Neuro-Fuzzy Approaches for Forecasting Electrical Load Using Additional Moving Average Window Data Filter on Takagi-Sugeno Type MISO Networks
Felix Pasila*, Ajoy K. Palit**, and Georg Thiele***
*Electrical Engineering Department, Petra Christian University, Siwalankerto 121-131, Surabaya, Indonesia
**Department of Electrical Engineering, ITEM, University of Bremen, D-28359 Bremen, Germany
***Department of Electrical Engineering, IAT, University of Bremen, D-28359 Bremen, Germany
The paper describes a neuro-fuzzy approach with additional moving average window data filter and fuzzy clustering algorithm that can be used to forecast electrical load using the Takagi-Sugeno (TS) type multi-input single-output (MISO) neuro-fuzzy network efficiently. The training algorithm with additional moving average filter is efficient in the sense that it can bring the performance index of the network, such as the sum squared error (SSE), down to the desired error goal much faster than the simple Levenberg-Marquardt algorithm (LMA). The fuzzy clustering algorithm allows the selection of initial parameters of fuzzy membership functions, e.g. mean and variance parameters of Gaussian membership functions of neuro-fuzzy networks, which are otherwise selected randomly. The initial parameters of fuzzy membership functions, which result in low SSE value with given training data of neuro-fuzzy network, are further fine tuned during the network training. Finally, the above training algorithm is tested on TS type MISO neuro-fuzzy structure for long-term forecasting application of electrical load time series.
-  J. S. R. Jang, “ANFIS: Adaptive network Based Fuzzy Inference System,” IEEE T. SMC., 23(3): pp. 665-685, 1993.
-  A. K. Palit and R. Babuška, “Efficient training algorithm for Takagi-Sugeno type Neuro-Fuzzy network,” Proc. of FUZZ-IEEE, Vol.3, pp. 1538-1543, 2001.
-  A. K. Palit, G. Doeding, W. Anheier, and D. Popovic, “Backpropagation based training algorithm for Takagi-Sugeno type MIMO neuro-fuzzy network to forecast electrical load time series,” Proc. of FUZZ-IEEE, Honolulu, Hawai, USA. Vol.1, pp. 86-91, 2002.
-  A. K. Palit and D. Popovic, “Computational Intelligence in Time Series Forecasting, Theory and Engineering Applications,” Springer-Verlag, London, pp. 230-246, 2005.
-  F. Pasila, “Forecasting of Electrical Load using Takagi-Sugeno type MIMO Neuro-Fuzzy network,” Master Thesis, University of Bremen, Germany, 2006.
-  M. Setnes, R. Babuška, and U. Kaymark, “Similarity measures in Fuzzy rule base simplification,” IEEE T. System, Man and Cybernetics, Vol.28, pp. 771-775, 1998.
-  D. Xiaosong, D. Popovic, and G. Schulz-Ekloff, “Oscillation resisting in the learning of Backpropagation neural networks,” Proc. of 3rd IFAC/IFIP, Belgium, 1995.
This article is published under a Creative Commons Attribution-NoDerivatives 4.0 International License.