中文摘要：

以条件风险价值（CVaR）作为风险量测，以对数正态分布模拟未来市场电价，使用实际购电数据，为购电商LDC（地方配电公司）建立一种新型的均值一CVaR模型．对其在3个市场的购电作优化组合和风险评估，并与均值一方差模型所得结果作比较．结果表明，所建立模型能在保证一定的成本约束下使LDC承担的CVaR风险最小，较均值一方差模型提供更能反映实际风险的结果．

英文摘要：

With CVaR （conditional value at risk） as the risk measurement, a novel mean-CVaR model for the local distribution company （LDC） was proposed. Based on the actual market data, future electricity prices are assumed to be lognormal distributed. Then this model was applied to optimize procurement portfolio and assess the risk for the LDC in three markets. Moreover, it was compared with the mean-variance model in the original reference. The simulation results demonstrate that the proposed model can guarantee the LDC to bear the minimum CVaR risk within the expected purchase cost. It provides the results more reflecting the real risk than the mean-variance model.