Quantum parallel multi-layer Monte Carlo optimization algorithm for controller parameters optimization of doubly-fed induction generator-based wind turbines

With larger searching space, intelligent algorithms could have insufficient global search capability and accuracy for optimization problems. This paper proposes a novel quantum parallel multi-layer Monte Carlo optimization algorithm (QPMMCOA) to optimize the rotor-side controller (RSC) parameters based on proportional-integral of a doubly-fed induction generator (DFIG) for achieving maximum power and improving generation efficiency. The QPMMCOA is proposed to find optimal solutions in an accurate small searching space. The QPMMCOA combines qubit probability amplitude with Monte Carlo random numbers to generate a diverse population and expand accurate search space. The optimization process of the QPMMCOA with strong global search ability and local development ability is divided into the rough search, precise search, and re-precise search. These three search processes are searched in the ever-shrinking feasible region. The QPMMCOA mainly seeks the optimized solution by continuously changing and narrowing the feasible region. The QPMMCOA is utilized to optimize a discontinuous step function and a multimodal function for confirming the efficacy and feasibility. With broader exploration and deeper development capabilities, the QPMMCOA can achieve the optimization result of the fitness function of the RSC is at least 0.51% lower than other algorithms. Compared with other algorithms, the QPMMCOA-based RSC can enhance the average power coefficient of the DFIG-based wind power system by at least 0.0028%. (C) 2021 Elsevier B.V. All rights reserved.

Automated summary

This paper introduces a novel quantum parallel multi-layer Monte Carlo optimization algorithm (QPMMCOA) to optimize the rotor-side controller (RSC) parameters based on proportional-integral of a doubly-fed induction generator (DFIG) for achieving maximum power and improving generation efficiency. The QPMMCOA combines qubit probability amplitude with Monte Carlo random numbers to generate a diverse population and expand accurate search space, showing strong global search ability and local development ability.