DocumentCode
3585551
Title
The Application of Neural Network Control Algorithm in RF-excited CO2 Laser Power Supply
Author
Shun Liu ; Youqing Wang ; Bo Li
Author_Institution
Nat. Eng. Res. Center of Laser Process., Huazhong Univ. of Sci. & Technol., Wuhan, China
Volume
2
fYear
2014
Firstpage
559
Lastpage
563
Abstract
In order to improve the injection power of the laser and guarantee the normal operation of the RF-excited Fast Axial Flow CO2 Laser, A kind of neural network control technology is proposed in a 2MHz laser RF power supply, which consists of two single-phase half-bridge MOSFET inverter modules. By using the SLPS Interface (SLPS), the RF power supply system with neural network controller is built and simulated. The result shows that the output power from the laser power supply can be up to 2000W, and a little of distortion (THD is 0.013%) is obtained. Moreover, it suggests that the neural network controller possesses better adaptivity than the PI controller to nonlinear, time-varying uncertain impedance characteristics of the load. So, the analysis of this paper shows that this kind of control algorithm scheme is fully applicable to the control design requirements of the laser.
Keywords
control system synthesis; gas lasers; neurocontrollers; PI controller; SLPS interface; carbon dioxide laser power supply; control algorithm; control design requirements; laser injection power; neural network control algorithm; proportional-integral controller; single-phase half-bridge MOSFET inverter modules; Discharges (electric); Inverters; Load modeling; Neural networks; Power lasers; Power supplies; Radio frequency; Half-Bridge; Lasers; Neural networks;
fLanguage
English
Publisher
ieee
Conference_Titel
Computational Intelligence and Design (ISCID), 2014 Seventh International Symposium on
Print_ISBN
978-1-4799-7004-9
Type
conf
DOI
10.1109/ISCID.2014.232
Filename
7082053
Link To Document