Bi-level cloud computing for spatial banking of sustainable mobile energy storage logistics fleet in renewable-dominated carbon-constrained urban grid

Renewable-dominated urban energy clusters require coordinated storage decisions across local systems, yet stationary batteries cannot move electricity between spatially separated surpluses and deficits. Existing formulations rarely combine sustainable mobile storage capacity sizing, correlated renewable-load uncertainty, fleet availability, battery degradation, environmental settlement, and distributed computing within one operational structure. This work developed a bi-level scheduling framework for a cluster of urban integrated energy systems. Local subproblems scheduled gas-fired generation, wind power, stationary storage, flexible demand, and grid exchange, while a cluster coordinator allocated a mobile energy storage logistic fleet, inter-system transfers, carbon trading, and green-certificate transactions. Wind and load dependence was represented by a Student-t copula; arrival, departure, and fleet-size uncertainty were incorporated through probabilistic scenarios reduced to 60 representative cases. The resulting stochastic mixed-integer linear program was tested on three interconnected systems over 24 h. A 60 kWh mobile fleet provides the most balanced lifecycle and operating outcome, reducing daily cost from USD 1332.21 to USD 1239.54, or 6.96%, while coordinated carbon and certificate trading reduces carbon dioxide emissions by 11.60%. Data aggregation lowers communication requirements by up to 76.3%, and the 24-system case is solved in 128.6 s with a 0.28% optimality gap. The results indicate that degradation-aware mobile storage banking can provide spatial flexibility, preserve distributed coordination accuracy, and support economically credible decarbonised operation in urban energy clusters.