Sliding mode analysis applied to improve the dynamical response of a spherical 3D wind sensor for Mars atmosphere

• Published in: Sensors and actuators. A. Physical. 2017-11, Vol.267, p.342-350
• Main Authors: Atienza, Maria-Teresa, Kowalski, Lukasz, Gorreta, Sergi, Jiménez, Vicente, Castañer, Luis M., Domínguez-Pumar, Manuel
• Format: Article
• Language: English
• Subjects: Atmospheric models; Closed loop systems; Diffusive representation; Enginyeria electrònica; Instrumentació i mesura; Mars; Mars atmosphere; Sensors; Sensors i actuadors; Sliding mode control; Sliding mode controllers; Studies; Temperature measurements; Termometria; Time constant; Time response; Time-varying systems; Wind sensors; Àrees temàtiques de la UPC
• ISSN: 0924-4247; 1873-3069
• DOI: 10.1016/j.sna.2017.09.044

•The improvement in the dynamical response under constant temperature control of a wind sensor for Mars is proposed.•Sliding mode controllers theory is used to justify the achieved benefit. Diffusive representation is used to obtain thermal dynamical models.•Two spherical wind sensor prototypes are compared experimentally to corroborate this effect. The objective of this paper is to show how the structure of a thermal wind anemometer can be tuned to ensure a fast response when the sensor works in closed loop configuration (constant temperature operation). If the thermal filter associated to the sensor structure has only one significative time constant, the resulting system time response, working in closed loop, is enhanced beyond the natural limit imposed by its own thermal circuit. This effect is theoretically explained using the theory of sliding mode controllers. Experimental corroboration is presented by comparing the results obtained with two prototypes of a spherical wind sensor for Mars atmosphere. It will be shown that in case of having only one significant time constant, the time response in closed loop is much faster than the value associated with that time constant. It will be experimentally shown that this effect is lost when the thermal filter has more than one significant time constant. Diffusive representation is used to obtain arbitrary-order models of the thermal structures of the sensors.