Novel operational matrices-based finite difference/spectral algorithm for a class of time-fractional Burger equation in multidimensions.

• A hybrid computational scheme is developed assisted by spectral and finite difference methods. • Some novel operational matrices have been computed via shifted Gegenbauer polynomials. • Validation is being made through convergence and error-bound analysis. • Numerical simulations are performed for fractional time-dependent one, two, and three-dimensional Burger's equations. • Comparative analysis is presented to show the efficiency and reliability of the proposed scheme. • Method found a useful for nonlinear problems arising in Mechanics and engineering sciences. In this work, an innovative computational scheme is developed to compute stable solutions of time-fractional coupled viscous Burger's equation in multi-dimensions. To discretize the problem, the temporal derivative is approximated through a forward difference scheme whereas the spatial derivatives are approximated assisted by novel operational matrices that have been constructed via shifted Gegenbauer wavelets (SGWs). The piecewise functions are utilized to construct the operational matrices of multi-dimensional SGWs vectors although related theorems are offered to authenticate the scheme mathematically. The proposed computational algorithm converts the model understudy to a system of linear algebraic equations that are easier to tackle. To validate the accuracy, credibility, and reliability of the present method, the time-fractional viscous Burger's models are considered in one, two, and three dimensions. An inclusive comparative study is reported which demonstrates that the proposed computational scheme is effective, accurate, and well-matched to find the numerical solutions of the aforementioned problems. Convergence, error bound, and stability of the suggested method is investigated theoretically and numerically. [ABSTRACT FROM AUTHOR]