基于结构化负荷模型的电力负荷概率区间预测

为了考虑电力负荷的不确定性,概率和区间预测成为电力负荷预测的重要方式之一.针对传统的负荷概率及区间预测方法没有考虑不同负荷成分的不确定性对电力负荷影响的问题,在分析电力负荷成分的基础上,基于结构化电力负荷模型提出一种电力负荷概率及区间预测方法.首先,对电力负荷的成分进行分析,针对不同负荷成分分别进行建模,构成结构化电力...

Probabilistic Interval Forecasting of Power Load Based on Structured Load model

Probability interval forecasting has become one of the main methods for power load forecasting because of the uncertainties of power load. In order to solve the problem that the conventional probability interval forecasting methods cannot consider the impact of the uncertainties of different components on the power load, a structural power load probability interval forecasting method is proposed based on the time-series state space model. Firstly, based on an analysis of the components of power load, models are established respectively for different load components. And then, based on the historical load data, the variational bayesian inference is used to train the posterior probability distribution of model parameters. Finally, the probability distribution of the future power load is predicted based on the trained model, thus realizing the forecasting of the power load probability interval. The proposed method is verified using the real power load data and compared with other methods. The experimental results show that the proposed method has a higher forecasting interval coverage probability and a narrower forecasting interval average width.