Training Recurrent Neuro-Fuzzy System Using Two Novel Population-Based Algorithms for Temperature Forecasting
Computer and Information Technology, International Conference on (2010)
Bradford, West Yorkshire, UK
June 29, 2010 to July 1, 2010
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CIT.2010.101
In this paper a new structure of a Mamdani recurrent neuro-fuzzy system (MRNFS) model is used to temperature forecasting problem. The model considers two recurrent properties, dynamic rules and the feedback connections which added in the defuzzification layer. The operational parameters of this model are trained using hybrid learning algorithm in which gradient descent (GD) algorithm is used to train the output membership functions (MFs) values and two novel population-based algorithms consist of the improved version of honey bees optimization (HBO) and breeding swarms (BS) algorithm are used to train the antecedent parameters of MRNFS model. The trained MRNFS is then used to predict the future weather conditions. This paper shows a comparison between improved HBO and BS for training the MRNFS model for temperature forecasting process. The simulation results demonstrate that the model can make predictions with high degree of accuracy and it is found that the proposed method is very effective.
breeding swarms, improved honey bee optimization, Mamdani recurrent neuro-fuzzy system, population-based algorithms, temperature forecasting
Z. Khanmirzaei, "Training Recurrent Neuro-Fuzzy System Using Two Novel Population-Based Algorithms for Temperature Forecasting," 2010 IEEE 10th International Conference on Computer and Information Technology (CIT), Bradford, 2010, pp. 438-445.