Open Access
Review

Table 3

Power quality improvement using AI techniques.

Ref. Issues Mitigation techniques used Results
[111] Power quality issues due to active/reactive power changes and grid voltage sag in HRES. PSO-based PI controllers (offline and online tuning). Online PSO-based PI controllers reduced ITAE by 15.24% and current THD by 5.32%, achieving the lowest error and THD among the different techniques.
[112] Power quality issues in a three-phase grid-connected inverter under voltage sag conditions. PSO-based tuning of PR controller. PSO-based control parameter tuning produces high-quality output grid voltage and current waveforms, achieving a very low current THD of 0.59% when PR control parameters are tuned using PSO.
[113] Current control instability in 3-phase PV grid-connected systems under dynamic isolation changes. PSO-based self-tuning linear current controller with SVPWM. Achieved fast dynamic response with sinusoidal grid phase voltage and line current. THD was reduced to 0.38%, well below the 5% limit in IEEE Std 1547-2003.
[114] Power fluctuations and harmonic distortion in microgrid operation during load changes. PSO-based optimal PI controller for DG units. Reduced THD from 14.26% (non-optimized) to 11.76%. The system exhibited fewer power fluctuations, improved voltage stability, and faster response.
[40] Unbalanced and distorted three-phase supply voltages and load currents in a power distribution network. PSO-based state feedback controller for UPQC. Maintained THD for source currents and load voltages within permissible limits (<5%) under varying conditions. Improved robustness and performance compared to LQR-based controllers.
[58] High THD levels in a grid-connected 100 kW PV system under varying solar irradiation. Fractional Order PI controller tuned using Grey Wolf Optimization (GWO). THD was reduced to 4.42% at 600 W/m2, 2.39% at 800 W/m2, and 1.22% at 1000 W/m2, meeting IEEE standards (<5%). DC link voltage stabilized at 1450V.
[115] High harmonic content under non-linear and unbalanced load conditions in a hybrid renewable energy grid-connected system. Hybrid Shunt Active Power Filter (HSHAPF) optimized with GWO-based Fractional Order PI controller. THD reduced from 27.43% (without filter) to 1.74% when GWO-tuned HSHAPF is used, with a good power factor improvement.
[116] High THD in supply current, poor dynamic response in SAPF ACO-based optimization of PI controller for SAPF THD was reduced to 3.85%, 2.92%, and 3.49% for phases A, B, and C, respectively. Settling time improved with ACO (287 ms vs. 1140 ms using conventional PI).
[117] Voltage distortion and high THD in autonomous microgrid ACO-tuned controller with synchronous reference frame PI control THD reduced from 35.43% to 4.63%. Voltage regulation improved with V-f control in islanding mode.
[78] High harmonic distortion and overshoots during DG injection or disconnection; poor power sharing. Grasshopper Optimization Algorithm (GOA)-based controller. Compared to the PSO-based controller, the GOA-based controller reduced the overshoot in active power from 180% to 94.4% and load power overshoot from 108.5% to 12.96%, significantly improving dynamic response.
[79] Harmonic distortion, voltage instability, and reactive power issues in MG due to RES variability. Adaptive Switched Filter Compensator (ASFC) with GOA-tuned PID controller. The proposed ASFC achieved a Voltage THD of 1.19% when compared with 3.79% for the traditional SFC, significantly reducing the processing time.
[86] Voltage and frequency regulation in an autonomous microgrid during load changes and island mode operation. Bacterial Foraging Optimization Algorithm
(BFOA)
The proposed power control strategy successfully regulated the voltage to 220 V and the frequency to 50 Hz after a short transient (voltage regulation achieved after 0.3s, frequency regulation after 0.4s).

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.