Wind-plant reliability has become increasingly important, as installations have reached 4% of U.S. generating capacity. Unplanned maintenance and component failures are a concern to both wind-plant owners as they directly affect revenue streams, as well as wind-turbine manufacturers who have to cover the costs of warrantied maintenance work.

Sandia leads efforts in wind-turbine reliability research—focused on reliability data gathering as well as specific research into blade reliability. The Continuous Reliability of Enhancement of Wind (CREW) project collects data from thousands of wind turbines nationwide to understand the primary drivers of wind-plant component failures. The Blade Reliability Collaborative (BRC) brings together industry, academia, and labs to improve wind-turbine blade manufacturing quality and determine the most cost-effective methods to mitigate environmental damage. Through this work, Sandia is ensuring that wind-energy technology will deliver economical, reliable, clean energy to the nation.

2013 CREW Benchmark: These wind turbines actively generate power 83% of the time and are available to generate 97.6% of the time.

2013 CREW Benchmark: These wind turbines actively generate power 83% of the time and are available to generate 97.6% of the time.

To characterize wind turbine and wind plant reliability performance issues and identify opportunities for improving reliability and availability performance within the wind industry, the Continuous Reliability Enhancements for Wind (CREW) Database and Analysis Program has been set up as Sandia. CREW information, based on analysis of industry partners’ proprietary data, is used by DOE to guide their understanding of wind industry performance and reliability, and to guide their R&D investment.

The wind industry can also use CREW’s public benchmark to self-assess their performance against their peers. Additionally, the information can guide industry action to improve equipment performance, improve operating practices, and provide common expectations across the industry. Additionally, research engineers can use the data to further understand the field conditions the industry faces.

Beyond providing industry with benchmarking information, CREW goals include further motivating the cultural shift to more effective monitoring, collection, analysis, and use of data as part of the CREW team’s “Full Data Picture” message.

The Sandia project team also partners with industry experts to provide the public with other important research and knowledge. Recent examples include “CMMS in the Wind Industry” by Management Resources Group (MRG), “Wind Industry Work Order Information Flow Survey” by Muir Data Systems, and “Wind Industry Segments & CMMS Value Propositions” also by Muir Data Systems.

The CREW Program also uses Sandia’s new and existing infrastructure for expanded analysis that characterizes reliability performance issues and identifies opportunities for improving reliability and availability performance within the wind industry. Examples of this work include further developing the industry’s understanding of turbine-to-turbine interaction and plant underperformance.

Reliability, Operations & Maintenance, and Standards

Wind-turbine blades are among the largest composite structures made in the world. The economics of energy production dictates that low-cost manufacturing methods be employed to produce cost-effective machines. This, combined with an often harsh operating environment, creates a challenge for blade designers and manufacturers, as well as wind-farm operators. In many cases, failures in the field can be traced back to the manufacturing floor.

Wind-turbine blades are among the largest composite structures made in the world. The economics of energy production dictates that low-cost manufacturing methods be employed to produce cost-effective machines. This, combined with an often harsh operating environment, creates a challenge for blade designers and manufacturers, as well as wind-farm operators. In many cases, failures in the field can be traced back to the manufacturing floor.

Blade reliability is one of the most costly elements of plant installation and operation because blade failure can cause extensive down time and lead to expensive repairs. The manufacturing process can be a significant source of defects due to the complicacy of producing very large composite structures with inexpensive materials and processes. Blades can also be damaged during the transportation and installation process. Finally, wind turbines operate in harsh environments where wind-blown particulates cause surface erosion and lightning stirkes can sometimes destroy blades. The summary effect of these issues is that blades have been responsible for 11% of turbine downtime in recent years, which is the third largest source of any component. Furthermore, the amount of downtime attributed to blade failure has increased over the past few years, and it is anticipated that as rotors continue to grow to larger sizes, blade reliability issues could become more prominent.

The Sandia-led Blade Reliability Collaborative (BRC), funded by the U.S. Department of Energy, seeks to understand the root causes of premature blade failure and unplanned maintenance, and the most cost-effective methods to insure that blades can survive their expected operational life.

The current research work under the BRC focuses on examining flaws and defects resulting from the manufacturing process, as well as damage incurred during operation. Major area of focus include:

  • the evaluation of nondestructive inspection (NDI) techniques and equipment in both manufacturing floor and field settings to assess their viability in terms of both accuracy and practicality.
  • understanding the effects of leading edge erosion by the collection of field measurements of erosion profiles and performance degradation, the development of aerodynamic models through wind tunnel testing, and computational fluid dynamics (CFD) modeling.
  • development of high-performance computing (HPC) models to analyze the loads experienced during operation and the effects that these loads have on the initiation and growth of damage in flawed blades.
Operational damage is also a cause of premature blade failure, whether it is from erosion of blade surfaces as shown on the left or lightning strikes as shown on the right.

Operational damage is also a cause of premature blade failure, whether it is from erosion of blade surfaces as shown on the left or lightning strikes as shown on the right.

