Mahdi Behroozmand
Mahdi Behroozmand has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Gholam Hossein Riahy
Department of Electrical Engineering, Amirkabir University of Technology, Tehran, Iran
2 shared publications Ashkan Pourghorbani
1 shared publication
Latest Publications
- Open Access
- 28 September 2026
- Published by Elsevier in SSRN Electronic Journal
Hybrid PI–Super-Twisting Current Controller for Robust Single-Stage Grid-Connected Photovoltaic Inverters
The proposed controller is implemented in a single-stage three-phase grid-connected photovoltaic inverter (GCPVI) and evaluated in the MATLAB/Simulink environment under three representative operating scenarios, including nominal operation, filter inductance degradation caused by aging, and severe balanced three-phase short-circuit faults. Its performance is systematically compared with conventional PI and Deadbeat predictive current controllers through both qualitative waveform analysis and quantitative performance indices, including total harmonic distortion (THD), current ripple, DC-link voltage ripple, tracking error, transient overshoot, settling time, FRT capability, together with implementation complexity considerations. Simulation results consistently demonstrate the superiority of the proposed controller over both benchmark methods under all investigated operating conditions. Compared with the conventional PI controller, the proposed Hybrid PI–STA controller reduces grid-current THD from 1.68% to 1.54%, decreases steady-state current ripple by approximately 20%, and lowers DC-link voltage ripple from 0.14% to 0.02%. Under a 30% filter inductance reduction, it maintains the smallest tracking error (0.19 pu) without controller retuning. During severe three-phase grid faults, it reduces DC-link voltage overshoot by approximately 71% and shortens the post-fault recovery time by 94.7% compared with the conventional PI controller while preserving implementation complexity comparable to that of classical PI control. These results verify that the proposed Hybrid PI–STA controller successfully combines the simplicity of conventional PI control with the robustness of higher-order sliding-mode control, providing an effective robust control solution for next-generation high-performance GCPVSs operating under uncertain and highly dynamic grid conditions.
- 28 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessHybrid PI–Super-Twisting Current Controller for Robust Single-Stage Grid-Connected Photovoltaic Inverters
The proposed controller is implemented in a single-stage three-phase grid-connected photovoltaic inverter (GCPVI) and evaluated in the MATLAB/Simulink environment under three representative operating scenarios, including nominal operation, filter inductance degradation caused by aging, and severe balanced three-phase short-circuit faults. Its performance is systematically compared with conventional PI and Deadbeat predictive current controllers through both qualitative waveform analysis and quantitative performance indices, including total harmonic distortion (THD), current ripple, DC-link voltage ripple, tracking error, transient overshoot, settling time, FRT capability, together with implementation complexity considerations. Simulation results consistently demonstrate the superiority of the proposed controller over both benchmark methods under all investigated operating conditions. Compared with the conventional PI controller, the proposed Hybrid PI–STA controller reduces grid-current THD from 1.68% to 1.54%, decreases steady-state current ripple by approximately 20%, and lowers DC-link voltage ripple from 0.14% to 0.02%. Under a 30% filter inductance reduction, it maintains the smallest tracking error (0.19 pu) without controller retuning. During severe three-phase grid faults, it reduces DC-link voltage overshoot by approximately 71% and shortens the post-fault recovery time by 94.7% compared with the conventional PI controller while preserving implementation complexity comparable to that of classical PI control. These results verify that the proposed Hybrid PI–STA controller successfully combines the simplicity of conventional PI control with the robustness of higher-order sliding-mode control, providing an effective robust control solution for next-generation high-performance GCPVSs operating under uncertain and highly dynamic grid conditions.
- Open Access
- 4 August 2026
- Published by Elsevier in SSRN Electronic Journal
Integrated Modeling and Head-Aware Dispatch of a Wind Pumped Hydro Energy Storage with Droop-Based Frequency Support
We develop and validate a unified wind-pumped hydro energy storage (PHES) framework that co-optimizes day-ahead energy management and primary-frequency response within a single, physics-informed control stack. A head-aware supervisory layer shifts charge–discharge thresholds via a linear SOC-head surrogate and enforces explicit anti-chatter measures (deadband, dwell, ramp limits). On top, an SOC-gated droop layer delivers frequency support while honoring actuator and energy constraints. The model is exercised over five 24-h scenarios: variable and sustained wind, fast fluctuations, low-head (offline) operation, and a frequency event, and is evaluated using discharged, curtailed and unserved energy, round-trip-efficiency change, equivalent full cycles (EFCs), switching events, and frequency metrics. Under nominal head, reservoirs sized to avoid day-ahead limitation render head penalties non-binding; baseline and proposed policies yield near-identical Eout, Ecurt, and Edef while chatter is restrained. Under low head the controller correctly suppresses generation, shifting performance into Edef without destabilizing modes. During the frequency event the framework preserves primary capability (nadir ≈ 49.85 Hz, RoCoF ≈ −0.023 Hz/s, recovery < 1 min, droop-support energy ≈ 0.0062 MWh), matching the baseline. Sensitivity-guided tuning exposes a reproducible parameter set (droop ≈ 4 MW/Hz; deadband 0.05 Hz; SOC thresholds 0.2–0.3 and 0.7–0.8; head-gain ≈ 0.05) that balances prudence and responsiveness and transfers directly to islanded or weak grids. The contribution is an explicit head-aware energy policy unified with an SOC-gated droop in a single reproducible model, achieving baseline-equivalent frequency performance without optimization overhead and enabling transparent, physics-based tuning.
- 4 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessIntegrated Modeling and Head-Aware Dispatch of a Wind Pumped Hydro Energy Storage with Droop-Based Frequency Support
We develop and validate a unified wind-pumped hydro energy storage (PHES) framework that co-optimizes day-ahead energy management and primary-frequency response within a single, physics-informed control stack. A head-aware supervisory layer shifts charge–discharge thresholds via a linear SOC-head surrogate and enforces explicit anti-chatter measures (deadband, dwell, ramp limits). On top, an SOC-gated droop layer delivers frequency support while honoring actuator and energy constraints. The model is exercised over five 24-h scenarios: variable and sustained wind, fast fluctuations, low-head (offline) operation, and a frequency event, and is evaluated using discharged, curtailed and unserved energy, round-trip-efficiency change, equivalent full cycles (EFCs), switching events, and frequency metrics. Under nominal head, reservoirs sized to avoid day-ahead limitation render head penalties non-binding; baseline and proposed policies yield near-identical Eout, Ecurt, and Edef while chatter is restrained. Under low head the controller correctly suppresses generation, shifting performance into Edef without destabilizing modes. During the frequency event the framework preserves primary capability (nadir ≈ 49.85 Hz, RoCoF ≈ −0.023 Hz/s, recovery < 1 min, droop-support energy ≈ 0.0062 MWh), matching the baseline. Sensitivity-guided tuning exposes a reproducible parameter set (droop ≈ 4 MW/Hz; deadband 0.05 Hz; SOC thresholds 0.2–0.3 and 0.7–0.8; head-gain ≈ 0.05) that balances prudence and responsiveness and transfers directly to islanded or weak grids. The contribution is an explicit head-aware energy policy unified with an SOC-gated droop in a single reproducible model, achieving baseline-equivalent frequency performance without optimization overhead and enabling transparent, physics-based tuning.