Quasi-Combinatorial Microgrid Energy Optimization Using Secure Affine-Based Blind Quantum Computing

Quantum computing possesses unique superiority in complicated energy optimization, especially in cases of microgrids with diverse discrete resources that renders corresponding energy optimization problems represents a quasi-combinatorial nature. Given that limited quantum resources are currently available as cloud services, we propose an affine-based blind quantum computing method for outsourcing quasi-combinatorial energy optimization (QCEO) problems of microgrids. First, we reveal the privacy disclosure issue posed by directly outsourcing local QCEO problems to a quantum cloud platform and design a cost-effective and easy-to-implement blind quantum computing scheme to preserve privacy-related details. Next, in the reformulation process from QCEO to quadratic unconstrained binary optimization (QUBO), we customize the secret keys composed of the coefficient matrices and vectors for affine transformation and binary expansion. The poor network structure of the original Ising model corresponding to ill-conditioned unencrypted QUBO is optimized while encrypting by the customized secret keys. Finally, with microgrid application, numerical experiments verify the effectiveness and scalability of the proposed method to accelerate the decision-making of local QCEO with privacy preservation.