[1M4-GS-10] AI application

[1M4-GS-10-05] Electrical Distribution Networks Optimization by QAOA

〇Yuki Kimura¹, Shuhei Sugimura¹, Keisuke Fujii² (1. MEIDENSHA CORPORATION, 2. Osaka University)

Keywords:Electrical Distribution Networks, Quantum Approximate Optimization Algorithm (QAOA)

Recently, with the expansion of the introduction of renewable energy, the management of electrical distribution networks has become complicated due to the rapid spread of distributed power resources. The distribution network optimization problem is a combinatorial problem of determining the supply route so as to minimize the power loss as much as possible by controlling the switching states of the switch gears included in the distribution network. However, as the number of switch gears increases, the number of combinations grows exponentially. Conventionally, the meta-heuristics method has been used for the problem of such a huge number of combinations, but it has the problem that it may fall into a local optimum. In recent years, there has been active research into applying quantum computers to such problems that are difficult to search globally. As one of them, variational quantum algorithms such as VQE and QAOA using universal quantum computers are attracting attention. Variational algorithms can solve advanced combinatorial optimization problems as they are without any special formulation by using the versatility of universal quantum computers, and may be able to handle various types of optimization problems. In this paper, we show how QAOA is applied to distribution network optimization problems and the results of actual application.