Research on flexible load configuration of hydrogen production and storage and operation optimization of distribution network based on bi-level optimization

INTRODUCTION: High penetration of wind and photovoltaic generation causes voltage deviation, line congestion, and renewable curtailment in distribution networks. Hydrogen production and storage offer flexible regulation capabilities that can effectively improve system operational stability. OBJECTIVES: This study aims to construct a bi-level optimization framework for capacity allocation and day-ahead scheduling in a wind–PV–hydrogen coupled system. The goal is to enhance voltage quality, reduce thermal loading, and increase renewable energy utilization. METHODS: The upper level minimizes investment cost, network loss, and voltage deviation through a particle swarm optimization algorithm. The lower level solves a mixed-integer dynamic scheduling model in YALMIP considering grid trading cost, curtailment penalty, and peak-valley regulation benefits. RESULTS: Simulation results show that the minimum node voltage rises from 0.913 p.u to 0.952 p.u, while the maximum line loading decreases from 0.903 to 0.714 after introducing hydrogen flexibility. The system load curve becomes smoother and renewable consumption significantly improves. CONCLUSION: The proposed bi-level approach effectively enhances system flexibility and reduces operational constraints in distribution networks. The method offers practical support for coordinated planning of high-penetration renewable energy systems.