Reliability rocks: the utility purchaser has the hard but interesting task of assessing the reliability of new technological concepts with no de facto track record. But how do utilities evaluate a design? OceanWise put the question to two of the most experienced utilities in the world of green energy: Vattenfall and Dong Energy.
By Anne Korsgaard
Villads Haar Jakobsen (VHJ) works as Project Engineer at Vattenfall on the East Anglia project, where he is responsible for the mechanical area, and Torben Kenneth Hansen (TKH) is Head of the Department for Renewables Turbine Leads at Dong Energy. They both have a job full of decisions at the heart of offshore renewables.
"Of course, as a buyer we will go for the product that has been tested as thoroughly as possible", explains Villads Haar Jakobsen, who does not hide his aspirations to push forward for more and better testing and higher degrees of openness when it comes to test methods and test results.
The good news is that more and more information is being released as competition grows. And information on durability, for example, will become an important competitive factor as will a manufacturer’s ability to document product reliability through robustness and reliability testing.
"Improved tools for risk assessment and more thorough testing are some of the tendencies that I observe in the sector", relates Torben Kenneth Hansen, who has a say in the project execution of all offshore wind turbine contracts at Dong Energy.
The utilities try to avoid unplanned, expensive O&M visits to the offshore wind turbines by assessing the reliability of the products. Photo: Heidi Lundsgaard.
What is the greatest challenge in choosing wind turbines for future projects?
VHJ: Our main concern and criteria when choosing is reliability. Therefore, one of the biggest challenges right now as a buyer is to assess the reliability and evaluate the new technologies that have not yet got a track record and years of production.
TKH: I see a challenge in that we are looking at more advanced and integrated technologies and still larger wind turbines on the market. Despite the risks, we throw ourselves into these new technologies because the overall project economy improves with size due to increased output and fewer positions in relation to swept area, especially offshore.
What can convince you of the reliability of a new technology?
VHJ: Increased openness concerning the information about test methods and results can make me go part of the way. Of course, the more data we can see on how the test design has been made and how the components, groups of components and nacelles actually performed in the test, the better we feel informed and the more likely we are to choose the product. Since we have NDAs with all the companies involved, this is possible – at least with some of the manufacturers. It will also be an advantage if a wind turbine has performed well in testing at an independent test facility like LORC, for instance.
TKH: In the future, an important criteria before buying a wind turbine will be how well the manufacturer does the job of convincing us about value creation of future wind power plants from a product’s test results. It will definitely be an advantage if a manufacturer encloses test models and results for a new design concept, where we can see clear evidence that the risks have been truly mitigated. It is central to us that the wind turbine has a certain degree of proven technology when we consider a brand.
What are the strongest tendencies that you observe in the offshore wind sector?
VHJ: I see the tendency of the market becoming a buyer’s market. This gives us, the utilities, an increased influence on, amongst other things, the information that the manufacturers are willing to share. I am pleased to observe that the quality of the documentation that we receive about the validation and test results for new designs is also improving.
TKH: The wind sector is turning into a professional sector that is run like an industry with all that this involves, including precise specifications and demands on products and contracts. Both the quality and the quantity of tests performed on new components and designs are increasing and the tools available for risk assessment of a product are also improving, says Torben Kenneth Hansen.
What are the main criteria when you qualify a wind turbine?
VHJ: When we make an assessment of a wind turbine we look at the technical data provided by the manufacturer and how the life has been verified in the field and by accelerated life testing. Once again, reliability is the key word. So anything that can back up and substantiate the reliability from day one is our focus.
TKH: We set up a number of criteria in the tender material concerning technical aspects i.e. whether the design uses proven technology. Also business factors, such as experience in the sector, construction execution capabilities and manufacturing security and quality, are considered. The most important subject in the technological field is reliable performance on site from day one, a high up-time so that we can rest assured that the turbine produces while the wind is present, which then again results in competitive overall wind power plant economy, and convincing performance forecasts.
Accelerated tests: The bathtub model
In the beginning, when a new concept or component is being tested, a large number of flaws will occur and be corrected. Then follows a long period of relative calm and quiet with a constant failure rate, only to see flaws and breakdowns occur when the component reaches a life of 20 years, which is the calculated lifetime for an offshore wind turbine. The wear out failures are acceptable inasmuch as the wind turbine design is made for a 20 year lifetime.
Two main types: robustness testing (e.g. HALT) and reliability testing (e.g. CALT)
Testing robustness is about finding the weak spots in a component. This means increasing the relevant load parameter above the limit of load that it was designed for until the component fails. It is then assessed whether the fault is relevant and if it is relevant follows a redesign of the component and a further HALT test again. If you work this way, you will get a lesser rate of faults in the first phase of the bathtub curve: also called infant mortality.
Reliability testing is all about accelerating realistic load parameters in a way that allows to demonstrate 20 years of life time in a short period of time - down to three months. That way you reach the wear out phase of the component in the bathtub model.