Numerical and statistical analysis of transport performance of viscous flows over a wedge with heat transfer by using response surface methodology: sensitivity analysis

In this study the sensitivity analysis of viscous flow past a symmetric wedge with thermal effect is studied numerically and statistically. The partial differential equations (PDEs) are transformed into nonlinear ordinary differential equation by using similarity transformations, and these transformed ODE's are solved with the help of MATLAB built-in code bvp4c. The numerical outcomes are compared with previous existing work and were found to be correct. Then, using response surface methodology (RSM) we developed a correlation between the independent parameters and dependent output responses. Using RSM we obtained the ANOVA results for the coefficient of determination for Cf-x and Nu(x). The ANOVA shows the results of R-2 = 100% and Adj R-2 = 100% for Cf-x and R-2 = 98.73% and Adj R-2 = 97.82% for Nu(x). These results show the best correlation between input parameters and output responses. Then, we performed sensitivity analysis and found that the Falkner-Skan exponent (m) is more sensitive to the skin friction coefficient (Cf-x), while Prandtl number (Pr) is more sensitive to the Nusselt number (Nu(x)).