Robust Predictive Current Control of CBDFG Based on Ultra-Local Model

For cascaded brushless double-fed generators (CBDFG), both vector control (VC) and deadbeat predictive current control (DPCC) exhibit the potential for achieving exceptional performance. Nonetheless, the presence of inaccuracies or parameter changes in CBDFG—attributable to factors such as temperature fluctuations and saturation—may lead to a decline in control performance. To fortify the system resilience, this paper introduces a novel solution: robust predictive current control based on ultra-local model (RPCC-UL). This innovative method amalgamates the strengths of vector control and deadbeat predictive control while replacing the conventional CBDFG mathematical model with an ultra-local model. This substitution yields a marked improvement in parameter robustness. Comparative analysis between the proposed RPCC-UL and traditional DPCC is conducted, with both simulation and experimental results converging to affirm its effectiveness.