Control and Integration of PV and Energy Storage Systems for STATCOM Applications

Abstract

The growing penetration of photovoltaic (PV) systems and the displacement of synchronous generators are creating voltage regulation and stability challenges at both the distribution and transmission levels of the electric power system. This dissertation develops advanced control and integration strategies for converter-based static synchronous compensators (STATCOMs) that transform PV inverters and modular multilevel converters (MMCs) into multifunctional grid assets capable of simultaneous renewable energy harvesting, dynamic voltage support, and energy storage management. At the distribution level, a layered 𝑑𝑞-frame vector control architecture is developed for a two-level voltage source converter (VSC) operated as a PV–STATCOM, enabling seamless transitions among Full PV, Partial STATCOM, and Full STATCOM operating modes with headroomaware supervision for daytime and nighttime operation. This baseline control is extended to weak, unbalanced, and harmonically distorted grids through a dual-loop flexible compensation architecture with an adjustable flexibility factor that enables continuous transition between harmonic rejection and compensation, with stability proven across the full parameter range via Bode and Nyquist analysis. Under the worst-case combined disturbance scenario at a shortcircuit ratio of 3, the PCC voltage THD is reduced to 1.1% and the voltage unbalance factor from 11.2% to 3%. An SVM-based islanding detection framework with dual-SVM PHEV discrimination achieves detection times of 40–60 ms—over 30 times faster than the IEEE 1547 requirement—and the complete grid-following to grid-forming to reconnection lifecycle is demonstrated without mode-boundary transients. A finite-control-set model predictive control (FCS–MPC) architecture replaces the conventional cascaded PI structure, directly selecting the optimal converter switching state at each sampling instant without a modulator. An online grid-inductance estimator, integrated within the MPC prediction loop, converges within two sampling periods after step changes in grid impedance and enables reconstruction of the grid voltage behind the supply impedance, reducing the voltage THD used for prediction from 17–18% to 1–3%. Controller hardware-in-the-loop (C–HIL) validation on a Typhoon HIL 404 platform confirms real-time executability within the 60 𝜇s sampling period. At the transmission level, a 60 kVA MMC-based STATCOM is designed with analytically derived stability boundaries in the 𝑖𝑑 –𝑖𝑞 current plane. A comparative assessment of four sequence extraction methods (DDSRF, DSOGI, IDSOGI, FMB) under distorted grid conditions yields a hybrid FMB+IDSOGI strategy that combines fast current-loop dynamics with robust voltage synchronization. The MMC is integrated with PV generation and battery energy storage through a Triple Active Bridge (TAB) dc–dc converter, with Jacobian-based decoupled power-flow control enabling independent PV and ESS current regulation and a coordinated dispatch algorithm managing energy flow with SOC-aware PV curtailment. The integrated system is validated through electromagnetic transient simulation at full-scale parameters and experimental testing on a scaled-down laboratory prototype across six operating scenarios, including fault-tolerant controller fail-over under active power flow.