Regarding the problem of high complexity of collaborative optimization and scheduling of microgrids

this paper proposes a hierarchical load-carbon scheduling frame for micro-grid clusters based on meteorological similar daily regression model. Firstly

a multiple linear regression model is proposed based on similar days of load pattern and meteorological factors

which provides a foundation for collaborative optimization. Then

a hierarchical scheduling model consisting of an upper-layer with energy-carbon coordination and a lower-layer with load optimization is constructed. This method decomposes the global optimization problem into the upper-layer with carbon trading and tie-line coordination and the lower-layer with economic operation and load matching in order to reduce the complexity. The upper layer and lower layer interact with each other through dynamic feedback coordination for hierarchical load-carbon collaborative scheduling of micro-grid clusters. Experimental results demonstrate that

compared with the model predictive control method and method considering spatiotemporal correlation of solar and wind power

the proposed method significantly improves the collaborative operation efficiency of micro-grid clusters

reduces system carbon emissions

and enhances overall economic performance. Additionally

the reliability of the micro-grid cluster is effectively improved in three typical operational scenarios.