Frequency-Adaptive Current Control with Kalman Filter-Based Observer for Multiple Grid-Connected Inverters Under Harsh Grid Distortion

As renewable energy source-based distributed generation is more widely connected to the grid, stable current control and power quality improvement in grid-connected inverters (GCIs) become more important. To satisfy increasing power demand, multi-inverter systems connected to the grid in parallel are being widely adopted. However, parallel operation may degrade current quality and stability because of inverter interactions under harsh grid conditions. In particular, grid voltage harmonics, voltage imbalance, and frequency variations can also impair current control performance and system stability. To address these concerns, a frequency-adaptive current controller integrated with a Kalman filter (KF)-based observer is developed to ensure a stable operation of multiple GCIs. Moreover, a stability evaluation is presented for multi-inverter systems by using admittance-based stability analysis. A Kalman filter-based state observer is applied to improve the estimation accuracy under noisy measurement conditions. In addition, a moving average filter-based phase-locked loop (MAF-PLL) is applied to improve the detection accuracy and reliability of the grid frequency and phase angle under harsh grid conditions to ensure an effective frequency-adaptive control design. The effectiveness and performance of the proposed current controller are assessed through the PSIM simulations. The simulation results show that the MAF-PLL reduces the maximum frequency fluctuation from ±7 Hz to ±1.1 Hz. In addition, the KF-based observer reduces the RMS estimation error to 0.0001 A. On the other hand, those values are 1.3 A with the conventional observer and 0.0003 A with the LQR-based observer, respectively. The practicality of the proposed scheme is also confirmed experimentally using 2 kW parallel multiple GCI prototype systems under harsh grid conditions.