High resolution Fabry-Pérot interferometer - dynamic system modeling and nanopositioning control system design.

• Ano de defesa: 2014
• Autor(a) principal: Molina Arcila, Ana María
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Computer systems engineering; Engenharia de sistemas de computação; Identificação de sistemas; Systems identification
• Link de acesso: http://www.teses.usp.br/teses/disponiveis/3/3139/tde-23122014-155735/

This work represents the research project to obtain the degree of Master of Sciences in Electrical Engineering, specializing in Systems Engineering, at the Escola Politécnica of the Universidade de São Paulo, in São Paulo, Brazil. The main objective of the project is to design the mirror nanopositioning controller of the state-of-the-art Fabry-Pérot interferometer to be installed in the Brazilian Tunable Filter Imager (BTFI) on the Southern Astrophysical Research (SOAR) telescope in Chile. A three-input-three-output multivariable prototype of the Fabry-Pérot system is comprised of three high-range Amplified Piezoelectric Actuators (APA) of 360 m stroke and three 400 m range capacitive measurement systems. A characterization of the instrumentation of the system, which consists of capacitive sensors and capacitance-to-voltage converters, piezoelectric actuators, power drivers of the piezoelectric actuators and data acquisition system was done as part of the identification and study of the system. With the characterization of the system, a sixth-order complete system model was built on top of a second-order piezoelectric actuator parametric model, required for the design of the controllers. Subsequently, the scientific specifications were translated to a control problem and the design of a robust controller was made following the Linear Quadratic Gaussian/Loop Transfer Recovery (LQG/LTR) method. Also a Proportional-Integral controller tuned using a genetic algorithm was designed to be used as benchmark. Finally the built controllers were validated in the real system. Results show that both controllers achieve the performance requirements of following reference signals and having null steady-state error. However, the robust controller is by far the best suited for the Fabry- Pérot instrument in terms of performance and stability because of its higher bandwidth and robustness to modeling errors.