Multi-power-level collaborative optimization considering high-entropy energy predictive supply in intelligent connected transportation system

To maximize the efficient use of distributed energy, this paper proposes a multi-power-level collaborative scheduling model for Intelligent Connected Transportation System (ICTS), integrating high-entropy energy output forecasting. First, the interdependency between high-entropy energy and traffic flow is examined, and a dynamic spatial–temporal transformer model is applied to predict high-entropy energy output. Second, a hierarchical scheduling model for ICTS is developed based on the power-load characteristics of ICTS. The high-power-level scheduling layer employs a two-stage adaptive robust optimization (ARO) model to optimize the output of wind turbines (WT) and photovoltaic (PV) systems. Meanwhile, the low-power-level scheduling layer utilizes chance-constrained goal programming (CCGP) to address the uncertainty of high-entropy energy, based on predicted error distribution. The multi-power-level system is interconnected through energy routers to ensure operational stability. Finally, case study results demonstrate that the proposed strategy enhances the robustness and economic performance of ICTS, providing a reference for collaborative optimization in ICTS under uncertain environments.