Solving the Two-Dimensional Conduction Heat Transfer Problem by Using Artificial Neural Network method

In the last decade, artificial neural networks (ANNs) have made significant progress and started to meld into various knowledge fields. In the modeling and simulation fields, physic-informed neural networks (PINNs), which emerge as derivatives of ANNs, are being extensively explored their potential. Hence, this paper tried to take a closer look at ANNs for solving the partial differential equations (PDEs) in a simple 2D heat transfer problem while the time is integrated by Euler’s method. Later, results were compared to the finite difference (FD) method and exact solution. The paper’s results show simple ANNs are able to solve the problem consistently with a limited degree of success.