This paper proposes a mathematical optimization approach based on semidefinite programing (SDP) for optimal operation of a microgrid (MG) equipped with renewable energy resources (RERs) as well as energy storage systems (ESSs) including compressed air energy storage (CAES) and electric vehicles (EVs). The SDP converts nonlinear and non-convex models proposed for day-ahead optimal operation into a convex approximation, which is easily implemented by commercial software. An optimization problem is formulated such that it minimizes the costs related to operation and environmental effects limited to several technical limitations. It is clear that there is an intrinsic deviation between predicted and actual uncertainty variables in MG. This paper presents a stochastic optimal operation model based on Information Gap Decision Theory (IGDT) with risk averse strategy to overcome this information gap and to help Microgrid Operator (MGO). As demand response programs (DRPs), the shiftable load and time of use (ToU) are considered for enhancing the flexibility of MG. Employing the presented model in the 21-bus test system shows that CAES and EVs, as novel ESSs, have good capability to reduce the costs. The total cost, including operation cost and emission cost can be reduced by 3.8% compared to without considering these ESSs in the day-ahead optimal operation.

本文提出了一种基于半定规划 (SDP) 的数学优化方法，用于微电网 (MG) 的优化运行，该微电网 (MG) 配备可再生能源 (RER) 以及包括压缩空气储能 (CAES) 和电力的储能系统 (ESS)。车辆（EV）。SDP 将针对日前最优操作提出的非线性和非凸模型转换为凸近似，这很容易通过商业软件实现。优化问题被公式化，使得它最小化与操作和环境影响相关的成本，受限于若干技术限制。很明显，MG 中预测的不确定性变量和实际的不确定性变量之间存在内在偏差。本文提出了一种基于信息差距决策理论 (IGDT) 的随机最优运行模型和风险规避策略，以克服这种信息差距并帮助微电网运营商 (MGO)。作为需求响应程序 (DRP)，考虑可转移负载和使用时间 (ToU) 以提高 MG 的灵活性。在 21 总线测试系统中采用所提出的模型表明，CAES 和 EV 作为新型 ESS 具有良好的降低成本的能力。与在日前优化运行中不考虑这些 ESS 相比，包括运行成本和排放成本在内的总成本可降低 3.8%。在 21 总线测试系统中采用所提出的模型表明，CAES 和 EV 作为新型 ESS 具有良好的降低成本的能力。与在日前优化运行中不考虑这些 ESS 相比，包括运行成本和排放成本在内的总成本可降低 3.8%。在 21 总线测试系统中采用所提出的模型表明，CAES 和 EV 作为新型 ESS 具有良好的降低成本的能力。与在日前优化运行中不考虑这些 ESS 相比，包括运行成本和排放成本在内的总成本可降低 3.8%。