摘要
负荷预测误差风险直接影响电网公司经营效益和供电的安全性和稳定性。通过金融风险管理工具VaR方法研究了电网企业负荷预测误差风险的问题。首先由灰色周期外延组合预测模型得出电量预测值,进而得到预测误差序列,在验证预测误差服从特定分布的基础上,在给定风险水平情况下,通过划分相应分位数求得VaR值,最后通过预测误差损失函数计算得出预测值在一定置信度和风险水平下的置信区间。本预测误差风险模型具有较高的普适性和灵活性,对任意给定的风险水平,都可以求出相应风险水平下的置信区间,满足不同风险偏好决策者的需要。实例分析表明该方法简单易用,且具有较高的精度,可以为电网公司负荷预测和风险管理提供参考。
The load forecasting error risk directly influences benefits of power companies and the security and stability of power supply. The load forecasting error risk is studied with the risk management tool of VaR. The conception of VaR is illuminated, and the forecasting error sequence is obtained with the gray-periodic extensional combinatorial model. After verifying the distribution of the error and giving a risk level, the VaR is figured out through the quartile, and the confidence intervals are determined under certain confidence degree and risk level. The model is universal and flexible in setting a risk level. Any risk level has a corresponding risk interval, which can satisfy different kinds of decision makers. The case analysis illuminates that the method is simple and effective, and can be easily used to guide the work of load forecasting and risk management for power companies.
出处
《现代电力》
2009年第3期87-90,共4页
Modern Electric Power
