Wyniki wyszukiwania - Biblioteka Nauki

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

3 Systems Science

Znaleziono wyników: 3

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Robust PI controller design for an uncertain recycle system

100%

Patre B. M. , Deore P. J.

Systems Science

2005

tom Vol. 31, no 4

63-74

In this paper, a method is proposed for design of a robust controller for interval process systems in the presence of parametric uncertainties. The method uses a necessary condition and sufficient condition for stability of interval polynomial. These conditions are used to derive a set of inequalities in terms of the compensator parameters which can be solved to obtain a robust controller. The method proposed is simple, involves less computational complexity and provides an easy way to obtain a robust compensator. It is applied to design a robust proportional-integral controller for a recycle system in the presence of process gain uncertainties. The results show the efficacy of the proposed method.

Comparison of neural network and decision tree induction in classification of radar returns from the ionosphere

100%

Salankar S. S. , Patre B. M.

Systems Science

2009

tom Vol. 35, no 4

61-67

This study compares the performance of decision tree (CART) and two neural network configurations, namely a well-known Multilayer Perceptron Neural Network (MLP NN) and the Radial Basis Function Neural Network (RBF NN) on the radar ionosphere databases. The task is to discriminate between radar returns from the ionosphere into "good returns" (evidence of structure) and "bad returns". It is shown that the proposed RBF neural network based classifier, consistently, has 100% accuracy on "bad" instances and 99.18% accuracy on "good" instances of testing data sets. The results prove that the proposed RBF NN classifier clearly outperforms the MLP NN and Decision Tree based classifiers on the testing datasets even after attempting different data partitions.

Robust variable structure model following controller for load frequency control

80%

Patre B. M. , Bandyopadhyay B. , Unbehauen H.

Systems Science

2002

tom Vol. 28, nr 4

43-59

A systematic procedure for design of variable structure model following load frequency controller for a single-area power system has been presented. The variable structure model following control (VSMFC) technique presented in this paper not only guarantees invariance to a class of parameter variations and disturbances but also possesses other attractive features of variable structure systems (VSS) such as robustness to unknown disturbances, simplicity in design, and reduced order dynamics when in sliding mode. The simulation results of the variable structure model following control strategy to load frequency control are presented by changing all parameters by 30% to 50% from their nominal values. These results show that systems performance is robust to parameter variations and disturbances.