A restrained optimal perturbation method for solving a kind of inverse problem with variable coefficient

Abstract

A restrained optimal perturbation method is proposed to study one-dimensional variable coefficient backward inverse heat conduction problem. We determine the initial temperature distribution from final measurement data. Owning to the ill-posedness of this problem, a regularization term is introduced in the objective function, which based on the thought of regularization technique. And we give a brief description about the application in Genetic regulatory networks. Numerical experiments show that the method is feasibility in the determination of initial condition. © 2014 Springer International Publishing Switzerland.