KCF Technologies Blog

Condition-Based Maintenance Can Mean Smooth Sailing for Wind Turbines

A SmartDiagnostics® vibration sensor mounted on a gearbox inside a wind turbine.
A 17-page study by four researchers from the University of Massachusetts and one from the Electric Power Research Institute in Palo Alto, Calif., "Condition monitoring and prognosis of utility scale wind turbines" appeared in 2006 in the quarterly Energy Materials, printed jointly by Maney Publishing and the Institute of Materials, Minerals, and Mining in London, England. The study's goals were to review, "The state of the art in condition monitoring in wind turbines, and related technologies,"and to summarize, "technology needs and future challenges for the development of condition monitoring and prognosis for large wind machines, both onshore and offshore."

Beginning by noting the great potential of predictive maintenance, the study observes that in the 21st century, "wind energy is a commercial technology;" condition-based maintenance, "must increase the profitability of this technology," and, "must contribute more than the alternative (such as run to failure, or periodic manual inspection)," if it is to be universally embraced by the industry.  And, the bar for that acceptance is set high:
"A condition monitoring system must produce actionable information to be useful.  That is, the indications of fault must be sufficiently specific and credible that the operator will order the maintenance action requested by the condition monitoring system based on its recommendation alone.  The challenge is in raising the fraction of faults detected, detecting these faults as early as possible, and correctly identifying the faulty component, all while reducing false positive indications to an acceptable level."
 Much of the report concerns the relative efficacy of predictive maintenance techniques on the various wind turbine components, and analogies in the high levels of reliability required in comparable U.S. Air Force maintenance standards for rotors, turbines, and gearboxes in fixed- and rotary-wing aircraft.  Also taken into account are different standards for maintenance among insurers in Germany, elsewhere in Europe, and the U.S.

It concludes by reviewing, "each of the four major subsystems of wind turbines: rotors, drive-train and generator, electronic controls and power electronics, and support structure.  In terms of value added by condition monitoring systems, the drive-train has the greatest potential, with the rotor a close second....The value added by monitoring the condition of the other systems is less certain."

While asserting, "Wind energy technology has advanced greatly over the past several decades, reducing the cost of energy to the point where it is competitive with other forms of electric power generation in many situations," the researchers concede, "Further advances in technology are required to enable commercial use of wind turbines in more difficult sites such as offshore locations."

"As wind turbines become larger and are located in more remote areas, both onshore and offshore, the value of condition monitoring increases.  Condition monitoring will be enabling for the largest offshore turbines."

2 comments :

  1. Thank you for sharing. Condition Based Maintenance is a major component of Predictive Maintenance (PdM) and asset health

    ReplyDelete
    Replies
    1. Thank you for your comment.

      CBM certainly is a major component of PdM and asset health. However, did you know that, according to Department of Energy statistics, more than 55% of maintenance activities are reactive in nature? This is a staggering statistic especially when you consider that these practices cost the US economy as much as $2.5 trillion per year.

      Here's an interesting question for anyone out there in cyberspace...if the dictionary definition of maintenance is "to keep in an existing state, preserve from failure or decline," why do so many companies continue to use reactive maintenance?

      Delete

KCF Making the Things You Work with Smarter