INVESTIGADORES
SANCHEZ PEÑA Ricardo Salvador
artículos
Título:
Active Noise Hybrid Time-Varying Control for Motorcycle Helmets
Autor/es:
ROSA CASTAÑÉ; RICARDO S. SÁNCHEZ PEÑA
Revista:
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY
Editorial:
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Referencias:
Año: 2010 vol. 18 p. 602 - 612
ISSN:
1063-6536
Resumen:
—Recent noise at work regulations in the EU (2003) have been established to prevent noise induced hearing loss (NIHL). This imposes better performance results to traditional feedback active noise control (ANC) in motorcycle helmets, which suffer from well known limitations. Here two new ideas are applied to this problem. First, an hybrid (feedforward/feedback) linear time invariant (LTI) controller is designed for a motorcycle helmet ANC, which improves the resulting attenuation. This is achieved by adding an extra pair of microphones which measurethe external noise that is then used as the feedforward input signal. In addition and to increase even more the resulting performance, the air velocity is measured in real-time and used as the parameter which schedules a linear parameter varying (LPV) feedback (FB) controller. This is combined with the previous feedforward (FF) controller, resulting in a time-varying hybrid controller. Both hybrid, LTI and LPV controllers are designed using linear matrix inequality (LMI)-based optimization. Two experiments have been carried out to measure the relation between external noise spectra and velocity: a wind tunnel test and a freeway ride experience.The resulting controllers are tested in a simulation which uses actual data obtained from the freeway experiment. The resulting attenuations in this motivating study seem promising for futurecontroller tests to be performed in real-time, with the adequate hardware.