skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Real-time anomaly detection for very short-term load forecasting

; ORCiD logo;
Publication Date:
Sponsoring Org.:
OSTI Identifier:
Grant/Contract Number:
Resource Type:
Journal Article: Published Article
Journal Name:
Journal of Modern Power Systems and Clean Energy
Additional Journal Information:
Related Information: CHORUS Timestamp: 2018-01-06 12:23:37; Journal ID: ISSN 2196-5625
Springer Science + Business Media
Country of Publication:
Country unknown/Code not available

Citation Formats

LUO, Jian, HONG, Tao, and YUE, Meng. Real-time anomaly detection for very short-term load forecasting. Country unknown/Code not available: N. p., 2018. Web. doi:10.1007/s40565-017-0351-7.
LUO, Jian, HONG, Tao, & YUE, Meng. Real-time anomaly detection for very short-term load forecasting. Country unknown/Code not available. doi:10.1007/s40565-017-0351-7.
LUO, Jian, HONG, Tao, and YUE, Meng. 2018. "Real-time anomaly detection for very short-term load forecasting". Country unknown/Code not available. doi:10.1007/s40565-017-0351-7.
title = {Real-time anomaly detection for very short-term load forecasting},
author = {LUO, Jian and HONG, Tao and YUE, Meng},
abstractNote = {},
doi = {10.1007/s40565-017-0351-7},
journal = {Journal of Modern Power Systems and Clean Energy},
number = ,
volume = ,
place = {Country unknown/Code not available},
year = 2018,
month = 1

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record at 10.1007/s40565-017-0351-7

Save / Share:
  • This paper presents a practical real-time implementation of weather adaptive short-term load forecasting for distribution power utilities. The implementation is accomplished by utilizing a comprehensive load forecasting model consisting of time series, nonlinear load-weather functions and a residual load function represented by an ARMA (Auto-Regressive Moving Average) model. The model parameters are estimated and updated on-line using the WRLS (Weighted Recursive Least Squares) algorithm. A variable forgetting factor (VFF) technique is incorporated in the WRLS algorithm for improved model tracking and numerical performance in real-time operation. A software package, STLF, is developed with the proposed implementation method for distribution powermore » utilities. Practical operation of the STLF in several power utilities has demonstrated great success. Off-line testing and on-line operation has consistently shown satisfactory performance with the mean absolute error (MAE) mostly less than 2% for a less than 24-hour ahead forecast and less than 2.5% for a less than 168-hour ahead forecast.« less
  • Three practical techniques--Fuzzy Logic (FL), Neural Networks (NN), and Auto-regressive model (AR)--for very short-term load forecasting have been proposed and discussed in this paper. Their performances are evaluated through a simulation study. The preliminary study shows that it is feasible to design a simple, satisfactory dynamic forecaster to predict the very short-term load trends on-line. FL and NN can be good candidates for this application.
  • The application of time series analysis methods to load forecasting is reviewed. It is shown than Box and Jenkins time series models, in particular, are well suited to this application. The logical and organized procedures for model development using the autocorrelation function make these models particularly attractive. One of the drawbacks of these models is the inability to accurately represent the nonlinear relationship between load and temperature. A simple procedure for overcoming this difficulty is introduced, and several Box and Jenkins models are compared with a forecasting procedure currently used by a utility company.
  • The intrinsic uncertainties associated with demand forecasting become more acute when it is required to provide an invaluable dimension to the decision making process in a period characterized by fast and dynamic changes. In this paper, estimates of the peak demand, pertaining to a typical fast growing system with inherit dynamic load characteristics, have been derived from three classical time-series forecasting methods. These demand estimates are compared with corresponding actual values.
  • A knowledge-based expert system is proposed for the short term load forecasting of Taiwan power system. The developed expert system, which was implemented on a personal computer, was written in PROLOG using a 5-year data base. To benefit from the expert knowledge and experience of the system operator, eleven different load shapes, each with different means of load calculations, are established. With these load shapes at hand, some peculiar load characteristics pertaining to Taiwan Power Company can be taken into account. The special load types considered by the expert system include the extremely low load levels during the week ofmore » the Chinese New Year, the special load characteristics of the days following a tropical storm or a typhoon, the partial shutdown of certain factories on Saturdays, and the special event caused by a holiday on Friday or on Tuesday, etc. A characteristic feature of the proposed knowledge-based expert system is that it is easy to add new information and new rules to the knowledge base. To illustrate the effectiveness of the presented expert system, short-term load forecasting is performed on Taiwan power system by using both the developed algorithm and the conventional Box-Jenkins statistical method. It is found that a mean absolute error of 2.52% for a year is achieved by the expert system approach as compared to an error of 3.86% by the statistical method.« less