Smart embedded sensor systems for offshore wind turbines

This is made possible by incorporating sensors into the wind turbine blades to discover changes and damage in the blade structure at an early stage before the blade breaks. Remote monitoring of damage to off-shore wind turbines could save money, because the intervals between inspections can be increased. Moreover, operation becomes safer and more stable as it possible to check from a remote place whether a blade has been damaged after a heavy storm. The project is called SESS and stands for Smart Embedded Sensor Systems.

Many unnecessary inspections can be avoided

The new sensor system is to monitor "the health" of each wind turbine, sending the information to the operator on shore. If the operators think it is necessary, they can send technicians to carry out repairs before a stoppage occurs. This will lead to significant savings in operating costs through the whole life of the turbine. It will also be easier to decide what to do with a wind turbine once it gets "old." Can it continue to work, does it pay to rebuild it, or should it be taken down and sent to recycling?

Risø DTU is helping develop the sensor system. Models describing the reaction of composite materials to various load conditions are tested in the laboratory, and classify the various defects and damages that may occur. Efforts are also made to model the reaction of the blade to the damage condition, and the loads during operation.

Interfering before minor damages become fatal

The idea is to develop a model which follows the principle of the old saying "Many of one thing makes another larger thing": Many small changes in the blade material should be monitored one at a time in order to assess the overall impact. In this way one can predict when a particular blade should be repaired.

This tool makes it possible to design new types of blades, and to predict how damage will develop. The sensors also give the opportunity to develop prognostics calculating the remaining life of the turbine, based on its current condition. Operators of large wind farms could use the tool to decide how best to maintain the blades during operation.

The models are developed and verified through mechanical tests of materials, structural sub-components and blades at full scale.

Sensors are used in all tests to provide detailed knowledge of the progress of the test. Typical sensors can include fibre optics, piezoelectric materials, strain gauges, etc. They keep an eye on the different types of damage that may occur to the blade.

Many functions within one sensor system

"We want to develop an integrated sensor system which can carry out many tasks simultaneously. It will be needed in a future remote monitoring system for wind turbine blades. No single type of sensor will be able to supply all the necessary information to estimate the condition of the turbine blade because there are many potential sources of errors and damage in blade materials," says Malcolm McGugan and Kaj Borum from the Materials Research Division at Risø DTU. They emphasize that the sensor system has to be very reliable; it must not cause wrong alarms, leading the operator to send technicians to a wind turbine without reason.

"The advantage for us is that we can test the blades throughout the entire scale from the microscopic damage of material to tests on the blade at full scale at Risø's Experimental Research Facility for Blade Structure," the two scientists say, continuing: "The intention is to make the new sensors a part of the data flow which is already transmitted from a wind turbine, in accordance with the wind industry IEC standards. Adapting the integrated sensors for these standards, we have created a fine basis for monitoring key components of the wind turbine, including the turbine blades."

To be used on buildings, bridges, aircraft and ships

The new "health" sensors will probably be introduced gradually. As more experience is gained with the sensors in connection with wind turbines, it will be possible to use the sensors in other fields too, such as the aviation industry. Other obvious uses are: bridges, ships and buildings. We believe that such systems could predict, for instance, whether the roof of a sports hall after heavy snow is in danger of collapsing due to the weight of snow.

In September, Risø DTU held a course with focus on this broad application. The course title was "Structural Health Monitoring -- SHM," and this type of monitoring is expected to have a great future when it comes to monitoring and protecting many types of constructions.