Full-adaptive THEN-part equipped fuzzy wavelet neural controller design of FACTS devices to suppress inter-area oscillations

By incorporating Self-Recurrent Wavelet Neural Networks (SRWNNs) into Takagi–Sugeno–Kang (TSK) fuzzy model, this paper not only develops a novel Indirect Stable Adaptive Fuzzy Wavelet Neural Controller (ISAFWNC), but also uses it as a supplementary damping controller of Flexible AC Transmission System (FACTS) devices. In the proposed approach, the SRWNN is employed to construct a full-adaptive self-recurrent consequent part for each fuzzy rule of a TSK fuzzy model. A Stable Back-Propagation (SBP) algorithm with the aid of an Adaptive SRWNN-Identifier (ASRWNNI) is then employed to adjust fuzzy rules in real-time operation while the closed-loop stability is guaranteed by a Lyapunov-based approach. The proposed controller is thus able to handle the plant uncertainty by both the concepts of fuzzy logic and ASRWNNI while the local details of non stationary signals can be decomposed in terms of the dilation and translation parameters of the self-recurrent wavelet neural networks. A Genetic Algorithm (GA) based approach is proposed to choose the initial values of the dilation and the translation parameters of the wavelet and thus to increases the training speed and convergence rate of the proposed control scheme, since the BP convergence rate depends on the selection of the initial values of the network parameters. Simulations results of both two-machine two-area and benchmark four-machine two-area power systems, respectively equipped with a Static Synchronous Series Compensator (SSSC) and a Unified Power Flow Controller (UPFC) demonstrate the effectiveness of the proposed ISAFWNC design.