Kalman-Filter-Estimation Based Sliding Mode Control of Three-Phase LCL-Filtered Grid-Tied Inverter Using only Grid-Injected Current Sensors

The sliding mode control (SMC) has been broadly used in Grid-Tied Inverters (GTI) with LCL filter, owing to its strong robustness against parameter deviation. However, compared with conventional current control methods, SMC normally requires extra sensors to detect state variables for the designed controller. Although adopting an observer seems an effective way to save sensors, as inputs of an observer, the PCC voltage and grid-injected current sensors are typically indispensable. In this paper, a PCC-voltage-embedded model is employed in a Kalman filter to further estimate the fundamental voltages at PCC for the SMC controller. In consequence, only the grid-injected current sensors are necessary in the proposed strategy. Moreover, by applying the proposed controller, a stronger robustness against grid impedance variation and an improved transient performance can be achieved. A 3-kW experimental device designed on the Dspace DS1202 is developed to verify the feasibility of the proposed strategy.