Durability and Fatigue Challenges in Wind,
Wave and Tidal Energy
Engineering Integrity Society
Wednesday, 29 November 2006
BAWA, Filton, Bristol
The following abstracts are available
and further ones will be added shortly:
Thermoelastic stress analysis and acoustic emission monitoring during full scale tests of wind turbine rotor blades
Dr Geoff Dutton - Energy Research Unit, RAL
The application of thermoelastic stress measurement techniques and acoustic emission monitoring during a wind turbine blade fatigue test is described. The most highly loaded regions of the blade are identified and a novel adaptation of the thermoelastic technique made to allow the identification of all known areas of developing damage in the blade. The three resulting techniques are compared and contrasted.
Thermoelastic stress analysis (TSA) allows the measurement of the surface stress distribution on a blade during cyclic loading. TSA can indicate stress concentrations and developing sub-surface damage long before any visible surface indications develop. The technique has the advantage of requiring only low load magnitudes and can be used to validate finite element model stress distributions during the early stages of a blade test or to characterise the spread of damage during failure. The author has been involved in developing a novel variation of the technique that enhanced the damage detection capability, resulting in the identification of all known damage within a test blade.
Acoustic emission (AE) monitoring can be used to proof-test a blade before and after static or fatigue loading. Since AE detects the vibrations due to damage propagation it has the potential to be developed into a condition monitoring technique for operational turbines. Results from the same test blade described above are presented.
Remote torque measurements and gearbox condition monitoring for wind turbines
Jarek Rosinski, Martin Rosinski, Andrew Wechsler – Transmission Dynamics
Unpredictable dynamic torque excitation in gustily winds often results in significant overloads of geared transmission systems in wind generators.
Design of future transmission systems for wind turbines critically depends on detailed knowledge of in-service loading. Ideally, in-service load measurements should be carried out over a prolonged time to capture reliable long-term information over the four seasons and under adverse weather conditions.
Transmission Dynamics designed bespoke systems for in-service load measurements and gearbox condition monitoring for wind turbine applications. These systems make use of remote data transfer via the GPRS network making significant improvement in the data collection process. Unique condition monitoring system developed for wind turbine applications make use of combined torque and gearbox vibration measurements to improve the reliability of diagnostics procedures.
Techniques & pitfalls in the analysis of fatigue histories in wind turbine hub assemblies
Roger Haines - Garrad Hassan
A typical (1 MW or more) wind turbine rotor hub assembly consists of a hub casting connected by large diameter bearings to three composite blades and by a flange to a shaft. All four connections are made through rings of bolts or studs. The bearings are usually of the four-point or eight-point contact type. The hub may rotate more than 100 million times in a lifetime. An examination of this simple description of the problem class reveals a number of features for which there is no universally agreed method of analysis. This contribution explores the characteristics of the problem class and highlights some of the pitfalls.
The simulated loads used for analysis typically comprise of the order of more than 100,000 seconds of each of 18 independent components of loads. Simply combining these loads in an automated fashion using unit stresses to obtain stress time histories may not be valid, because of significant non-linearities introduced by the structure, particularly the bearing races where the contact angles of the balls change sign as the overturning moment passses through zero.
The severity of the non-linearities and the implications for the appropriate methodology depend to as great extent on the type of structure used for the hub. “Star”, “ball” and “delta” shapes and various types of stiffening rib each have their own effect.
In cases where linear methods are not valid, auxiliary load terms can be introduced which bear a more linear relationship to certain stress terms.
The bolted flanges introduce non-linearities at relatively high operational loads. Methods for handling local loss of contact area are well established, but the more difficult phenomenon of local slipping between a stiffened and unstiffened structure can be just as important.
The use of strain gauges to validate the analyses is advantageous. Strain gauges around the hub rims to determine ovalisation behaviour and multiple strain gauges to measure dynamic axial and bending strains in the fasteners can be both practicable and informative.
Compression-compression fatigue behaviour of composite blade
Daniel Ng - Hexcel
Hexcel and in particular the site at Linz has been supplying prepreg since the early 1980’s and is a dedicated manufacturing plant designed for supply to the industrial markets.
During the last several years we started conducting our own tests for fatigue associated with qualifying materials to the wind industry. Through this period we have evaluated several different jigs and encountered many problems associated with compression – compression fatigue.
The main aim of the seminar will be to pass on some of the problems associated with this test and to discuss the few jigs out in the market place.
Some relations to the standard S-N curve will be made as a conclusion to the discussion.
