This paper proposes an adaptive control for enhancing the compensating capability of the Dynamic Voltage Restorer (DVR). The proposed feedforward-based Backpropagation (BP) architecture is optimized by Coronavirus Herd Immunity Algorithm (CHIA) and implemented for the fundamental quantity extraction. The Fractional Order PID (FOPID) is utilized for voltage regulation of DC and AC bus under dynamics. The drawback of traditional system SRF-PI has manual intervention for tuning and mathematical modeling. The conventional SFR-PI based DVR is not a failsafe mechanism for supply voltage variation scenario. The proposed controllers, NN-CHIO and FOPID-STO overcome the issues of SRF-PI and significantly increase the tracking ability of the designed system. In the proposed design of DVR, NN layer weights are optimized by Corona Virus Herd Immunity Algorithm (CHIA) and FOPID gains are optimized by Sooty Tern Optimization Algorithm (STO). The FOPID is assessed by the integral time square error (ITSE) performance index. The FOPID (PIλDδ) has additional tuning coefficients compared to classical PI tuning and achieved less settling time, rise time, and overshoot. The controller effectiveness is examined under voltage imperfection such as sag/swell, unbalanced, and distortion. Compared to the conventional control scheme, NN-CHIO and optimized FOPID based DVR performs better by improving the output voltage profiles. The proposed system shows its effectiveness by producing less THD. Moreover, simulation and experimental results are demonstrated for verifying the effectiveness of the DVR using the developed intelligent controller.