.png){kind=icon}
.png){kind=icon}
Return
Event-Triggered NPC Three-Level Inverter Model Predictive Compound Control
.png){kind=icon}
PRE
AI
L
S
S
C
Z
M
DOI：10.1109/tpel.2026.3699432.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Aiming to ensure the stability of a neutral-point-clamped (NPC) three-level inverter under disturbances and parametric uncertainties, including parameter mismatches and changing of reference values. This article proposes a model-predictive hybrid control strategy based on event triggering and an extended state observer (ESO). First, the event-triggered strategy updates the control signal only when the Lyapunov-based triggering conditions are violated, significantly reducing redundant switching operations and computational requirements. Simultaneously, the ESO provides real-time disturbance estimation and compensation, thereby enhancing system robustness, improving output-voltage regulation, and maintaining neutral-point voltage balance. Second, a comprehensive performance evaluation function is proposed to objectively compare the impact of different event triggering conditions on system performance. Finally, experimental results obtained on a DSP 28377D controlled NPC three-level inverter platform demonstrate that the proposed ET-MPC+ESO strategy is optimum in comparison with traditional MPC, MPC+ESO, and ET-MPC.
Keywords:
Event trigger
extended state observer (ESO)
model-predictive control
neutral-point-clamped three-level inverter
Journal
IF:
6.5
Papers:
1.7W
Citations:
8.3W
Organization
Cited Papers