Efficiency Comparison of Neural Networks, Support Vector Machine and Neuro-Fuzzy with Fast Fourier Transform for Waveform Analysis and Classification
Article Sidebar
Main Article Content
Abstract
This research presents a comparison of the efficiency of three models, namely, the model using Artificial Neural Networks technique (ANN), Support Vector Machine models (SMOreg) and Neuro-Fuzzy using ANFIS technique models in form of pattern recognition for waveform analysis and classification between good signals and bad signals data set. In principle, the three models are proposed. It used Fast Fourier Transform data for learning and testing process for fi nding the best model. The three experiments were ANN, SMOreg and ANFIS. Results from the MSE training were 1.45E-08, 5.60E-08, 3.32E-09, the MAE
were 8.28E-05, 4.31E-05, 3.20E-05, the MAPE were 2.99636, 0.69080, 0.83541, respectively, and the MSE test results were 8.30E-09, 2.66E-07, 3.21E-09, the MAE were 8.61E-05, 3.32E-04, 3.38E-05, and the percentage obtained from MAPE were 2.5807, 10.9384 and 0.8061, respectively. It was found that the ANFIS technique was the best. With the smallest overall error value than the other models, which provides the highest percentage of Efficiency Index (EI) in the testing process were 99.999993 %. The experimental result of testing showed that all data sets can distinguish between good and bad data sets correctly.
Article Details
References
[2] Pecharanin, N., Sone, M., and Mitsui, H. (1994). An Application of Neural Network for Harmonic Detection in Active Filter. Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN’94). Vol. 6, pp. 3756-3760. DOI: 10.1109/ICNN.1994.374807
[3] Kittisak, K. and Somchat, J. (2010). The Use of Power Quality to Forecast Reliability in the PEA’s Distribution System Using Artificial Neural Networks. The Journal of KMUTNB. Vol. 20, No. 1, pp. 55-63
[4] Porramate, I. and Somchat, J. (2008). Application of Artificial Neural Network in Optimal Power Flow. The Journal of KMUTNB. Vol. 18, No. 2, pp. 14-23
[5] Adisorn, K., Prajuab, I., and Wutthichai, S. (8-10 March 2017). Application of Adaptive Network-Based Fuzzy Inference System with Fast Fourier Transform for Waveform Analysis and Classification. 2017 International Electrical Engineering Congress (iEECON). 8-10 March 2017, Pattaya, Thailand. DOI: 10.1109/IEECON.2017.8075886
[6] Apichit, K., Jirasak, C., and Surapan, Y. (2006). Real Time ECG Compression/Decompression System with Multirate Digital Signal Processing. Thesis of King Mongkut’s University of Technology North Bangkok (Medical Instrumentation). pp. 6-7
[7] Jos Arrillaga, Bruce C. Smith, Neville R. Watson, and Alan R. Wood. (1997). Power System Harmonic Analysis. John Wiley and Sons. pp. 7-31
[8] Dech, T. and Phayung, M. (2011). Ensemble Data Classification Based on Decision Tree, Artificial Neuron Network and Support Vector Machine Optimized by Genetic Algorithm. The Journal of KMUTNB. Vol. 21, No. 2, pp. 293-303
[9] Lin, H. C. (2004). Intelligent Neural Network Based Dynamic Power System Harmonic Analysis. International Conference on Power System Technology, 2004. PowerCon 2004. Vol. 1, pp. 244-248. DOI: 10.1109/ICPST.2004.1460000
[10] R. Rojas. (1996). Neural Network: A Systematic Introduction. Springer. Berlin: Springer-Verlag Berlin Heidelberg.
[11] Narimol, S. and Somchat, J. (2006). Short-term Load Forecast using Artificial Neural Networks. The Journal of KMUTNB. Vol. 16, No. 2, pp. 37-42
[12] Chanwit, T. and Sirivit, T. (2013). Fault Detection and Protection of Induction Motor Using Artificial Neural Network. The Journal of KMUTNB. Vol. 23, No. 1, pp. 42-51
[13] Adisorn, K., Panida, L., and Ariya, N. (2015). An Artificial Neural Network with Technical Indicators for Stock Investment: Application to the Stock Exchange of Thailand. Information Technology Journal. Vol. 11, No. 1, pp. 53-63
[14] J.-S.R. Jang and Chuen-Tsai Sun. (1995). Neuro-Fuzzy Modeling and Control. Proceedings of the IEEE. Vol. 83, Issue 3, pp. 378-406. DOI: 10.1109/5.364486
[15] Nunthakorn, C., Panida, L., and Ariya, N. (2015). A Neuro-Fuzzy with DCA and VA for Stock Investment: Application to the Stock Exchange of Thailand. Information Technology Journal. Vol. 11, No. 1, pp. 64-73
[16] Shevada, S. K., Keerthi, S. S., Bhattacharyya, C., and Murthy, K. R. K. (2000). Improvements to the SMO Algorithm for SVM regression. IEEE Transactions on Neural Networks. Vol. 11, NO. 5, pp. 1188-1193. DOI: 10.1109/72.870050