Cost Reduction and Life Extension of Offshore Wind Farms (CORLEX)
C Lindley - Corus RD&T, Swinden Technology Centre
CORLEX is a 2-year Dti-funded project – funded under the ‘Technology programme – Renewable Energy’. The project is looking into various aspects of offshore wind farm design, operation and maintenance and is coordinated by Corus RD&T. Project partners include Atkins Process, TWI Ltd, Camcal Ltd, RWE Npower and Sheffield Forgemasters Engineering Ltd (SFEL), each bringing their own area of expertise to the project.
A 30% reduction in costs is necessary for offshore wind to be economic. One area where savings could be made is to reduce the costs of the support structure (tower and foundation) by:
(i) Cheaper/faster fabrication of towers & foundations (utilising novel welding techniques);
(ii) Extended structural lives through improved designs and fabrication methods;
(iii) Light-weighting of turbines and supporting structures.
The objective of CORLEX is therefore to help reduce the capital and operating costs of offshore wind farms. The project has thus been designed to address a number of issues via 5 main work packages:
WP1 - Review of current designs
The review of existing tower and foundation designs covers various wind tower designs, fabrication approaches, options for welding/casting and installation methods. Added to this - cost, weight & manufacturing issues are also being addressed. Investigation of SFEL’s novel cast node design proposal (WP5) is being undertaken – yielding additional data, which will help SFEL produce a prototype. The review is also focussing on the steel property requirements, as it is possible that higher strength steels may be required.
WP2 - Welding technologies
Traditional welding methods can be slow and expensive, while also limiting the design
lives of such fatigue-critical structures. The results from the Corlex test programme will help determine whether innovative new welding technologies have the potential to reduce capital costs and extend design lives, but more development is needed to assess their viability for the Offshore Wind industry. The techniques which have been highlighted for investigation include multi-wire submerged arc and reduced pressure electron beam welding.
WP3 - Structural health monitoring (SHM)
CORLEX aims to identify the benefits to be gained from the use of SHM in the offshore wind energy industry and if initial results prove positive, will help to provide a specification for a suitable system. Such SHM data systems may also be used to justify reduced levels of conservatism in design, which may provide opportunities for cost reductions in fabrication processes and material specifications.
WP4 - Risk based life management (RBLM)
It is hoped that CORLEX will identify the potential for gains to be achieved from RBLM (in conjunction with SHM) in offshore wind farms. The EU FP5 project 'RIMAP' has provided a generic framework for the development of risk-based inspection and maintenance methodologies across a wide range of industries in the European regulatory context. The RBLM/SHM output from CORLEX could be used as the basis for the development of a RIMAP-based offshore wind energy industry 'workbook' for RBLM.
WP5 - Casting technology and design optimisation
There is a need to develop the engineering design of cast nodes, which could be used on these steel sub-structures and more particularly, to define the costs (via the production of a full-scale trial node) and to use this as a vehicle to introduce new cost saving ideas. The economics are a vital issue as presently wind turbine installations are struggling severely to meet the required economic criteria for wide scale development. A number of novel designs have been assessed by SFEL – much of their effort is currently being concentrated on the production of the prototype cast transition node.
It is hoped that the potential economic benefits are significant: the results from CORLEX helping to save the industry a substantial amount of money and increasing the competitiveness of the UK offshore wind energy sector.
Fatigue analysis approach of Pelamis main structural elements
Charles Taylor - Ocean Power Delivery
Presented is an overview of the development of Pelamis Wave Energy Converter (WEC) to date, highlighting Ocean Power Delivery's (OPD) evolving approach to the design challenges of offshore wave energy.
The key purpose of the prototype structure was to provide a robust platform on which to demonstrate and develop the systems behind Pelamis. Third party verification of structural design and durability, based upon accepted oil and gas standards, played a crucial role in securing insurance and building credibility for the technology and the industry.
Unlike typical moored floating structures, Pelamis is designed to resonate and absorb energy. OPDs in-house Pelamis simulation software (PEL) has formed the basis for our performance and fatigue analysis. The machines response is modeled using nonlinear hydrodynamics to generate a time series of load at an arbitrary point. A rain flow analysis of this data is used to calculate fatigue damage based on detailed finite element modeling accepted S-N curves.
As application specific data is collected, confidence in the durability of the design will allow development of new standards appropriate to the new industry. The first stage of this is underway and OPD have been instrumental in the development of the DNV's “Guideline on Design and Operation of Wave Energy Convertors”.
