The main goal of the WindFarmSHM research project is the development, validation and optimization of a monitoring strategy to be applied at the level of the wind farm, which should be able to evaluate the structural condition of a set of wind turbines and their consumed fatigue life. The accomplishment of this goal will imply the development of an experimental campaign, the construction of numerical models and the implementation of new data processing methodologies.

The experimental campaign involves the simultaneous instrumentation of several wind turbines in the same wind farm. This is being performed in an onshore wind farm located in the central region of Portugal, in a flat terrain close to shore. It consists of 5 VESTAS V100-1.8 MW wind turbines, with a 100m diameter 3 blades rotor, supported by a 93.3m steel tower, clamped in a reinforced concrete slab supported by 16 concrete piles.

The research project includes the monitoring of 3 wind turbines, loaded by winds with different turbulence intensities motivated by the neighbouring wind turbines, during a period of about 2 years. In one of the wind turbine the following equipment has already been installed:

• 2 alternative systems to characterize the dynamic behaviour of the tower: a commercial system and a customized low-cost system based on MEM accelerometers designed and assembled in FEUP;
• strain gages at two cross sections of the tower base to characterize the stresses during different operating conditions;
• a set of fiber optic strain gages at the blades roots to estimate bending moments;
• MEM accelerometers placed at the blades (10 meters from the root) to characterize their dynamics;

Complementary information about the wind characteristics and the wind turbine operation are obtained from the SCADA system.

This work describes alternative approaches for experimental estimation of the bending moments applied in the tower of the already fully instrumented wind turbine

The monitoring system installed on the tower is composed by 6 2D rosette strain gages, 4 temperature sensors connected to digitizers and a field computer then linked with a modem for remote access to the data. Having in mind the evaluation of the static bending moments diagrams along the tower, the six strain gages are distributed in two tower sections: one with four sensors at 6.6m from the tower base and another section with 2 sensors at 7.8m from the tower base. The temperature sensors and the measurement of the strain in the direction perpendicular to the tower axis are important to evaluate alternative procedures to minimize the temperature influence on the measured longitudinal strains.

The adopted solution for blade strains monitoring is a commercial system. Each blade is instrumented with 4 fiber optic strain sensors and temperature sensors for compensation of the temperature effects. Each set of sensors is connected to a central acquisition system installed in the hub, which is then linked with a modem for remote access to the data.

Special attention is given to tuning of the strain signals processing, which includes temperature effects compensation and signal calibration according to IEC 61400-13.

The main goal of the blades strain monitoring is fatigue assessment of the blades and condition monitoring using automatically tracked modal parameters evolution. But from the bending moments estimated at the blades roots it is also possible to estimate the total wind load at the rotor and subsequently the FA static bending moments at the tower base. These results will be compared with the results obtained with the tower strain gages.

Advanced models of wind turbine are currently being developed in FAST. The experimental results reported in this contribution are being used for calibration of these numerical models.