Browsing by Author "Wang, Chih-Hu"

Now showing 1 - 3 of 3

Adaptive Two-Stage Fuzzy Temperature Control for an Electroheat System
(2009-03-01)
Wang, Chih-Hu
;
Lin, Chun-Hung
;
Lee, Bore-Kuen
;
Liu, Chien-Nan Jimmy
;
Su, Chauchin
In this paper, we construct a simple adaptive two-stage fuzzy temperature tracking controller for a commercial electroheat system. Being not well isolated from the environment, it is hard to build an analytical model for the system due to heat convection caused by a fan for heat mixing in the chamber, heat leakage to the environment, variation of the environment temperature, and uncertain nonlinear heating dynamics. Also, as a commercial product by using cheap heater coils and without equipping a cooling system, this system usually has a bad transient response such as a long rise time, a large overshoot, and a long settling time. Moreover, temperature tracking at the steady-state phase is not easy to maintain due to heat interaction with the environment. Here, in the presence of the unknown system dynamics, we use a two-stage fuzzy controller to improve the transient response. Furthermore, a simple fine-tuning adaptive control scheme is proposed to overcome environmental influence and ensure tracking of the temperature setting. Simulation study and experimental results show good performance of the adaptive fuzzy temperature control system.
H-infinity output feedback control of stochastic T-S fuzzy model with state-dependent noise
(2008-01-01)
Ho, Shih-Ju
;
Wang, Chih-Hu
;
Lee, Bore-Kuen
;
Su, Chauchin
;
Liu, Chien-Nan Jimmy
Estimator which guarantee H-infinity robustness performance for the considered nonlinear stochastic systems can be obtained by solving bilinear matrix inequalities. Then, to solve the bilinear matrix inequalities, a two-stage method is adopted to separately obtain the controller gain matrices and the eatimator gain matrices by solving two sets of linear matrix inequalities. An example is given to verify the derived results.
Long-Range Prediction for Real.-Time MPEG Video Traffic: An H(infinity) Filter Approach
(2008-12-01)
Wang, Chih-Hu
;
Chen, Bor-Sen
;
Lee, Bore-Kuen
;
Lee, Tsu-Tian
;
Liu, Chien-Nan Jimmy
A novel prediction scheme is proposed for real-time MPEG video to predict the burst and long-range dependent traffic. The trend and periodic characteristics of MPEG video traffic are fully captured by a proposed stochastic state-space dynamic model. Then a recursive H. filtering algorithm is proposed to estimate traffic for long-range prediction. Simulation results based on real MPEG traffic data show that the proposed scheme has superior performance and lower complexity than some adaptive neural network methods, such as TDNN, NARX, and Elman neural networks.