Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Generation of voltage references using Multilayer Feed Forward Neural Network

Dehini R., Bassou A., Chellali B.

Przegląd Elektrotechniczny

2012

R. 88, nr 4a

289-292

Among the dynamic voltage restorer (DVR) objectives, the protection of sensitive loads against voltage swell/sag voltage, as well as correction of unbalanced problems. The DVR acts as a voltage source, which is connected in series with a sensitive load and controlled to generate the required compensation voltages. Many researches were presented in literature to identify the voltage disturbance; one of them is Synchronous Reference Frame (SRF) method. This paper provides a DVR with identification method based on artificial neural network - the (MLFFN) Multilayer Feed Forward Neural Network. To investigate this novel method performance, it is confronted with the conventional (SRF) method. All of the studies have been carried out using the simulation with the MATLAB Simulink Power System Toolbox. The results of simulation study of new DVR detecting method presented in this paper are found quite satisfactory to eliminate the voltage disturbance from utility voltage.

Przedstawiono dynamiczny układ przywracania napięcia DVR jako zabezpieczenie przeciw zapadom napięcia. Układ działa jako źródło napięcia połączone w szereg z obciążeniem. W artykule opisano metodę identyfikacji wykorzystującą sztuczne sieci neuronowe. Przedstawiono też symulacje systemu.

Control imrovement of shunt active power filter using an optimized-PI controller based on ant colony algorithm and swarm optimization

Berbaoui B., Ferdi B., Benachaiba C., Dehini R.

Journal of Automation Mobile Robotics and Intelligent Systems

2010

Vol. 4, No. 4

19-25

In the last years, there has been a increase currents harmonics on electrical network injected by nonlinear loads, such as rectifier equipment used in telecommunication system, power suppliers, domestic appliances, ect. This paper makes a comparison of the effectiveness of the two methods on particular optimization problem, namely. The tuning of the parameters for PI DC link voltage to a shunt active power filter. The simulation results demonstrates that the optimized PI controller by ant colony (ACO) presents a advantage of little response time and best control performances compared to the optimized PI with Particle swarm (PSO). This comparison is shown on redu cing harmonic current supply (THD).