The Sandia project team has recently

  • designed and manufactured two small-scale validation blades containing highly characterized flaws;
  • made detailed surface scans of utility-scale wind blades with eroded leading edges;
  • conducted detailed investigation of manufacturer process to determine width of adhesive bond line at shear web to spar cap flange interface;
  • tested a new, in-house designed pierced-bladder, phased-array probe holder that can be used with wide-area scanning techniques;
  • performed wind tunnel testing of airfoils with leading-edge erosion; and
  • developed new analysis methods to model wind blade leading-edge erosion.
The connection between flaws like wavy fibers (top) and porosity (middle) among others, and the diagnostic tools that can be used to find them both on the factory floor and in the field is a major focus of the BRC research work.

The connection between flaws like wavy fibers (top) and porosity (middle) among others, and the diagnostic tools that can be used to find them both on the factory floor and in the field is a major focus of the BRC research work.

Our future project goals are to

  • improve operational wind plant reliability by collecting data on blade defects and damage to guide research,
  • advance blade standards by quantify uncertainty in analysis tools,
  • inform reliability and performance testing through blade testing and model validation, and
  • increase operational wind plant reliability by addressing operational damage caused by leading-edge erosion and lightning.
Manufacturing Supply Chain: Targeted Effects of Manufacturing Defects

Researchers start by measuring defects in real wind-turbine blades that have failed in the field, such as wavy fibers shown at the top. Then material test coupons can be manufactured (middle) and tested to validate computer models, which can then be used in the design process.

Researchers start by measuring defects in real wind-turbine blades that have failed in the field, such as wavy fibers shown at the top. Then material test coupons can be manufactured (middle) and tested to validate computer models, which can then be used in the design process.

Defects in wind–turbine blades are site, material, and blade specific and have been shown to have a wide range of effects depending on their location, material, type, and size. Furthermore, detecting defects in thick, multiple-material laminates has proven to be challenging using off-the-shelf inspection equipment developed for other industries.

If defects cannot be detected through inspection procedures, then they must be accounted for in the design of the blade (i.e., the blades must be “overdesigned,” which means more mass, more cost, and less efficient wind turbines). An increased understanding of the effects of blade defects along with better flaw–detection methods will enable higher confidence in manufactured products and lower design margins, leading to lighter, more reliable blades.

This project seeks to understand wind-turbine blade defects both in terms of

  • the confidence level with which they can be detected as well as
  • their ultimate effects if they are not found and remediated.
Drawing from experience inspecting composite aviation parts, Sandia has created a set of wind-blade-specific inspection specimens that mimic commonly seen blade flaws—allowing manufacturers to test and refine their equipment to better diagnose blade flaws.

Drawing from experience inspecting composite aviation parts, Sandia has created a set of wind-blade-specific inspection specimens that mimic commonly seen blade flaws—allowing manufacturers to test and refine their equipment to better diagnose blade flaws.

We will accomplish this by performing a probability of detection (POD) experiment using nondestructive inspection equipment and realistic, flawed blade specimens, along with mechanical testing and computational modeling of flawed blade coupons and substructures.

The POD experiment will consist of designing and producing a set of flawed wind–blade laminate and substructure specimens that will be blindly inspected by technicians using a variety of inspection equipment.

This project’s ultimate goal is to reduce uncertainty in as-manufactured wind–blade composites, enabling higher reliability and energy capture.

The Sandia project team has recently

  • designed and built multiple-scale mechanical test facility, enabling us to test wind blade composites at a large enough scale to examine relevant damage mechanisms;
  • developed new methods for probabilistic blade design with the inclusion of defects, along with performing significant characterization and analysis of flaws in wind blade composites; and
  • designed and manufactured initial NDI screening set to conduct first-ever comprehensive wind turbine NDI evaluation.
manufacturing3_SAND2014_0996P

Multiple-scale wind-blade material test frame (top) along with model of bending test specimen (bottom).

Our future project goals are to

  • improve blade standards by quantifying the probability of detection for common wind blade manufacturing flaws and operational damage,
  • develop next-generation blade inspection technology to increase probability of detection of wind blade flaws and damage, and
  • improve blade standards by characterizing the effects of defects in wind blade laminates.
Building on a foundation of analysis capabilities, methods, and industry partnerships established in the Continuous Reliability Enhancement for Wind (CREW) Database and the Blade Reliability Collaborative (BRC), as well as incorporating Sandia’s high-performance computing (HPC) resources and expertise, this work fosters the emergence and establishment of standards that influence wind turbine design, wind industry data collection and storage techniques, and analysis methods.

The IEA Task 33 Working Group on Reliability Data unites experts from around the world, with the common purpose of “standardizing data collection for wind turbine analysis.”

The IEA Task 33 Working Group on Reliability Data unites experts from around the world, with the common purpose of “standardizing data collection for wind turbine analysis.”

Sandia works to influence standards development with sound information and to promote higher standards that

  • improve wind turbine design and performance,
  • improve quality, and
  • reduce performance risks and costs.

