Back to the programme printer.gif Print

Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'Advanced operation & maintenance' taking place on Thursday, 13 March 2014 at 11:15-12:45. The meet-the-authors will take place in the poster area.

Rohan Soman Institute of Fluid Flow and Machinery, Polish Academy of Science, Poland
Rohan Soman (1) F P Pawel Malinowski (1) Wieslaw Ostachowicz (1)
(1) Institute of Fluid Flow and Machinery, Polish Academy of Science, Gdansk, Poland

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Rohan Soman is a research fellow working in the area of smart structures for damage detection in metallic and composite structures in Wind Industry.
His previous experience has been in the area of SHM for long span bridges and also includes research in multi-disciplinary areas like use of Wireless Sensor Networks, Energy Harvesting.
His other research interests include SHM, Vibration based Damage detection, Modal Analysis, FBG strain sensors, Data Fusion and Optimization of Sensor Placement.


Neutral axis tracking for damage detection in wind-turbine towers


Wind Energy is seen as one of the most promising solutions to man’s ever increasing demands of a clean source of energy. The use of wind energy has received an impetus due to the advancements in the field of materials engineering. Newer, bigger wind turbines are now possible which are more robust, and lighter in weight. The main drawback of the wind energy is the high initial costs for setting up and maintenance. These high initial cost make the energy more expensive than the conventional energy sources like fossil fuels and nuclear and hence has not been widely accepted.


The wind turbine is a complex system consisting of many components, like the support structure, tower, nacelle, blades, gearbox, generator etc. All these systems are in co-ordination in order to ensure proper functioning of the wind turbine. Some of these components are more prone to failure, while others are more robust. These components also have different costs and time taken for the maintenance actions or the replacements of the components. If the rate of failure and the downtime or the cost of replacement are taken into consideration, the generator, gear box and the blades are most important components.
Thus the present research focusses on the damage detection of the composite blades.The traditional methods for health monitoring and damage detection was visual inspection, but in the case of these massive off-shore wind turbines it is not always possible. In addition, it is highly subjective and in-accurate for detection of internal cracks which are often not accessible. In order to overcome these shortcomings many Non-Destructive Evaluation (NDE) techniques have been proposed. Like the use of IR Thermography, Lamb Waves, Performance monitoring, Acoustic Emission etc. These methods have shown promise in simulated studies and experimental validations, but are not always applicable for in-service monitoring. The need for closures, leads to monetary loss, which is not always welcome by the Wind Farm owners. Thus a new technique which is able to perform in-service monitoring is required. The vibration based damage detection techniques are a possible solution to this problem and are looked at in this research.

Main body of abstract

The vibration-based damage identification (VBDI) has received great attention in SHM of large structures bridges, towers etc. The use of global level sensors for VBDI allows low cost technique for SHM of structures, but is insensitive to lower levels of damage and if the damage occurs away from the sensors. Furthermore it is more challenging to use the methods, where the input excitation cannot be measured. In order to improve the sensitivity to damage, a use of local level sensors is on the increase. The recent advancements in Fibre Optic Sensors and the excellent characteristics shown by these sensors have generated a lot of interest in the use of dynamic strain measurements for SHM of structure. Unfortunately, strain sensors are local level sensors and are highly sensitive to the ambient loading on the structure and the ambient condition changes which may lead to false detections. Thus, there is a need for a robust damage indicator, which is sensitive to lower levels of damage, while being robust to overcome the effects due to ambient condition changes. Furthermore, the method should also be able to detect the damage in the output only mode (excitation is not measured).
In this paper, a robust metric for damage detection of towers is proposed. The metric chosen for damage detection is the position of Neutral Axis (NA) of the tower structure. The NA is the property of the cross section of the tower independent of the bulk temperature effects, and the ambient wind loading. The position of the neutral axis can be assessed by measuring the strains on opposite surfaces of the tower in bending. The estimation of the NA of the tower subject to unknown loading both in magnitude and direction is presented here. The discrete kalman filter is employed for the estimation of the NA subject to wind loads in the presence of measurement noise from the sensors. The study is undertaken on a validated FE model of the 5MW wind turbine tower. The robustness of NA tracking as a damage indicator has been studied for different damage locations, damage extents, and different loading conditions.


The most critical part of the damage detection methodology is the selection of the threshold change in the NA location before an alarm is raised. The selection process is essentially a trade-off between false positive detections and false negative detections of damage. The false-negative damage detection reduces the availability as the system has to be shut-down as precaution till inspection has been carried out, on the other hand, a false positive detection may lead to oversight of some damage which can lead to failure of the entire system causing great economic losses and even be a threat to human life. The initial selection of this threshold should be carried out based on engineering judgement and then can be updated through visual inspections.
In nutshell, the study undertaken indicates that the NA tracking through the use of discrete kalman filter is a robust damage detection strategy. The use of local level sensors improves the sensitivity of the system to damage, and hence allows lower levels of damage to be detected. In addition, through the use of extended kalman filter, the local effects due to loading and measurement noise are easily overcome. The effect of the ambient temperature changes can be overcome through the use of temperature compensated FBG sensors. Furthermore, the choice of the damage indicator makes the system independent of the loading conditions, and as such the wind loads need not be measured. The selection of the appropriate damage indicator has thus allowed a better damage detection as compared to the VBDI approaches which have been commonly used in structural engineering

Learning objectives
New Damage Indicator for wind turbine tower, subjected to unknown loads in direction and magnitude.
Use of extended kalman filter for improving the robustness of the damage indicator in the presence of measurement noise

[1] Doebling, S.W., Farrar, C.R., Prime, M.B., 1998, A Summary Review of Vibration based Damage Identification Techniques, Shock and Vibration Digest, 30(2) 91-105
[2] Ciang, C. C., Lee, J. R., & Bang, H. J. (2008). Structural health monitoring for a wind turbine system: a review of damage detection methods. Measurement Science and Technology, 19(12), 122001