The increase in utility-scale PV generation and the displacement of synchronous machines reduce system strength, reactive power headroom, voltage resilience, and overall power system stability, motivating a robust comparison of various mitigation technologies beyond static load-flow or PV assessments. RMS time-domain simulations are performed for balanced and unbalanced contingencies, and performance is quantified using post-fault voltage dip depth, undervoltage area (V < 0.9 pu.), recovery time to nominal, and RoCoF. These metrics are aggregated into a single weighted composite severity score, which is then normalised to the baseline to form the dynamic voltage resilience index (DVRI) and the Frequency Disturbance Relative Index (FDRI). The results show that the converter-based reactive power support devices deliver the fastest and most controllable post-fault voltage restoration, with the STATCOM achieving the lowest composite penalty and best DVRI under severe fault conditions but the poorest FDRI during PV plant trip/reconnection events. The synchronous condenser (SC) improves post-fault recovery through excitation driven reactive capability and increased short-circuit contribution, but its recovery to nominal voltage levels is slower and can produce negative-sequence current under unbalanced fault conditions whilst producing the smallest frequency disturbance and best FDRI. The SVC provides effective steady-state regulation but becomes less effective during extremely low voltages due to the voltage-dependent reactive power output, and its FDRI remains close to baseline. The BESS-GFM is dependent on the inverter current limits and the control priorities, which influence both voltage recovery and response times, achieving an FDRI scoring second to the SC. These metrics are combined into baseline normalised composite indices (DVRI and FDRI) using explicitly dimensionless sub-metrics (dip magnitude, exposure area, and recovery delay for voltage and deviation magnitude, windowed RoCoF, and exposure for frequency). Equal weights are used as a neutral baseline, and a weight sensitivity study is included to confirm that technology rankings are robust to plausible variations in weighting choice.
Pather et al. (Tue,) studied this question.