With the gradual penetration of clean energy into microgrids and nanogrids, a two-stage optimization scheduling strategy based on multi-agent game theory is proposed to address various issues such as collaborative optimization among nanogrids, microgrids, and main grids, improvement of clean energy utilization efficiency, promotion of clean energy consumption, and economic reliability of energy use. This strategy covers diversified trading and microgrid operator pricing mechanisms. First, multiple nanogrid load models with significant energy consumption characteristics are established, forming a nanogrid system model that supports autonomous scheduling of multiple nanogrids and energy trading between them. Secondly, considering the diversity of electricity supply and demand, a microgrid operator model is constructed for different energy supply scenarios, and a competition mechanism between the main network and the operator is introduced. In this game model, microgrid operators play the role of leaders, while the network management system acts as followers, achieving a game equilibrium of one master and one slave. By optimizing the output methods of microgrid operators, the energy potential of nanogrid systems is stimulated. Finally, the rationality and effectiveness of the proposed scheme are verified through numerical examples. This scheme has successfully reduced the operating cost of the nanogrid system, improved the clean energy consumption capacity of the microgrid system, and increased economic efficiency. It provides practical and effective guidance for the sustainable development of microgrids and nanogrids.