Conference programme

Back to the programme printer.gif Print

Delegates are invited to meet and discuss with the poster presenters during the poster presentation sessions between 10:30-11:30 and 16:00-17:00 on Thursday, 19 November 2015.

Lead Session Chair:
Stephan Barth, ForWind - Center for Wind Energy Research, Germany
Michael Kalkum Vaisala, United States
Mark Stoelinga (1) F Matthew Hendrickson (1)
(1) Vaisala, Seattle, United States

Share this poster on:

Printer friendly version: printer.gif Print


Poster Download poster (11.76 MB)


Comparison of a Global Fleet of Triton Sodars to Collocated Met Towers


As the wind industry continues to mature, development of new wind energy projects requires wind resource measurements that are cost-effective, rapidly deployable to remote locations, and reliably accurate across the rotor plane of modern utility-scale wind turbines. The purpose of this talk is to present results of Vaisala’s validation analysis of a set of 30 Triton/met pairs gathered by a number of Triton users over the past 6 years. The analysis presented here will address two specific questions:
1. Can the Triton essentially act as a met tower, in benign terrain with typical meteorological conditions?
2. Can the Triton improve uncertainty of the wind resource at hub height and above, compared to extrapolation from a 60-m met mast?


Over the course of the past six years, Vaisala has acquired data samples from deployments of Triton sodars in close proximity to met towers (hereafter referred to as Triton/met pairs). Roughly two thirds of these Triton/met pairs were provided by customers, either in response to a Vaisala requests for data, or in connection with customer inquiries about Triton behavior that required investigation by the Triton developers. The other approximately one third of the pairs were associated with either internal or external R&D efforts to validate Triton measurements or to test technology upgrades.

Main body of abstract

Using a large dataset of 30 Triton sodars and collocated met towers deployed around the globe, we have conducted a unique validation experiment that reflects the performance of commercially deployed sodar wind profilers in diverse real-world conditions. When the mean wind speed differences for all measurement pairs are aggregated, the average relative difference is -0.20%, and the percent root mean-square of the differences is 1.51%. This is consistent with an estimated uncertainty of the sodar units of approximately 1%, if the met tower measurement uncertainty is assumed to be independent and approximately 1% as well, a reasonable assumption for a large set of met towers maintained by many different users. It was also found that filling in unrecovered data with wind speeds extrapolated using sodar-derived shear values improves the wind estimates from the sodar in two ways: First, it results in hub-height wind estimates that exhibit uncertainties less than half that of estimates sheared up from met towers – a reduction from 2.6% uncertainty to just over 1%. Second, the additional uncertainty in mean wind speed estimates incurred by a bias in the speed at unrecovered times can be reduced to a mean error statistically not different from zero, and a root mean-square value of 0.59%, with no degradation in the accuracy of the Triton mean wind speed at recovered times.


The findings mentioned here support the use of sodar wind profilers as stand-alone wind measurement devices for use in wind energy resource assessment measurement campaigns.

Learning objectives
- Attendees will learn how to compare sodar profiler data to met tower data.
- Attendees will learn how, with proper treatment, sodar data can be used to derive better wind speed estimates at hub height than via extrapolation from a met tower.
- Attendees will learn how uncertainty from imperfect sodar data recovery can be substantially reduced using the Sodar’s own measured wind shear estimates.