离散椭球分布下两阶段WCVaR风险利润优化模型及应用

Two Stages WCVaR Risk-profit Optimization Model under the Ellipsoidal Discrete Distribution and Application

本文研究随机变量非完全分布下的两阶段风险-利润优化问题。采用最坏情况下条件风险(Worst-case Conditional Value-at-Risk:WCVaR)度量指标,在离散椭球分布下建立了两阶段WCVaR约束下利润期望最大优化模型,运用优化对偶方法将复杂的Max-Min结构化简,理论上证明了简化模型和原模型的同解性,以发电商电能分配组合优化为数值实例,验证了模型和计算方法的有效性。

This paper presents two-stage risk-profit optimization problem under the know part information of random variable. Taking worst-case Conditional Value-at-Risk（ WCVaR） as a measuring index,we establish two-stage profit expectation maximization model under WCVaR constraint. By means of the dual method,the complex structure of the Max-Min becomes simple. The optimal solution between the original problem and the reduced optimization problem is proved to have the same solution. Taking optimal allocation of generation assets in power markets as numerical experiments,numerical results show the validity of the proposed model and computation method.