Adaptive fuzzy integrated LMS based control strategy for voltage power quality enhancement

Abstract

In this paper, an adaptive Takagi Sugeno (TS) Fuzzy integrated Mixed Order Normalized Least Mean Square/Fourth Algorithm (MONLMS/F) is proposed for the unresolved voltage power quality issues in distribution system. The proposed control utilizes the cost function of mixed order of errors, which are based on second and fourth order and the feature of cost function overcomes the demerit of traditional basic least mean square method, which is based on second-order error. The implemented control technique has two main objectives. Firstly, the fundamental estimation of correct active and reactive signals from polluted grid voltage for computing the reference load voltages and second is the DC link voltage regulation. A correct weight estimation is achieved by Mixed Order Normalized Least Mean Square/Fourth Algorithm (MONLMS/F) and voltage regulation is performed by Takagi Sugeno Fuzzy (TSF) approach. The proposed integrated control enhances the capability of system against voltage deviation from the reference voltage and successfully eliminates the harmonics from polluted grid voltages. A bio-inspired optimizer like Spotted Hyena (SH) is integrated with TSF for self-tuning of the membership functions (MFs) and fuzzy rule to adapt the system variations and regulate fluctuations in the DC link for dynamic conditions. However, auto-tunned TS-Fuzzy have a fast transient response of DC and AC link voltage based on dynamic indicators like settling time (ts=0.12 s), undershoot (Us=3.4 %) and peak overshoot (Mp=5.6 %) are compared to classical PI controller. Nonetheless, the proposed TS-MONLMS/F control strategy outperforms the other classical control with low oscillations at any power quality scenario with improved dynamic response. The significant MATLAB simulation and experimental results are evaluated, and its performance is demonstrated with the dSPACE (DS1202) processor for experimental realization.