Abstract
This chapter develops a new nonlinear model, ultra high frequency polynomial and trigonometric higher order neural networks (UPT-HONN) for control signal generator. UPT-HONN includes UPS-HONN (ultra high frequency polynomial and sine function higher order neural networks) and UPC-HONN (ultra high frequency polynomial and cosine function higher order neural networks). UPS-HONN and UPC-HONN model learning algorithms are developed in this chapter. UPS-HONN and UPC-HONN models are used to build nonlinear control signal generator. Test results show that UPS-HONN and UPC-HONN models are better than other polynomial higher order neural network (PHONN) and trigonometric higher order neural network (THONN) models, since UPS-HONN and UPC-HONN models can generate control signals with error approaching 10-6.
Reference45 articles.
1. Alanis, A. Y., Sanchez, E. N., & Loukianov, A. G. (2006). Discrete- Time Recurrent Neural Induction Motor Control using Kalman Learning. In Proceedings of International Joint Conference on Neural Networks (pp.1993 – 2000). Vancouver, Canada: Academic Press.
2. Real-Time Discrete Neural Block Control Using Sliding Modes for Electric Induction Motors
3. Arai, M., Kohon, R., Imai, H. (1991). Adaptive control of a neural network with a variable function of a unit and its application, Transactions on Inst. Electronic Information Communication Engineering, J74-A, 551-559.
4. Centralized Indirect Control of an Anaerobic Digestion Bioprocess Using Recurrent Neural Identifier
5. A New Neuro-FDS Definition for Indirect Adaptive Control of Unknown Nonlinear Systems Using a Method of Parameter Hopping