Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Przegląd Elektrotechniczny

2 2016

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

The T-polynomial approach for LQG control applied to a Switched Reluctance Motor (SRM)

Correla W. B., Carvalho Almeida R. N., Silva W. A., Torrico B. C., Pinto V. P., Reis L. L. N. dos

Przegląd Elektrotechniczny

2016

R. 92, nr 2

161--167

This paper proposes a tuning procedure that combines the well known T-polynomial, from predictive control, with the LQG algorithm. The proposed scheme was successfully applied to a speed loop of a switched reluctance motor (SRM), which is known by its specific issues related to driving and control. Such characteristics make the SRM a strong set up to exploit the proposed solution. Results are compared with both the predictive controller approach GPCBC and the classical LQG/LTR, analysed regarding the transient and disturbance rejection. Power analysis of the control signal shows a power efficiency improvement favourable for the proposed approach.

Opisano sposób sterowania silnikiem reluktancyjnym wykorzystująca metodę T-wielomianu z algorytmem LQG – linear quadratic Gaussian. Metodę porównano z innymi metodami jak GPCBC I LQG/LTR pod kątem tłumienia zakłóceń i stanów chwilowych.

Battery discharge forecast applied in Unmanned Aerial Vehicle

Souza D. A., Pinto V. P., Nascimento L. B. P., Torres J. L. O., Gomes J. P. P., Sá Junior J. J. M., Almeida R. N. C.

Przegląd Elektrotechniczny

2016

R. 92, nr 2

185--192

This paper proposes a comparative study methodology for the prediction of Li-Po (Lithium Ion Polymer) batteries discharge in UAVs (Unmanned Aerial Vehicles) using four approaches based on Artificial Neural Networks (ANNs) using Multilayer Perceptron (MLP) and Extreme Learning Machine (ELM) techniques, Polynomial Regression, and Kalman Filter (KF). The information estimates are important to assist in making decisions on which missions can be addressed to UAVs when supplied by such batteries. The data series for the experiments are obtained from tests carried out on a test bench.

W artykule przedstawiono stadium porównawcze metodologii przewidywania procesu rozładowania baterii litowych wykorzystujących sieci neuronowe. Baterie te były stosowane w pojazdach typu dron.