Simulation of a Helicon plasma source using a multivariable fuzzy model

• Published in: IEEE transactions on plasma science Vol. 26; no. 1; pp. 104 - 112
• Main Authors: Zibo Zhang, Rayner, J.P., Cheetham, A.D., Lund, T.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.02.1998; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Automatic control; Computers in experimental physics; Exact sciences and technology; Genetic algorithms; Instruments, apparatus, components and techniques common to several branches of physics and astronomy; Magnetic field measurement; Mathematical models; Neural networks, fuzzy logic, artificial intelligence; Physics; Physics of gases, plasmas and electric discharges; Physics of plasmas and electric discharges; Plasma; Plasma materials processing; Plasma production and heating; Plasma simulation; Plasma sources; Power system modeling; Predictive models; Simulation; System testing; Fuzzy control; Material processing; Simulation; Process control; Plasma devices; Plasma sources; Genetic algorithms; Fuzzy systems
• ISSN: 0093-3813; 1939-9375
• DOI: 10.1109/27.659539

This paper reports work on the development of an automatic control system for a Helicon plasma processing source. The lack of a definitive physical model for the plasma physics of the source and the power coupling mechanism to the plasma precludes the use of traditional control algorithms. This paper develops a fuzzy model that simulates the behavior of the plasma source using the process of genetic algorithms to identify and optimize the parameters of the fuzzy model. This type of model will eventually be used to test a fuzzy control system for the plasma source. In this work, an extensive set of experimental data was acquired where the magnetic field and input power to the plasma source were varied over a wide range while the electron number density was measured. From this learning dataset, the genetic algorithm derived the values of the parameters for the difference equation that describes the system. The fuzzy model so constructed was used to predict the behavior of the source from known input parameters. Comparing the predictions with experimental observations showed that the fuzzy model was generally able to predict the behavior of the plasma as its input parameters were varied with a precision of better than 10%.