Title :
Comparison between Adaptive and Non-adaptive HRBF Neural Network in Multiple Steps Time Series Forecasting
Author :
Mamat, Mazlina ; Samad, Salina Abdul
Author_Institution :
Inst. of Microengineering & Nanotechnol., Univ. Kebangsaan Malaysia, Bangi, Malaysia
Abstract :
A new piezoelectric/ electrorheological hybrid hydraulic ultra-precision step actuator that uses a hybrid mode of piezoelectric stack pump hydro-drive and electrorheological valve control to the precision hydraulic cylinder is presented. It achieves high frequency, ultra-precision hydraulic step drive. The working principle of the step actuator is analyzed and its experimental characteristics are tested. Experiments show that the piezoelectric/ electrorheological cooperate ultra-precision hydraulic step actuator provides new theory and method to the development of precision step actuator with high resolution and large remove, high-speed, big load ability, it provide a basic research reference in the field of ultra-precision drive technology.
Keywords :
fuzzy set theory; learning (artificial intelligence); least squares approximations; mean square error methods; pattern clustering; radial basis function networks; recursive estimation; time series; Mackey Glass data; Santa Fe competition set A data; adaptive HRBF neural network; adaptive fuzzy c-means clustering algorithm; adaptive learning; coefficient of determination test; exponential weighted recursive least square algorithm; hybrid radial basis function neural network; mean square error test; multiple steps time series forecasting; nonadaptive HRBF neural network; nonadaptive learning; Artificial neural networks; Computer networks; Electronic mail; Fuzzy logic; Glass; Iron; Neural nanotechnology; Neural networks; Research and development; Testing; HRBF; adaptive; non-adaptive; time series forecasting;
Conference_Titel :
Computer Research and Development, 2010 Second International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-0-7695-4043-6
DOI :
10.1109/ICCRD.2010.177
