector machine, Advances in Neural Information Processing Systems, Vol. 12, pp. 652-658, 2000.
    8. Tipping M. E., Sparse Bayesian learning and the relevance vector machine, J. of Machine Learning Research, vol. 1, pp. 211-244, 2001.
    9. Vapnik V. N., the Nature of Statistical Learning Theory, New York: Springer-Verlag, 1995.

 

View

Download

Article ID: 2011
 
DOI: 10.14419/ijet.v3i2.2011




Copyright © 2012-2015 Science Publishing Corporation Inc. All rights reserved.