COMPRESSIVE STRENGTHS PREDICTION MODEL FOR BAGASSE ASH CONCRETE
Onchiri, Richard Ocharo
MetadataShow full item record
This paper deals with development of regression models for prediction of compressive strength of bagasse ash (BA) concrete. The compressive strength of dry BA concrete was determined using two variables, namely, curing period and pozzolanic Portland cement (PPC) content for BA grain size < 0.075 and water – PPC ratio of 0.55. The polynomial regression models were developed by varying PPC replacements (0, 5, 10, 15, 20, 25 and 30 percent) and curing periods (7, 14 and 28 days). Effort was made to modify the polynomial models using a three step approach. First step was identifying the best curve fitting technique (least square and interpolation methods). The second step of model development was to compute adjusted R2 for the polynomial Regression models. The final step was model validation. Least square quadratic polynomial (LSQP) with degree of the regression n = 4 was established as the most effective and accurate model in compressive strength prediction of BA concrete.