Modified Artificial Neural Networks For Solving Fuzzy Differential Equations

Abstract

In this paper, we introduce a novel approach based on modified  neural networks  to solve fuzzy differential equations. Using modified  neural network makes that training points should be selected over an open interval  without training the network in the range of first and end points. Therefore, the calculating volume involving computational error is reduced. In fact, the training points depending on the distance selected for training neural network are converted to similar points in the open interval  by using a new approach, then the network is trained in these similar areas. In comparison with existing similar neural networks proposed model provides solutions with high accuracy. The proposed method is illustrated by three numerical examples.

Keywords: Fuzzy  differential  equation, Modified  neural  network,

Feed-forward  neural  network, BFGS Teqnique, Hyperbolic tangent   function.