Abstract

An inventory application of a LoRa-based wireless sensor network on the wind turbines is described to replace anemometers with the low-cost and precise networked wind speed estimators. It operates without mechanical issues associated with the cup anemometers and provides better data even in the presence of turbulence, tower shadow, and nacelle body. Each node in this scheme compresses it's local real wind speed and electrical mea- surements and transmits corresponding packets to a center. Due to the limited memory and computational power available on the sensor side, the modeling task carries out pri- marily by the central node to eliminate the need for the anemometer. The nodes use the resultant linear multivariable model to estimate the local wind speed instead of using the anemometer. The methodology can be applied to the AC or DC turbines with an excellent performance where the LoRa technology minimizes power consumption while maximizing the license-free communication range.