Short-term nodal load forecasting method considering redundant information

2018 3rd Asia Conference on Power and Electrical Engineering

能源学;电工学

Yang, Si^1 ; Zhao, Long^1 ; Han, Xueshan^2 ; Wang, Yong^2 ; Li, Wenbo^3 ; Sun, Donglei^1 ; Wang, Junxiong^2

Economic and Technology Research Institute of State Grid Shandong Electric Power Company, Jinan, China^1

Key Laboratory of Power System Intelligent Dispatch and Control of Ministry of Education, Shandong University, Jinan, China^2

State Grid Shandong Electric Power Research Institute, Jinan, China^3

关键词: Forecasting error;    Forecasting modeling;    Historical data;    Measured values;    Measurement equations;    Measurement redundancy;    State variables;    Support vector machine models;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/366/1/012046/pdf
DOI  :  10.1088/1757-899X/366/1/012046

来源: IOP

【 摘 要 】

The most existing short-term nodal load forecasting methods only focused on the application of separate historical data, ignoring the abundant redundant information in power system. In this paper a new nodal load forecasting method considered physical redundancy and measurement redundancy was proposed. Firstly, the effect of redundant information on short-term nodal load forecasting was analysed and the forecasting principle was given. Secondly, in order to form additional state variable measurement equations, two kinds of existing redundant information between state variable and measured values were analysed deeply. Then based on the forecasting principle and the analysis results, the forecasting model mainly imitated the form of state estimation was established and the specific forecasting process was given. Case studies demonstrated that compared with traditional support vector machine model, the proposed method could effectively decrease forecasting errors and improve forecasting results.