Evolutionary Optimization Algorithm Based Optimal Tuning of Cascaded Control Scheme for Single-Stage AC-DC Solid-State-Transformers

The cascaded control loops of a grid-connected Solid-state-transformer (SST) are conventionally tuned through small-signal-model (SSM) based root-locus or frequency-response analyses, which often leads to sub-optimal tuning of respective control loops. Evolutionary algorithm based optimal tuning of the cascaded control strategy for a single-stage AC-DC SST is investigated in this paper. Particle-swarm-optimization (PSO) and genetic-algorithm (GA) based optimization techniques are used for offline tuning of the SST’s secondary power balance control (PBC) through iterative minimization of the concerned control loop’s steady-state integral-squared-error (ISE) obtained via detailed co-simulation model. The obtained optimal controller tuning parameters are plugged into the single-stage SST-based interface’s full-scale, 22 kV utility grid-connected, 1 MVA simulation model to demonstrate the performance of the PSO and GA based optimally tuned controller compared to the conventional SSM based counterpart, which elicits the merits of evolutionary algorithm based optimal tuning of cascaded control.