Abstract: The integration of large-scale renewable energy sources into the power grid has significantly increased power volatility and uncertainty, posing severe challenges to the secure and economic operation of distribution networks. To address this issue, an optimal scheduling model based on discrete probability scenarios is proposed to tackle the uncertainty of control commands in distribution network dispatch for virtual power plants. Firstly, a generalized aggregation model for typical distributed resources such as energy storage systems, air conditioning loads, and electric vehicles was established to uniformly characterize their power regulation capabilities and operational constraints. Subsequently, the uncertainty of regulation signals is discretized into a set of probabilistic scenarios using historical data, and a bi-level optimization framework is introduced. In the decision-making level, the objective is to maximize the virtual power plant's profit by determining the optimal bidding strategy for energy and frequency regulation capacity. At the execution level, upon receiving regulation signals, the goal is to minimize operational costs by rapidly allocating power adjustment tasks among various distributed resources. Finally, case study simulations are conducted for validation, with results demonstrating that the proposed method effectively enhances the economic efficiency and operational flexibility of virtual power plants under diverse uncertain environments.