| Title |
Design of LERP-Based IT2TSK Fuzzy Prediction System with 2nd order Trend Difference and its Application to Electric Power Load Forecasting |
| Authors |
방영근(Young-Keun Bang) ; 이철희(Chul-Heui Lee) |
| DOI |
https://doi.org/10.5370/KIEE.2020.69.8.1237 |
| Keywords |
electric power load; linear interpolation; 2nd order trend difference; IT2TSK; footprint of uncertainty |
| Abstract |
This paper deals with the electric power load prediction system. The first step is, in original data, making of additional information necessary to the system design by using the linear interpolation method. Next is generating the valid input data in which the rapidly increasing trend is mitigated by the 2nd order trend difference method. Final step is designing the IT2TSK fuzzy prediction system that can properly handle the uncertainty involved in the generated data. To design IT2TSK fuzzy prediction system, IT2 fuzzy sets, K-means clustering algorithm, and least square method are used individually for expressing the footprint of uncertainty, tuning the IT2 fuzzy sets, and identifying the parameters of TSK regressive model. In simulation, the performance and effectiveness of the proposed system are verified by comparing with other system after analyzing the prediction results of two types of electric power load data. |