The Short-term Load Forecasting of Power System Based on Kalman Filter Algorithm
Recently,the frequency of crisis-break rises enormously,with electrical energy kind one of the most imminent.In order to enhance the security and stability of the power system,while guarantee the operation-friendly status,domestic and foreign scholars attach great importance to the study of loading forecasting method,among which the forecasting methods based on time series gets earlier application and has been well developed.However,forecasting methods (Such as least squares method) based on time series of short-term load of power system in the past exsited deficiencies in the forecasting accuracy and convergence rate,this has led to the poor sensitivity of the prediction model and the forecast result was not ideal.Therefore,this paper launched an in-depth research on auto-regressive AR model parameter identification method and proposed the application of Kalman filter estimation algorithm for identification of AR model.This model used auto-regressive parameter vector as state vector to build the state equation.This method does not have to know the priori information of model in advance,and the parameters of the forecasting model is updated in real time,this can improve the sensitivity of the forecasting model.Through simulation example,comparing to the east squares method,the method proposed here obtain better forecasting results.
Crisis situations Power load forecasting Kalman filter AR model Parameter identification
PENG Xiu-yan CUI Yan-qing GUAN Ruo-lin
College of Automation,Harbin Engineering University,P.R.China,150001 School of Astronautics,Harbin Institute of Technology,P.R.China,150001
国际会议
北京
英文
255-259
2012-10-17（万方平台首次上网日期，不代表论文的发表时间）