Abstract
This paper proposes a universal hybrid model-free quantum–transfer learning controller with enhanced online grey wolf optimization algorithm (GWO–QTL) for DC–DC boost converter. This system has the characteristics of non-minimum phase behavior, parasitic effects, and fractional-order dynamics because of high frequency operation. These characteristics make analytical modeling complicated and make it difficult to have a single traditional controller that will operate reliably over different converter types. This motivates the creation of a unified model-free control framework that is able to learn directly from the behavior of the converter without relying on the topology specific models. Reinforcement learning, where an agent interacts with an environment to learn an optimal policy, has great promise for regulating power converters, but has the disadvantages of being slow to converge, requiring a large amount of data, being prone to suboptimal local minima and lacking generalization to when operating conditions change. To address these drawbacks, the proposed approach presents a sequential multi-layer learning strategy that starts with: (i) deep transfer learning module that is implemented as the deep neural network architecture and is defined for the cross-domain policy reuse; followed by (ii) quantum-inspired action refinement mechanism that enhances efficiency of an exploration and reduces early convergence; and finally (iii) online GWO stage that is introduced as the last step in the development, which continuously searches for optimal initial gain settings to enhance learning speed and promote stable integration in the quantum–transfer learning structure. The controller is evaluated on boost converter as black box system without analytical modeling. Hardware-in-the-loop testing is performed to verify fast transient response, low steady-state error, strong disturbance rejection, and practical real-time feasibility.
Keywords
DC–DC boost converter, deep transfer learning, grey wolf optimization, hardware-in-loop, quantum-inspired reinforcement learning
Document Type
Journal Article
Date of Publication
1-1-2026
Article Number
e70263
ISSN
17554535
Volume
19
Issue
1
Publication Title
IET Power Electronics
Publisher
Wiley
School
School of Engineering
Funding Information
This work was supported and funded by the School of Engineering, Edith Cowan University.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Ghamari, S. M., & Aziz, A. (2026). A universal hybrid model-free deep quantum–transfer learning controller enhanced by grey wolf optimization for DC–DC boost converters with hardware-in-loop validation. IET Power Electronics, 19(1), e70263. https://doi.org/10.1049/pel2.70263