Authors : A. Jafarian and S. Measoomy Nia
Abstract: Fuzzy equations system plays a major role in various areas, therefore a new method for finding a crisp solution for this system is presented. So, researchers have applied an architecture of fuzzy neural network in which the corresponding connection weights are real numbers. The proposed neural net can get a fuzzy input vector and calculates its corresponding fuzzy output. Next a learning algorithm based on the gradient descent method has been defined for adjusting the connection weights. The given approach has been illustrated by several examples with computer simulations.
A. Jafarian and S. Measoomy Nia, 2013. A Learning Algorithm of Fuzzy Neural Networks for Solving A System of Fuzzy Equations. International Journal of Soft Computing, 8: 20-28.