DC Optimal Power Flow in Unit Commitment Using Quantum Computing: An ADMM Approach

This paper introduces an approach to solve unit commitment-based DC optimal power flow using the quantum-based alternating direction method of multipliers optimizer. The substantial interconnection of current generation and loads and the widespread incorporation of renewable energy have significantly altered and increased the complexity of the grid. As a result, efficient planning and operation of the power grid have become central focuses. This paper proposes an advanced method to solve DC-optimal power flow via quantum heuristic approaches leveraging the power of both quantum and classical computing. This is achieved by splitting the procedure to decompose a mixed binary quadratic problem into a quadratic unconstrained binary optimization sub-problem which is solved using a quantum approximate optimization algorithm. Following this, a continuous convex-constrained sub-problem is solved with a classical optimization solver. Within the DC-OPF, the startup cost and shutdown cost of each generator are incorporated in addition to the general cost function with the generator's cost coefficients. The DC-OPF is solved using the IBM Qiskit software development kit on an IEEE-14 bus system. The results obtained are then validated with the results obtained from classical methods using MATPOWER.