Estimação da Velocidade do Vento utilizando Transdutores Ultrassônicos baseada em Rede Neural Artificial

Abstract

The measurement of wind speed is of interest in several applications, such as wind power generation, mainly for the optimal determination of locations for the installation of wind farms. In the current scenario, the ways of measuring this quantity are diverse, however, when the design restrictions point to a measurement with sufficient accuracy, instruments based on ultrasonic sensors stand out, as they present a low uncertainty compared to conventional mechanical instruments. In this sense, its measurement principle requires the estimation of the Time of Flight (ToF), which is the travel time of the ultrasonic signal from the transmitting transducer to its detection in the receiving transducer. Therefore, the conventional techniques for estimating the ToF are those based on time intervals and digital signal processing, but there are, on the other hand, those based on artificial intelligence. In this article, an artificial intelligence model based on Artificial Neural Networks capable of estimating the wind speed is presented. Furthermore, the construction process is described, as well as its experimental results commented.