目的 为应对虚拟电厂(virtual power plant，VPP)在参与电能量和需求响应市场时所面临的新能源出力和负荷不确定性问题，提出一种考虑多重不确定性的鲁棒优化调度策略，旨在降低鲁棒优化的保守性并提高VPP的经济效益。 方法 构建基于条件风险价值(conditional value at risk，CVaR)的多面体不确定性集，在此基础上，考虑风电、光伏出力和负荷的不确定性，建立VPP参与电能量和需求响应市场策略的日前两阶段鲁棒优化模型。基于行列生成(column-and-constraint generation，C&CG)算法和拉格朗日对偶理论，将所建模型分为可利用求解器求解的主问题和子问题。最后，利用蒙特卡罗方法生成大量风电、光伏和负荷数据，对所提策略进行仿真分析，并与其他方案的优化结果进行对比。 结果 所提策略采用基于CVaR的多面体不确定性集，能够充分利用历史数据，相比于采用传统不确定性集的方案，VPP的总成本降低了约2%。 结论 所提策略可以显著降低鲁棒优化结果的保守性，并在多重不确定性条件下提升VPP参与市场的经济性。

Objectives To address the uncertainties of renewable energy output and load faced by virtual power plant (VPP) when participating in electric energy and demand response markets, a robust optimal scheduling strategy considering multiple uncertainties was proposed to reduce the conservativeness of robust optimization and improve the economic benefits of VPP. Methods A polyhedral uncertainty set based on conditional value at risk (CVaR) was constructed. On this basis, considering the uncertainties of wind power, photovoltaic output and load, a day-ahead two-stage robust optimization model of VPP participating in electric energy and demand response markets was established. Then, using a column-and-constraint generation (C&CG) algorithm and Lagrangian dual theory, the model was divided into a master problem and a sub-problem that can be solved by a solver. Finally, Monte Carlo method was used to generate a large number of wind power, photovoltaic and load data. The proposed strategy was simulated and analyzed, and compared with the optimization results of other schemes. Results The proposed strategy adopting a polyhedral uncertainty set based on CVaR can make full use of historical data. Compared with the scheme using traditional uncertainty set, the total cost of VPP is reduced by about 2%. Conclusions The proposed strategy can significantly reduce the conservativeness of robust optimization results and enhance the economy of VPP participation in the market under multiple uncertainties.

针对决策者风险态度(decision-makers risk attitudes, DMRA)和不确定性对社区综合能源系统调度策略的影响，提出了社区虚拟电厂(community virtual power plant, CVPP)的多目标调度模型。首先，建立了考虑DMRA的CVPP模型和经济-能源-环境多目标满意度模型。其次，考虑可再生能源、负荷和DMRA的不确定性，对信息间隙决策理论(information gap decision theory, IGDT)模型进行了改进。第三，在考虑DMRA的基础上，拓展自信双层语言术语下的改进VIKOR方法。最后，以某居民区为例，对该模型的有效性进行了验证。结果表明：1)基于DMRA的CVPP提供了切合实际的调度策略。2)实施需求响应后，居民成本和净碳排放分别降低了9%和91%，提高了能源供应商的利润和可再生能源的利用率，所构建的IGDT模型也改进了多个目标。3)改进后的IGDT模型的不确定性和偏差因素允许采用多种调度策略。同时，改进的VIKOR方法为决策者选择策略提供了一种新的方法。该模型为调度策略的选择提供了指导，同时也为鼓励可再生能源的使用提供了途径。

To address the impact of decision makers risk attitudes (DMRAs) and uncertainties on dispatch strategies in community-integrated energy systems, a multi-objective dispatch model of a community virtual power plant (CVPP) is proposed. First, a novel CVPP model that considers DMRA and a multi-objective economics-energy-environment satisfaction model was developed. The information gap decision theory (IGDT) model is then improved considering uncertainties of renewable energy, loads, and DMRA. Third, considering DMRA, an improved VIKOR method was proposed under self-confident double-hierarchy linguistic preference relations. Finally, the effectiveness of the proposed model was validated regarding a multi-scenario example of a residential area. The results indicate the following: 1) The novel CVPP provides realistic scheduling strategies based on the DMRA. 2) After implementing demand response, the resident cost and net carbon emissions are reduced by 9% and 91%, respectively. In addition, the energy supplier profit and renewable energy utilization rate are increased. The constructed IGDT model also improves multiple objectives. 3) The improved uncertainty and deviation factors of the IGDT model allow diverse scheduling strategies. Simultaneously, the improved VIKOR method provides a new way for decision makers to select strategies. This model serves as a guide for selecting scheduling strategies and encourages the use of renewable energy sources.