Marine Current Turbines: towards commercial reality
Peter Fraenkel - Marine Current Turbines
The presentation will include a review of previous demonstrator technology
leading to current prototype Seaflow which has been working for 3 years and
the latest production design: Seagen. Seagen's design is complete and
production of the first one is nearing completion - due only 1-2 months
after the seminar. It will be the world's largest installed and working
marine current turbine.
Fatigue loading of offshore wind turbines - the importance of integrated analysis
Tim Camp - Garrad Hassan
Offshore wind turbines, like their onshore counterparts, are flexible structures subject to various sources of excitation. The internal loads in the blades, drive train and support structure are critcally influenced by the dynamic response of the structure as well as by the applied aerodynamic and hydrodynamic loading. With design lifetimes of typically 20 years, wind turbines can fairly be described as "fatigue machines"!
For offshore wind turbines in particular, the dynamic behaviour of the support structure is very influential in determining the design fatigue loads. A particular example of this is the aerodynamic damping of wave-induced tower motion provided by the wind turbine rotor. This effect is significant in mitigating fatigue loads in the support structure and can only be captured during the design process using an integrated numerical model of the wind loading, wave loading and structural dynamics.
The presentation will show the latest design tool for space-frame support structures for offshore wind turbines and will illustrate the importance of integrating wind and wave loads in a single numerical model in order to calculate accurately structural fatigue loads.
Wave Loads on Offshore Wind Turbines - Comparison of Theory with Measurements at Blyth
Andrew Henderson - Offshore Wind Energy
The vast majority of offshore windturbines have been built in shallow waters, significantly different to where the offshore petroleum industry builds the majority of its structures and hence the environment where design codes and guidelines have been developed for. Although common sense might suggest that methods developed with deep waters in mind will be more than sufficient in shallower waters, this is not necessarily the case, since waves become less “linear” as they approach the shore.
Hence it is necessary to confirm that the procedures used are sufficiently accurate. Focusing on the wave loads and making use of the data generated by the Blyth offshore windfarm measurement programme, wave loads can be estimated from the measured surface profile and compared with the strain measurements, in terms of individual waves, sea states and lifetime.
This presentation will show that care must be taken in choosing how the wave loads are modelled when analysing the support structure, however in the shallow waters where non-linear effects are most severe, the relative contribution of the waves to the overall fatigue damage is lower.
The Certification views on the structural survivability / reliability
for wave and tidal energy converters
Claudio Bittencourt – DNV
Abstract (draft):
Defining the required level on structural survivability / reliability for wave and tidal energy converters is a challenge in itself. The objective is to obtain the right balance between safety, asset integrity, uncertainties on loading and influence of control system, maintenance regime, installation and decommissioning against cost / investments and revenue provided from power generation. The traditional certification process normally focus on safety issues, that is well short of the needs from the renewable energy sector.
The DNV OSS-312 Certification of Tidal and Wave Energy Converters establishes a certification process that extends beyond the safety aspects by reflecting the functional requirements of the marine energy converter. Thus, it is natural that the targets for structural survivability / reliability takes into account the same functional requirements. The OSS-312 applies the Qualification of New Technology principles, the technical requirements from the Guidelines for planning and operation of wave energy converters and standards defined in the Guidelines.
The Battelle structural stress method for fatigue analysis of welded joints
Pingsha Dong - Battelle
John Draper - Safe Technology Limited
Welded joints are characterised by a sharp stress concentration at the weld toe. It is not possible to calculate actual stresses at this stress concentration. For this reason the fatigue strength of welded joints is often defined using a nominal stress some distance from the weld toe. In terms of nominal stress, different welds have different fatigue strengths. The engineer has therefore to make two decisions - how far from the weld toe should the nominal stress be defined, and what S-N curve applies to a particular weld. When using results from finite element models, the calculated stress can also be mesh-sensitive.
The Battelle structural stress method was developed to remove these subjective decisions. The method uses a stress calculated at the weld toe, which is then corrected for the effects of plate thickness and type of loading. It is shown that with this parameter, termed the equivalent structural stress, all welds can be defined using a single S-N curve (one curve for welds in steels, one curve for welds in aluminium alloys). The method has been correlated against more than 3500 fatigue test results, using butt and fillet welds in a wide variety of plate thicknesses, and also using spot welded joints in shear and in peel.
Extensive work with finite element models has shown that the method is highly mesh-insensitive, allowing accurate fatigue life estimates from coarse mesh models.
The method is patented, with the Verity® trade mark, and is being adopted by international design codes. It has been licenced to Safe Technology Limited and is included in fe-safeTM.
Home: http://www.e-i-s.org.uk