灰色组合模型在中长期电力负荷预测中的应用

Combination gray model in mid-term and long-term load forecasting

针对中长期电力负荷预测使用的历史数据较少,且影响因素较为复杂等特点,提出了一种基于最优权重组合(weightoptimal combination,WOC)的灰色组合模型。该模型利用最优权重组合将灰色线性回归模型、二次指数模型、等维新息模型、残差灰色模型4种模型进行组合,采用二元最优组合的迭代寻优算法来求取WOC中每个模型的权重,充分发挥几种模型各自的优点。通过在湖南省某地电力系统规划中的实际应用,验证了此方法有较好的实用性和有效性。

For mid-term and long-term load forecasting use small amount of historical data and the factors are very complicated,a gray combination model based on optimal weight （WOC） was proposed.This model uses the optimal weight combination to combine the following model,such as the gray linear regression model,quadratic exponential model,dimensional information model and residual gray model.It uses the dual optimal combination of iterative optimization algorithm to obtain the weight of each model in WOC and gives full play of the merits of those models.Through a practical application of power system planning in Hunan province,it is verified that this method has a high performance in practicality and effectiveness.