By Topic

A Clustering based Genetic Fuzzy expert system for electrical energy demand prediction

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Ghanbari, A. ; Dept. of Ind. Eng., Univ. of Tehran (UT), Tehran, Iran ; Ghaderi, S.F. ; Ali Azadeh, M.

Constructing an intelligent system is a demanding work, it can involve many different aspects of learning, adaptation, and control under uncertainty. Fuzzy logic is one of the major tools for scientists and engineers working in this field since it can deal with complex engineering problems which are not easy to solve by orthodox methods. On the other hand increasing worldwide demand for electrical energy requires development of advanced intelligent prediction tools. This article presents a novel load forecasting approach by integration of Genetic Fuzzy Systems (GFS) and Data Clustering for constructing a load forecaster expert system. Firstly, all records of data are inputted into the K-means model and will be categorized into k clusters. Then, all clusters will be fed into independent GFS models with the ability of rule base and data base extraction. For the purpose of evaluating our Clustering based Genetic Fuzzy System (CGFS) we apply it on Iran's load forecasting problem and compare the results with the same GFS without clustering (WGFS) and also one of the popular neuro-fuzzy type approaches called Adaptive Neuro-Fuzzy Inference Systems (ANFIS). For carrying out the comparisons we adopt three common evaluation statistics called the root mean square error (RMSE), the mean absolute error (MAE), and the mean absolute percentage error (MAPE). All evaluations indicate that CGFS provides more accurate results than WGFS and ANFIS. Besides, number of the rules generated by CGFS is lower than the other two approaches and this shows the ability of CGFS the extract optimum number of rules and as its consequence such expert system will have higher interpretability.

Published in:

Computer and Automation Engineering (ICCAE), 2010 The 2nd International Conference on  (Volume:5 )

Date of Conference:

26-28 Feb. 2010