Interval type-2 recurrent fuzzy neural system with asymmetric membership functions for chaotic system identification

In this paper, we propose an interval type-2 recurrent fuzzy neural system with asymmetric membership functions (AIT2RFNS). The proposed AIT2RFNS having the dynamic fuzzy rules and asymmetric fuzzy membership functions to enhance the performance of the interval type-2 fuzzy neural system. The AIT2RFNS is implemented as seven-layer network which consists of six feed-forward layers and a feedback layer. The feedback layer is embedded in the network by connecting to the layer 2 of the network. The feedback units act as memory elements which endue the network with the ability of copping the temporal problems. For training the AIT2RFNS, the particle swarm optimization algorithm is adopted to exam the performance. The chaotic system identification is done to show the effectiveness and the performance of the proposed AIT2RFNS. In addition, the comparison result is presented to show the superiority of AIT2RFNS.