With elements of reliability and performance engineering, this work can leverage the knowledge of DOE researchers and the results of DOE research turbines (including future offshore units) to influence the development of standards that increase the performance of all wind turbines and wind plants.

Sandia’s current and recent work in this area includes contributions to the industry through

  • the International Electrotechnical Commission (IEC) 61400-26-1, -2, and -3 Availability Standards,
  • the IEC-61400-1 Design Requirements Standard,
  • the International Energy Agency (IEA) Task 33 working group on Reliability Data, and
  • the American Wind Energy Association (AWEA) Operations and Maintenance Working Group’s best practices document for data collection.
One of the key challenges facing the industry is to develop reliable methods to detect damage in the rotor blades—and to detect them early enough to improve operations and repair/maintenance decisions leading to reduced costs and increased revenues. Sandia is addressing this challenge by performing research in the areas of structural health monitoring and prognostics management. The principal motivation of this research is to reduce operations and maintenance (O&M) costs, improve wind-plant reliability, and reduce downtime. A particular focus is to mitigate the large rise in costs for offshore O&M due to access difficulty, weather, high sea states, etc. using structural health monitoring and prognostics management.

With the overall goal of significantly reducing O&M costs, we are performing research to develop a cost-effective SHPM (Structural Health and Prognostics Management) system that can

ensure operations in a desired (designed) safe state of health,
aid in planning of maintenance processes versus more costly unplanned servicing,
avoid catastrophic failures through advance warning, and/or
improve energy capture by avoiding unnecessary shutdown and increasing overall plant availability.

Sandia has developed a technology roadmap for combining structural health monitoring and prognostics assets into a SHPM system with application to wind-turbine rotor blades. The roadmap describes six technology R&D thrust areas along with a maturation plan for each thrust area divided into five stages. Sandia has performed research in several areas specified within this roadmap.

A broadly applicable and cost-effective approach has been developed for simulating damaged conditions. The multiple-scale simulation of damage approach provides a means to understand (1) sensing for damage detection based on the operating response, (2) state of health or severity of damage based on local sensitivity loads analysis, and (3) the effect of damage mitigating control strategies on state of health and damage progress.

A broadly applicable and cost-effective approach has been developed for simulating damaged conditions. The multiple-scale simulation of damage approach provides a means to understand (1) sensing for damage detection based on the operating response, (2) state of health or severity of damage based on local sensitivity loads analysis, and (3) the effect of damage mitigating control strategies on state of health and damage progress.

At the core of this work is a simulation-based approach. Sandia has been simulating damaged blade conditions using a multiple-scale damage modeling and simulation methodology that was developed at the project’s onset. This simulation approach combines structural health monitoring and prognostic management to bridge the gap between being able to detect and characterize the presence of damage (via sensing) and then being able to make revenue-optimizing O&M decisions (by evaluating the effect of damage on state of health) within this single multiple-scale framework. This approach has provided a cost-effective capability to simulate and study a wide range of damage conditions. The focus of the simulations has been on issues that have led to significant turbine downtime and unavailability in the field (e.g., common rotor/blade issues including bondline damage, rotor imbalance, and pitch-angle errors). Sandia has conducted these simulations under a wide range of operating and inflow conditions to better understand the effect of actual operating conditions on damage detection capabilities.

A recent research focus has been on turbine operating strategies, that is, research on what to do once damage is identified in the rotor blades. Sandia is examining smart loads-management strategies (e.g., derating strategies or prognostic control) with respect to several metrics including performance, economics, and ease of implementation. We are evaluating these strategies along with progressive damage modeling to understand the viability of safely operating damaged turbines in order to maximize energy capture versus unnecessary shutdown. Two smart loads-management strategies based on derating are shown in the figure below. Sandia is assessing and comparing the economics of these two strategies—along with analyzing the effect of seasonal variability in wind speeds—to understand the economic benefits of derating in high-wind-speed months versus lower-wind-speed months.

An analysis of smart loads management strategies indicates that limiting the loads (for example, the root bending moments) can be accomplished through small changes in the control pitch schedule. A more greedy derating approach (Derated “A”) results in a power-curve performance that is only affected in the vicinity of the rated wind speed while a more conservative loads-reduction strategy (Derated “B”) limits the power to a constant peak value by initiating blade pitch below rated wind speed.

An analysis of smart loads management strategies indicates that limiting the loads (for example, the root bending moments) can be accomplished through small changes in the control pitch schedule. A more greedy derating approach (Derated “A”) results in a power-curve performance that is only affected in the vicinity of the rated wind speed while a more conservative loads-reduction strategy (Derated “B”) limits the power to a constant peak value by initiating blade pitch below rated wind speed.

In FY14, a SHPM Working Group was initiated and is being led by Sandia through this project. Participation in this initiative is welcomed from all stakeholder groups with interest in this research including those from industry, universities, and national laboratories.

For more information, please visit the Structural Health Monitoring page.