Application of Non-Linear Partial Least-Squares Regression in the Prediction of Carbonization Depth of Concrete
Based on partial least-squares multinomial regression, this paper had a prediction on carbonization depth of conctete. Taking water-cement ratio (i.e. water-ash ratio), cement content (i.e. application amount of cement material), and exposure time of concrete as independent variables , and carbonization depth of conctete as dependent variable , the forecast model of carbonization depth of conctete was obtained. It was found that, Press residual value decreased with the increase of number of latent variables, and the number of latent variables were three by Press residual value versus number of latent variables. The normal regression coefficient of exposure time of concrete was the largest wiithin three influence factors, this indicated that the influence of exposure time of concrete was largest to conctete carbonization depth; The determination coefficient of forecast model obtained in this paper was 0.9940, the error of forecast model was . The following conclusion can be drawn that, the model is accurate and credible, and the partial least-squares multinomial regression is a eximious non-linear method, and it is worthy to spread its application in the forecast analysis of conctete carbonization depth.
J. P. Jiang "Application of Non-Linear Partial Least-Squares Regression in the Prediction of Carbonization Depth of Concrete", Advanced Materials Research, Vols. 341-342, pp. 53-57, 2012