Validation and application of an adaptive transparent defuzzification strategy for fuzzy control

An extended parametrizable defuzzification method is implemented as a special transparent neural network, which can be considered as a global defuzzification approximater. The customization of the method to different applications and the analysing capability of the trained solution are the key features here. The network is validated by showing its convergence to various existing defuzzification methods. The results of an application example and a number of less known defuzzification methods are also discussed